LOCKE ON IDEAS By RONALD L. CLAYPOOL A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2017
2017 Ronald L. Claypool
To my Mom and my Son
4 ACKNOWLEDGMENTS I would like to thank my dissertation committee for their support and patience. In particular, Stewart Duncan has been a steady source of guidance, much needed and productive criticism, and moral support. John Biro and Gene Witmer have both provided extens ive feedback and support throughout this long process. I am very grateful that Fred Gregory was willing to step in so late in the game. I am also e xtreme ly thankful to my family and friends for their support M y mother and my son, Elizabeth and Richard Cla ypool, have been amazingly helpful and supportive during this endeavor. To select just a few of the many friends to whom I owe thanks, Andra Wilson and Yusun Kang were patient and helpful officemates; I am deeply indebted to Aubrey Spivey and Eugenio Zald ivar for reading large parts of this work and providing substantive and quite helpful input; I am very grateful for the support given by Julie Cholet and Cyrena Sullivan; and, last but by no means least, Andreas Falke has been a steadfast friend and source of encouragement during each year of my graduate school experience. Aubrey and Andreas have each, working separately and together, consistently demonstrated what it means to be a true friend. My thanks also to the University of Florida philosophy departme nt, each member of which has been warmly and thoughtfully supportive. Ja John Palmer and Greg Ray have been particularly helpful, but all the faculty, staff, and graduate students have been wonderful. Thank you. Finally, I am i ndebted to the University of Florida writing program and Santa Fe College for making the completion of this dissertation a financial possibility.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF ABBREVIATIONS ................................ ................................ ................................ .......... 8 ABSTRACT ................................ ................................ ................................ ................................ ..... 9 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ ...................... 11 1.1 The Plan ................................ ................................ ................................ ............................ 13 1.2 Conventions ................................ ................................ ................................ ...................... 16 1.3 A Word about Ideas and Things ................................ ................................ ....................... 16 2 LOCKEAN IDEAS AND IMAGES ................................ ................................ ........................... 19 2.1 Preliminary Points ................................ ................................ ................................ ............ 21 ................................ ................................ ......... 25 ................................ ................................ ........... 37 ................................ ................................ .............. 44 2.5 Images and Ideas ................................ ................................ ................................ ............... 49 3 ................................ ................................ ................................ ....................... 54 3.1 Preliminary Distinctions ................................ ................................ ................................ ... 55 3.1.1 Private Things ................................ ................................ ................................ ......... 56 3.1.2 Mental Objects ................................ ................................ ................................ ........ 56 3.1.3 Imagination or Intellect ................................ ................................ .......................... 63 3.1.4 Summary ................................ ................................ ................................ ................. 65 ................................ ................................ ................................ ...... 66 3.3 Simple Ideas ................................ ................................ ................................ ...................... 70 3.3.1 Empiricism and Atomism ................................ ................................ ....................... 70 3.3.2 Imagistic and Non Imagistic Simple Ideas ................................ ............................. 73 3.4 Complex Ideas ................................ ................................ ................................ .................. 77 3.4.1 Empi ricism and Atomism ................................ ................................ ....................... 77 3.4.2 Imagistic and Non Imagistic Complex Ideas ................................ ......................... 79 3.4.3 Concerns ................................ ................................ ................................ ................. 80 3.5 The Simple Combination Reading ................................ ................................ ................... 81 ................................ ................................ ................................ ... 86 4 REPRESENTATION ................................ ................................ ................................ .................. 88 4.1 Resemblance Theories of Representation ................................ ................................ ......... 90 4.2 Causal Theories of Representation ................................ ................................ ................. 101
6 ................................ ................................ ........... 108 4.4 Simple Combinations and the Limits of Representation ................................ ................ 116 ................................ ................................ ................................ .............. 120 4.6 Summation ................................ ................................ ................................ ...................... 128 5 LOCKEAN ABSTRACTION AND GENERAL IDEAS ................................ ........................ 129 5.1 Essay 2.11 ................................ ................................ ................................ ....................... 131 5.2 Of Modes and Substance Kinds ................................ ................................ ...................... 135 5.3 Separation or Partial Consideration ................................ ................................ ................ 140 5.4 Summary the Theory ................................ ................................ ................................ ... 155 5.5 Essay 4.7.9: Those Damn Triangles! ................................ ................................ .............. 160 ................................ ................................ ................................ .......... 168 5.7 Three Problems from Lowe ................................ ................................ ............................ 174 5.7.1 Recognition ................................ ................................ ................................ ........... 175 5.7.2 Individuation ................................ ................................ ................................ ......... 177 5.7.3 Re semblance ................................ ................................ ................................ ......... 178 5.8 Summation ................................ ................................ ................................ ...................... 181 6 REAL, ADEQUATE AND TRUE IDEAS, AND PERCEPTUAL ERROR ........................... 182 6.1 Real, Adequate, and True Ideas ................................ ................................ ...................... 183 6.1.1 Real Ideas (2.30) ................................ ................................ ................................ ... 183 6.1.2 Adequate Ideas (2.31) ................................ ................................ ........................... 189 6.1.3 True Ideas (2.32) ................................ ................................ ................................ .. 193 6.1.4 Conformity, and the Real, Adequate, and True ................................ .................... 198 6.2 The Problem of Perceptual Error ................................ ................................ .................... 201 6.3 Responding to the Problem ................................ ................................ ............................. 204 6.3.1 The Round Idea is not Simple ................................ ................................ .............. 205 6.3.2 The Round Idea is Real, Adequate, and True ................................ ....................... 211 6.3.3 The Distinction between Imagistic and Non Imagistic Ideas ............................... 211 6.4 Troubling Consequences ................................ ................................ ................................ 213 6.5 Results ................................ ................................ ................................ ............................. 216 7 IDEAS, MATHEMA TICS AND GEOMETRY ................................ ................................ ....... 218 7.1 Locke on Demonstrative Knowledge ................................ ................................ ............. 220 7.2 Trifling Truths? ................................ ................................ ................................ ............... 225 ................................ ................................ ................................ ............ 230 7.3.1 Simple vs. Mixed Modes ................................ ................................ ...................... 231 7.3.2 The Instructiveness of Geometrical Knowledge ................................ .................. 240 7.3.3 The Certainty of Geometrical Knowledge ................................ ........................... 241 ................................ ................................ ............... 249 7.4 Another Carsonian Objection ................................ ................................ ......................... 250 7.5 A Lockean Success ................................ ................................ ................................ ......... 257
7 8 CONCLUSIONS ................................ ................................ ................................ ....................... 258 8.1 My Reading of the Essay ................................ ................................ ................................ 258 8.2 Responses to Critics ................................ ................................ ................................ ........ 260 8. 3 Two Remaining Problems ................................ ................................ .............................. 261 8.4 Further Questions ................................ ................................ ................................ ............ 263 LIST OF REFERENCES ................................ ................................ ................................ ............. 265 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ....... 268
8 LIST OF ABBREVIATIONS Essay John L An Essay concerning Human Understanding Â§ Section Â§Â§ Sections
9 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy LOCKE ON IDEAS By Ronald L. Claypool August 2017 Chair: Stewart Duncan Major: Philosophy Essay concerning Human Understanding idea intended to be rather general, its lack of specificity and an absence of any further explicit cla apparent difficulties. For instance, if these ideas are all mental images (properly understood) However, I argue that Locke is not committed to the position that all of our ideas are imagistic. Instead, Locke is best read as That is, each of our non simple ideas is really nothing mo re than a straight forward combination of simpler ideas, in what can be understood as a basic language of mental discourse. Each term in such a combination is either a simple idea or a non simple idea which is in turn a simple combination. As such, each i dea is ultimately reducible to simple ideas The inherent structure of meaningful explanation, or combination, allows for the compositionality that Lockean atomism about ideas requires, without the component ideas having to be thought of as elements of a s ense experience. These simple combinations effectively establish the parameters
10 of the ideas and what they can properly be said to represent. Ideas are ultimately distinguishable by a comparison of their simple combination contents, and things are appropri ately represented by ideas if they correspond to the specifications of the simple combination. Moreover, although each idea can be reduced to a set of ideas that may be largely imagistic, this does not require that higher level ideas are mental images. Th is is compatible with the generality Locke clearly intends for the term idea and makes possible a clearer understanding of Locke on several important topics, such as abstraction, representation, and demonstrative knowledge.
11 CHAPTER 1 INTRODUCTION It is notoriously hard to get clear on idea In his Essay concerning Human Understanding Phantasm, Notion, Species, or whatever it is, which the Mind ca (1.1.8). 1 While this is clearly intended to be rather general, its lack of specificity and an absence of any further explicit clarification y critique of Locke to the present, it has been common to read Locke as having an imagistic understanding of ideas. That is, all of our ideas are mental image to be compatible with having ideas involving non sight modalities. This reading of Locke has a number of unwelcome consequences for his theory, starting with the objections Berkeley raised. ires ideas that are supposed to represent any number of things, each of which are fully particular (a man has a particular hair color, height, complexion, etc.), but if the idea is an image it seems that such an image would also have to be fully particular (an image of a man with a particular hair color, height, complexion, etc.), but then how can such an idea represent things that do not match it in these particular details? If, on the other hand, another, not strictly imagistic, reading of Locke can be de veloped perhaps it is possible to avoid these problems. In this dissertation, I develop just such a reading. and critics is his position on general ideas. This is cl osely related to the worry I just sketched. Some of our ideas are of particular things, like my idea of this computer, or an idea of a specific 1 John Locke, An Essay concerning Human Understanding edited by Peter H. Nidditch (Oxford: Clarendon Press, 1975). All parenthetical references are to the Essay.
12 cat. Other ideas are of general things, like computers in general, or cats; and it seems to make sense to say we have general ideas like computer and cat which are somehow supposed to refer to all and only computers or cats, respectively. There are a number of questions we can ask about these general ideas. For instance where do we get them? Traditionally, there a re two options that have received most of the attention: we either get these general ideas from experience (which is one of the main tenets of empiricism) or they are somehow implanted, or learned, by us before we are born (innatism). For now, and the rest of this dissertation, let us assume that the best answer is the former. 2 If that is so, how do we get general ideas from experience? After all, we just experience particular things: this computer, that computer, that cat over there, and those cats outside on the porch. Any empiricist theory seems to owe us an explanation of how it is possible to derive general ideas, which we do seem to have, from our experience of particulars. More than that, such a theory needs to explain what those general ideas are lik e, how they can be understood to stand for all the things they are supposed to stand for (particularly when the particulars stood for can be very different, ranging, say from a tiny house cat to a full grown lion, or even to the prehistoric smilodon), and what kind of confidence can we have that the general ideas we have as a result of whatever process applies adequately match up with the real world of particulars we experience. In the Essay Locke advances just such an, explicitly anti innatist, empirici st theory. In this dissertation, I examine what that theory has to say about general ideas, both in response to the kinds of concerns mentioned above and in regard to some of the things Locke thinks we can do with those ideas (that is, have knowledge of ge neral truths, particularly in mathematics and geometry, and even in morality). 2 innatism arguments, but empiricism in the sense I mention above is a critical commitment of his theory that has important ramifications in the following chapters.
13 What are they? How do we get them? How do they work together? How can we make sense of some what reading of Locke on ideas is the best, most accurate? Throughout this dissertation Essay in effect, the theory of the Essay My goal is to offer what could be thought of as a best reading of what the Essay says. While I do often argue that what the Essay says is not as problematic as many commentators have thought, it is not my purpose to argue that Locke was right Where ther e is a tension between what the Essay says, or, more accurately, the implications of what the Essay says, and what Locke might have actually thought, I always go with the Essay over Locke. My interest is primarily in what the Essay says rather than what Lo cke might have really thought. 3 I seek to understand how the Essay is best read and along the way, as it turns out, I 1. 1 The Plan As I have already said, some of the longest standing worries directly from the assumption that he must be thinking of ideas imagistically. Recently, Michael Ayers has taken the position that for Locke all of our ideas are images. 4 In Chapter 2, I argue that and a careful consideration of the Essay shows that his position is mistaken; not all of our ideas are images, though, of course, some are. Discussion of the imagism issue opens up the question of how we are to understand Lockean ideas that are not image s. Moreover, since some of our ideas are images, how are we to 3 Wherever I say Essay 4 Michael R. Ayers, Locke, Vol. I: Epistemology (London: Routledge, 1991), 44 51.
14 understand how imagistic ideas work with non imagistic ones? In Chapter 3, I c onsider these questions and develop what I call my simple combination reading of Locke, which I argue is compatible idea and accommodates both imagistic and non imagistic ideas. This reading enables a fresh look at representation, or how it is that ideas are supposed to stand for the objects Lock e clea rly intends them to represent. While it applies to all ideas, t his is a particularly serious issue for general ideas, which are supposed to stand for a wide range of particulars. In Chapter 4 I argue that a careful reading of the Essay shows that what Locke has in mind is something of a two part theory: a causal understanding of the representation done by simple ideas and a use theory for non simple ideas That is, the representational relationship is one that we establish in using x to represent y. This stands in contrast to the leading reading contender resemblance theories. Although my reading is a causal theory, and could be considered a modified causal theory it is quite a bit different from more common varieties, because one of t he causes involved is the mind, or, as Locke says, the understanding. abstraction to which I turn in Chapter 5 What Locke says about abstraction is not the clearest part of the E ssay and I try to explicate his position. I argue, against Vere Chappell, 5 that Locke has a single theory of abs traction rather than two. I also argue that Lockean abstraction does not require a strict separation of ideas, but accommodates a less demandin g partial consideration of some idea, or set of ideas. Also, Locke says that the general idea of triangle imperfect, that cannot exist; an Idea wherein some parts of several different and inconsistent 5 The Cambridge Companion to John Locke edited by Vere Chappell (Cambridge: Cambridge University Press, 1994), 26 55.
15 Ideas are put together 4 7 .9) This has caused a great deal of consternation among commentators, as it seems to put Locke in a hopeless bind. Locke seems to be saying that we cannot have such a general idea, but in the rest of the Essay he clearly seems to think we can. In fact, his project of securing our knowledge of, say, mathematics and morality depends upon our being able to have such general ideas. However, I argue that a careful reading of the context in which this claim is made dispels the problem for Locke. Locke makes the puzzling cl aim that some of our ideas are real, adequate, and true. This has received surprisingly little discussion from commentators. However, Antonia LoLordo points Loc ke with a serious problem with perceptual error. 6 In Chapter 6, a fter a careful reading of the Essay and an explication of what Locke means when he says that an idea is real, adequate, and true, I argue that he really has no problem with perceptual error a t all. However, consideration of simplified taxonomy of ideas is not clearly compatible with the fundamentals of his empiricism. Finally, in Chapter 7, I take up what Locke says about systems of de monstra tive knowledge which raises a number of important questions. He clearly identifies two such systems, mathematics (including geometry) and morality. Unfortunately, what Locke says about the latter is tantalizingly brief, and I forego consideration of it for now However, what he says in regard to mathematics is enough to generate issues. For instance, Emily Carson raises a number of critiques, and one general objection, concerning the nature of geometrical ideas and the source 6 Philosophy and Phenomenological Research Vol. 77, No. 3 (November, 2008), 705 724.
16 of the informative nature of claims involving them. 7 Again, a careful reading and consideration of the text of the Essay along with the employment of the simple combination reading, disarms while it is not perfectly clear exactly where Locke thought th e informativeness of our mathematical and geometrical ideas resides, there are several options that seem to work, and I argue that any of them are sufficient to deal with the problems Carson raises. 1. 2 Conventions All parenthetical references are to the Essay and correspond to the book, chapter, and section of the referenced material For emphasis within a quotation, I use bold font Wherever bold text appears the Outside of quotations I use italics for emphasis. I also use italics to indicate an idea, and rely upon context to make it clear whether the italics are there for emphasis or to point out that I mean an idea. I use quotations to indicate that I am referrin g to a word, rather than an idea or an existent. For instance, my idea of Jim [the idea] which represents my friend Jim [the actual person]. 1. 3 A Word about Ideas and Things The quot such a way that Locke is taken to imply that anything that the Mind can be employe d about in thinking is an idea. Such a reading would push Locke toward Berkeleyan idealism by denying 7 British Journal for the History of Philosophy Vol. 10, No. 3 (Aug., 2002), 359 Locke Studies Vol. 5 (2005), 19 Intuition and the Axiomatic Method ed. Emily Carson and Renate Huber (Dordrecht: Kluwer Academic Publishers, 2006), 3 Locke Studies Vol. 7 (2007), 21 46.
17 any distinction between ideas and, say, things existing in the world independently of our minds. This, however, is not what Locke means to say, and I will offer my brief argument to that effect in this section in the hope of leaving the matter suff iciently settled that the rest of the dissertation can proceed without further time spent on the issue. First, the list of things Locke gives as rough synonyms of idea (phantasm, n otion, and s pecies ) strongly suggests that he intends to use idea to stand for some kind of mental entity clear connotations of something that is only in the mind. He follows the passage already quoted ranted me, that there are such Ideas in one is conscious of them in in (1.1.8). So, Locke is fairly explicit that ideas are something in the mind, and he does n ot intend for the term to also apply to external things. Second, Locke is concerned to make it clear that some of our ideas correspond to things are thought to represent while some are not, and that our claims about the world might be said to give us true knowledge of the world outside us. 8 It would make no sense for Locke to worry about such correspondence to, resemblance of, and true knowledge of the world outside us, if he thought that there were no world beyond our minds and ideas. Moreover, he must think that we can be employed in thinking about such external things, else we could not consider said correspondence and resemblance, etc. So, Locke must thin k that there are at least two categories of things, ideas and external things, both of which we can be employed about in thinking, and these two categories are not 8 I discuss each of these points more fully in the following chapters.
18 intended to be co extensive. Accordingly, in Chapter 3, I will describe ideas as mental obje cts about which we can think without presuming the external existence of those objects. 9 Of course, this does not tell us what kind of objects these are, and it is to that task I now turn. 9 explicit eschewal of metaphysical considerations. The more specific t the external world, whether it is direct or indirect, is a question that I le ave almost entirely unaddressed in this dissertation, but hope to return to in future work.
19 CHAPTER 2 LOCKEAN IDEAS AND IMAGES As I have already mentioned, in the introduction to the Essay Locke gives an often quoted explanation as to why he is using the term idea : It being that Term, which, I think, serves best to stand for whatsoever is the Object of the Understanding w hen a Man thinks, I have used it to express whatever is meant by Phantasm, Notion, Species, or whatever it is, which the it (1.1.8). This is also normally taken to stand as an explanation of what Locke means by that term. Locke seems to be telling us that for him, and the Essay idea what kinds of things can be the objects of understanding. Unfortunately, Locke is not explicit about the nature of ideas, particularly in the general sense he seems to have in mind in the foregoing passage. Aside from potentially being objects of the under standing, what can we say about Lockean ideas as such? What Locke says is not particularly helpful in answering this question at least not in any direct sense. However, based on what Locke does say in the Essay and reasonable expectations one might have ab out such a theory, we can make some headway on clarifying a general Lockean commitments on ideas in general. For instance, it will not do to have a Lockean theory of i deas that requires some ideas to be something other than mental images if Locke is committed to the position that all ideas are images in some sense. This example is deliberate in that the focus of this chapter is the question of whether Locke thinks that all ideas are images or that, while some ideas are images, some are not.
20 One trend in Locke commentary, arguably implicit in Berkeley and Hume, and explicitly developed more recently by Michael Ayers, is that Locke thought all our ideas were mental images of some sort, which is the imagist thesis. As will become clear, the (often merely tacit) theory; objections raised against it often presuppose the imagist thesis. The other trend is that Lockean ideas also include non imagistic mental objects more akin to concepts. I support the second trend. I hold that it is implausible and unnecessary to think that all Lockean ideas are images, and that there are at least two ty pes of ideas operating in the Essay understood than Locke explicitly construes the notion. 1 In the following, in Â§1, I will offer some preliminary comments on what Locke says about ideas as such and how that might be ar upon the issue at hand; I will then, in Â§Â§2 4, either clearly wrong or, at best, insufficiently supported, and, along the way, support my own not strictly imagist thesis; and, finally, in Â§5, I offer some thoughts about how our mental images and ideas might relate. The following is largely negative (even polemic) in nature, because this seems the best way to effectively establish the positive position tha t for Locke some ideas are images while others are not. (1) The non imagistic ideas seem to naturally follow from what Locke says in general throughout the Essay and (2) there seems to be no good reason to accept the more restricted, imagistic, thesis as it is defended in what seems to be its most developed form in 1 strictly as variations on those locutions) simply mean the position that not all ideas are images. Obviously, this is not the same as the position that no ideas are images, which is hardly a tenable position and certainly not one that I am advancing.
21 Ayers. Therefore, if I am successful in defending these two claims, I take myself to have adequately established the more general, not strictly imagist, position. 2. 1 Preliminary Points Given the anti innatist program of Book I of the Essay it is perhaps unsurprising that in Book II Locke is less concerned with what ideas are than in how we get them and what we do with them. At the very beginning of the part of the Essay concerned specifically with our ideas as such, after offering a sample list of the ideas someone might have, 2 Locke offers his basic story of how we come to have the various ideas, and sorts of ideas, Locke presumes we do all have. Of course, according to Locke, the fundamental source of all our ideas is experience. As he puts it, external sensible Objects ; or about the internal Operations of our Minds perceived and reflected on by our selves is that, which supplies our Understandings with all the materials of thinking One way to think the strictly imagistic thesis gets off the ground is to point out that the kind of observation Locke mentions here would deliver phenomenal experienc es, which would basically amount to images, broadly construed. An idea of a particular white patch gathered from observing a bit of snow would most plausibly be a remembered visual experience, or, more accurately, a component of such an experience. Similar ly, an idea of a particular kind of calm summer night also amounts to remembering part of a particular experience. In each case that remembered thing is most direc tly thought of as a kind of image called up before the mind. However, simple observation of external objects and internal operations does not by itself result in the ideas we have. Clearly, for a start, the mind must also have the capacity to isolate 2 Whiteness Hardness Sweetness Thinking Motion Man Elephant Army Drunkeness
22 par ticular elements of our observed experiences, as suggested by the two examples I just gave. The actual experience of the snow or the summer night would, of course, involve much more than just the whiteness or the sense of calm. This is acknowledged by Lock Ideas we have, or provides us with the raw material of which all of our ideas are made, or is the essential source of the process, or processes, that result in the ideas we have. Much of the rest of Book 2 of the Essay is concerned with spelling out how we go from our basic experience to having those various ideas. Along the way, Locke considers faculties of the mind that are used to manipulate and, to at least some degree, alter the basic material furnished by experience faculties including retention, discernment, comparison, composition, enlargement, and abstraction. 3 As a r esult, we have a wide variety of ideas, from those ideas of particular experiences, to simple ideas like whiteness or solid to simple modes like three to mixed modes like freedom and army to ideas of substance kinds like human gold and elephant The question of this chapter is: does whatever we do with the direct results of experience, which might consist of just a wide range of images, broadly, but reasonably, construed, result in ideas of a different sort; that is, ideas that are not plausibly cons idered to be images? I think t he because that is the best way to make sense of both what Locke says in the Essay and the ideas we actually have. The rest of this chapter constitutes my argument for that answer. Before moving ahead, though, there are a few things Locke says that bear upon the issue at hand. 3 Respectively: 2.10, 2.11.1 3, 2.11.4 5, 2.11.6 8 (for both composition and enlargement), and 2.11.9 11.
23 at what time a Man has first any Ideas, is to ask, when he begins to perceive; having Ideas .9). Our having ideas comes with our experience of things, either external objects in the world or our own internal processes, and it does this automatically. Perception here seems to mean just being aware enough of that experience to be receiving ideas. I t seems hard to imagine how one could count anything less as experience. In this paragraph of the Essay Locke is taking one more shot at the notion of innate ideas, but it might also be a support for the imagistic thesis. If perception is understood to co nsist of having an experience, in some phenomenal sense, it might be necessary to think that what one is having an experience of be a kind of image, broadly construed. So, to have ideas would be to have such images. Clearly, this conclusion depends upon The power of Perception is that which we call the Understanding Perception, which we make the act of the understanding, is of three sorts: 1. The Perception of Ideas in our Mi nds. 2. The Perception of the signification of Signs. 3. The Perception of the Connexion or Repugnancy, Agreement or Disagreement, that there is between any of our Ideas All these are attributed to the Understanding or perceptive Power, though it be the two latter only that use allows us to say we understand (2.21.5). So, the power of perception just is the understanding, and that power is manifested in three ways. The first way is, apparently, the one most directly related to the topic of this chapter, but we do not get much clarification from the simple identification here. The other two sorts of perception, though, are helpful. The third involves our grasping the most basic relation between two or more ideas; that is, whether those ideas agree in the s ense Locke has in mind. This perception might be thought to amount to a kind of phenomenal experience that comes with an immediate matching
24 component of experienc ing the agreement or disagreement in question. A case could perhaps be made for this qualifying as an emotional image of sorts. However, the second sort of perception, ry. When I perceive what is signified by a given sign, it is not at all clear that what I perceive could even be an image in all cases. I will return to this below, but it seems that when I perceive what is y be an image of any sort. To say that it must be an image basically amounts to begging the question. So, while the above passage is not understanding, or perhaps grasping; we perceive that ideas agree or disagree when we grasp that match or dismatch; and we understand a sign when we grasp what it signifies. This sense is neutral, I think, between imagism and the more general position that not all ideas are images. What Perception is every one will know better by reflecting on what he does himself, when he sees, hears, feels, etc. rather general understanding of perception, which is supported by the rest of the discussion in that chapter. Note also that t etc. sense perception and that involved in thinking, whatever that turns out to be. idea and perception in such a general, all encompassing, way supports the g eneral, not strictly imagistic, position rather than that of the imagist reading. To further support the general reading I favor, I now turn to considering an imagist reading directly.
25 2. 2 Ayers The most prominent 4 One might think it is not such a quasi sensations, 5 but Ayers means to say that for Locke all ideas are sensory images or quasi s case for this imagist reading is made 6 (when not actual we have a separate intellectual faculty over that of the imagination, and, therefore, whether or not purely physical mechanisms might explain human thought. 7 After pointing out that this question 8 The implication here is that to be an imagist means holding that ideas are all sensory images, more or less. 9 4 Michael Ayers, Locke, Volume I: Epistemology (London; Routledge, 1991), 51. 5 However, as I will argue in Chapter 3, it is actually not likely that all ideas are images even when they occur in present sensation or reflection. Unity and existence seem to be ideas that occur with all of our experience but it is implausible to think, or that Locke thought, that they are images. 6 Ibid., 44 51. 7 Ibid., 44. 8 Ibid. 9 The thought process is:
26 commitments Locke has that stem from the text of the Essay but upon commitment s Ayers takes Locke to have as an imagist. 10 Ayers lays out a number of points that he thinks support his conclusion that Locke should be considered a member of the imagist camp, discusses and rejects what he takes to be the most obvious objection to this 11 I will discuss the first of res pectively. difference of opinion, fairly long standing by the time Locke wrote, as to the nature of the processes involved in the formation of our concepts of the objects of 12 The Aristotelians, and to some extent Descartes, Ayers tells us, thought there was such a distinct faculty, while others, like Gassendi (most of t he time), Hobbes and Glanvill, thought, apparently, that no other faculty than imagination was needed. The issue was therefore a metaphysical one as well as epistemological because the imagination was commonly thought to be a merely material process, or o firmative answer [to (1)], the grounds for holding him an imagist (3) Apparently, being an imagist is entailed by holding that all ideas are sensory images. (4) So, it would seem that Ayers thinks holding that all ideas are se nsory images is a sufficient condition for being an imagist. 10 It seems to me that a better term would be imaginationist, since the issue is really about whether we only have a faculty of imagination, rather than that plus intellect. 11 Ibid., 49. 12 Ibid., 44.
27 metaphysical considerations, but, as we will see, Ayers seems to think the materialist dualist debate is playing a role in giving us cause to think Locke was an imagis t. Quoting the passage from the Essay with which I opened this chapter, Ayers identifies remark seems to have been modelled on a number of similar remarks by Gas 13 The when we are thinking of anything. It gets given a number of other names. It is is now a familiar and well worn term, a nd suffers less from ambiguity than the others. 14 Though there are of course similarities, what Gassendi says seems noticeably different not say that what he is talking about is an a specific thing, which he does explicitly say is an image, though it has been called by other names. Locke, on the other hand, explicitly seems to be explaining how he will be using a most the similarity between the passages might suggest is that Locke had read other works and felt free to adapt what others had said to h is own purposes. It does not, by same thing. In fact, given that the passages are not the same, I suggest that the similarity emphasizes the difference between th 13 Ibid., 45. 14 Pierre Gassendi, Opera Omnia 1658:I (Lyons) 92, quoted on Ayers, 45.
28 Mind can be e metaphysical theory neutral sense, staying away from the question of whether or not there was a pure, non material intellect involved. T his may be so, but in his apparently deliberate decision to not say that what the mind thinks about are images, Locke seems to be expanding the theory neutrality to also cover the question of what sorts of things the mind can think about. 15 In any case, th ere seems to be very little reason to think that a comparison of the two passages supports the contention that Locke was an imagist. Essay might be thought to help. Although the first cont ext would seem to argue against his position, Ayers between idea and objec 16 Ayers points out that as far as secondary qualities are concerned, at Resemblances of cause not theory does not require for representational purposes that our ideas be images that resemble that 15 I am not here taking any position at all as to how well aware of Gassendi Locke actually was. For much more on Journal of the History of Ideas Vol. 45, No. 3 (July Sep., 1984), 339 359. 16 Ayers, 46.
29 which they represent. I certainly agree with Ayers on this point, 17 but wonder why we should not understand this to suggest that Locke did not think all ideas had to be images. If Locke disavows the need for imagistic resemblance, does this not support the notion that he was at least open to the possibility that some of our ideas are not images, but perhaps something like concepts? seems that Ayers should consider what Locke is saying here more carefully. It seems that Locke is granting us a necessary condition for the not strictly imagistic position: our ideas of some qualities do not have to resemble those qualities. Instead, thou First, in the chapter on retention, Locke clearly indicates that recalled sensory idea s are mental 18 Ideas The Pictures drawn in our Minds, are laid in fading Colours that Locke should indicate that sensory ideas are mental images. It is hard to see how sensory ideas could be anything but images, if image is understood generally enough to allow for sounds, smells, tactile impressions, and so forth. It is not my positio n that none them all to be images, which is the position Ayers is defending. indicate that remembered ideas are supposed to correspond to things currently to hand, and that memory as well as one 17 There is more on this issue in Chapter 4. 18 Ibid.
30 19 Ayers points out that retention is critical in the Lockean story of thought, and that Locke clearly indicates that he holds that all ideas that we have in the mind are (also?) in the memory. So, the st ory seems to be, what is retained in the memory are images, and we do not have any ideas that are not in the memory. Absent some kind of purely intellectual notion of the imagination, which is strictly imagistic. We need to retain our ideas in the memory, the only thing that is available for such retention is imagination, so Lockean ideas must be ese connections, an 20 I think we should not accept this line of reasoning. What Ayers seems to need here is something that says only sensory images are in the memory. Nothing Locke says indicates that he is restricting memory so severely. If he did say that, it seems plausible that this debate over the nature of Lockean ideas would have been settled long ago. Instead, Ayers is trying to get that restriction slipped in by pointing ou t that there is no separate intellectual (non imaginative) at a time when the issue over whether or not we humans have a distinct, purely intellectual faculty, which was presumed to be (necessarily?) immaterial, or merely an imaginative faculty, which was thought to be (potentially) material. Surely, Locke must have had a position on this issue. As Ayers points out, Locke does not anywhere say that we have a dist inct faculty of the intellect, so when he talks about our retaining things in the memory Locke must be relying upon 19 Ibid. 20 Ibid.
31 However, what we do clearly have from Locke is his explicit disavowal of any such metaphysical commitments: I shall not at present meddle with the Physical Considerations of the Mind; or trouble my self to examine, wherein its Essence consists, or by what Motions of our Spirits, or Alterations of our Bodies, we come to have any Sensation by our Organs, or any Ideas in our Understandings; and whether those Ideas do in their Formation, any, or all of them, depend on Matter, or no (1.1.2). Locke is clearly setting aside any consideration of whether or not there is a material o r immaterial, or combination thereof, component or components involved in how our minds this, and in harmony with the rest of the Essay (not withstanding occasional forays into is relying upon to secure our acceptance of Locke as an imagist. Of course, Locke does occasionally seem to talk as if he is taking a position on the metaphysical issues. For instance, in a passage Ayers notes, Locke seems to think mental quite materialistic. The full passage, though, seems a bit more tentati ve than Ayers suggests. In explaining how it is that one idea is regularly associated with another, Locke says: Custom settles habits of Thinking in the Understanding, as well as of Determining in the Will, and of the Motions in the Body; all which seems t o be but Trains of Motion in the Animal Spirits, which once set a going continue on in the same steps they have been used to, which by often treading are worn into a smooth path, and the Motion in it becomes easy and as it were Natural. As far as we can co mprehend Thinking, thus Ideas seem to be produced in our Minds; or if they are not, this may serve to explain their following one another in an habitual train, when once they are put into that tract, as well as it does to explain such Motions of the Body ( 2.33.6). In this and the following metaphor of a musician who cannot help but think of the rest of a tune once he hears the first few notes, Locke seems clearly to be engaged in a bit of metaphorical
32 discourse. He is not saying that association of ideas i s such passages commit Locke to materialism, it seems they emphasize his deliberate agnosticism. Certainly, it is possible that, in his heart of hearts, Locke was a materialist, or harbored materialist sympathies, but his theory as presented in the Essay is neutral on such issues. As far as that theory goes, Locke does not care what mechanisms make it possible that the mind functions in the way it does; what he is interested in is describing what is going on when we think, remember, and so forth. This neither argues against nor for thi nking of Locke as an imagist. If we are to take Locke seriously and evaluate the theory of the Essay on its own merits, which is my project, we must decide the issue of this chapter without resorting to metaphysical speculation. explicit agnosticism on such metaphysical issues but points seen to link his project with that of Hobbes. 21 Ayers observes that Locke is inclined to connect such things as association, religious enthusiasm and error with physical things, much as thinking to some higher, non material faculty of pure intellect operating with pu re, non sensory 22 The implication is that in so far as Locke seems comfortable with the possibility of there being a completely physiological explanation for thought, memory, association, etc., 21 Ibid. 22 Ibid., 47.
33 without ever discussing a separate intellectual capacit y, Locke seems to be siding with the imagist (materialist) camp. This line of argumentation seems weak, though. There seems to be reason to think that non metaphysical program. Again, it is possible Locke harbored sympathies for the materialist camp, but his explicit theoretical position is that he is not taking sides in that argument. Why then does he appear to favor hypotheses that are compatible with the one side and not the other? I think the answer to this is that where there is a disagreement between two ontologies, where both are supposed to adequately explain the phenomena (or when one is specifically avoiding that issue), any attempt to be agnost ic is going to favor the simpler ontology. 23 In this case, Locke can hardly deny that there are material mechanisms at play, 24 so the only thing he can really be agnostic about is whether or not there is a separate, immaterial faculty of the intellect, with or without a separate material faculty (as the imagination was considered in the imagist camp). It is not so much a case of either or, but of this or this plus that. So, if any hypothesis he entertains must be compatible with both just matter and matter an d intellect that is going to suggest a materialist/imagist bias, if position on this debate. 25 The above line of thought might be taken to present a problem for my own position that not all Lockean ideas are images; if that agnostic principle pushes Locke to talk in terms that 23 I call this the agnostic compatibility principle. 24 Locke is not Berkeley, after all. 25 razor to come down in favor of the materialist side of the issue, but he does not seem at all convinced that both ontologies do adequately explain the phenomena.
34 suggest the materialist/imagist metaphysics, does it not suggest that it would also drive Locke to think all ideas must be images? If we need ideas that are compat ible with both doctrines, the suggestion might be, it seems that the only ones that could do the trick are imagistic ones. I do not think this follows, though, for the actual thing about which Locke is trying to be neutral is whether we need to have a pure ly intellectual faculty to do all the work that our minds clearly do. For all Locke can grant on the theory presented in the Essay it might be the case that a purely physical faculty can deal with utterly non imagistic ideas. His agnosticism bars any stan ce on what material or immaterial faculties we have, not any position on what sorts of ideas we happen to have. This distinction is analogous and clearly related to the one he makes regarding the possibility of a purely material thing thinking. Locke deni es that we can know Ideas without revelation, to discover, whether Omnipotency has not given to some Systems of Matter fitly d isposed, a power to perceive and think, or else joined and fixed to Matter so disposed, a thinking immaterial some purely physical imaginative faculty and a non physical intellectual faculty, or if, perhaps, God has seen fit to give the imaginative and intellectual faculties both to merely physical things (which for all we know we might be), or some other distribution of faculties and substance sorts. However, be very Doubts about what it is, confirm the certainty of its being, though we must content our selves in the Ignorance of what kind of Being cannot know whatever it is that makes us capable of thought might be there can be no doubt that we do think. Similarly, we do, obviously, have ideas of all sorts, regardless of what mechanism we have that
35 makes this possible, and Locke has specifically ru led the question of what that mechanism (or mechanisms) actually is (or are) as out of bounds, but that does not mean he has to put off questions about what we do in thinking, or what kinds of ideas we have. So, while I think this agnostic compatibility p commitments that the pri nciple comes into play against him. Ayers goes on to point out the fact that if Locke thought there was a separate faculty of surprising that [an imagist] sh ould have refrained from the explicit assertion that thought can be efrained from discussing or even mentioning the difference and relation between the faculties of intellect 26 Locke does mention lots of other faculties, and seems to be implicitly engaged in the debate in question (as for instance in the d chiliagon ), but Locke does not mention intellect. So, Ayers seems to conclude, Locke must not have thought we have any such faculty therefore we only have the imaginative one. Ayers goes on to point out tes in several areas, particularly in regard to innate ideas and principles, and therefore some resemblance to Hobbes, who clearly was an imagist. Certainly numbe r and in the aforementioned metaphor about our memories being painted in fading colors, but we have no reason to think that Locke is thereby committed to the imagist cause as Ayers 26 Ibid.
36 ntellectual, or, for that matter, a purely imaginative, faculty as a deliberate application of his official agnosticism. As Ayers has helpfully underlined, there can be little doubt that Locke was aware of the metaphysical issues surrounding said faculties and his project is deliberately and explicitly non imagism, but a re sult of his official neutrality on this topic. not there was, or had to be, both a (probably) purely material faculty of imagination and a completely immaterial faculty of pure intellect (the dualist position) or just the former to explain human thought (the imagist); Locke was aware of this debate and must have had a position on this debate; and, it seems most plausible to read Locke as being on the side of the imagists. While the first point is certainly true and the second seems likely, 27 I have argued that the third is going too far. Not only is Locke explicitly committed to avoiding weighing in on the metaphysical issue being debated, but he seems to have allowed the p ossibility that it might be possible that a physical faculty might be able to deal with purely intellectual ideas perhaps due to super addition, for instance. So far, Ayers has simply given us no reason to think that Locke is theoretically committed to ima gism. 27 Essay
37 2. 3 The Argument against s Position Ayers next responds to what he says is the most common objection to the position he is taking: Locke talks about ideas like jealousy and lie and so on, but it does not make sense to think that these are images. Laid out formally, the argument Ayers considers is: (1) (2) images of these things. (3) 28 Since this is very similar to a big part of the argument I am advancing in this chapter, it is important that I address what Ayers says her e. Ayers has three responses to this argument: (a) it begs the question; (b) the most it shows is that an imagist like this would be allowing an impossibility, which merely puts Locke in good company; and, (c) this objection fails to appreciate what Locke meant by ideas of reflection or understand his doctrine of abstraction. It is not particularly clear what Ayers means by (a), and he does not offer any clear support for the assertion. As far as the formal argument is concerned, it looks like the only pla usible candidate for question begging status is (2). One might certainly think that the bald assertion that it is impossible to have these sorts of ideas as images amounts to begging the question, but one must wonder if anyone has ever done just that asser ted without support that it is impossible to have these kinds of ideas as images. While Ayers has offered this argument on 28 Ibid., 48. The numbered format is provided by me, for ease of discussio n, but the argument is presented by Ayers in support of his position.
38 does seem a bit odd to think someone would just flatly claim without support that it is impossible to have such ideas. If anyone has done this, though, I am certainly willing to agree with Ayers that it seems they have begged the question. A more substantial possibility is that the question b we did not have the resources to give support to the claim that Locke does have a good theory. I of ideas is at least plausible, and does accommodate a variety of ideas including clearly imagistic ones like red and sour and ideas that are less plausibly imagistic like, perhaps, unity being cause power and substance In the absence of the exposition of the theory as I am developing it, it might very well am not making any such assumption. A more serious problem with (a) is that the list of ideas in (1) seems too restrictive. My own version of the argument (below) under consideration includes ideas like animal, democracy adultery and incest in the list given in (1). Ayers is about to argue, under (c), that the ideas he has mentioned in (1) are compatible with an imagistic understanding, but I will suggest that the ideas of the variety listed in the previous sentence are not at all so compatible. In any case, the idea that such an argument, with the emended list in (1), would beg the question is now far from ob vious, and, moreover, it seems unlikely that anyone would put forth the argument as Ayers phrases it with the objectionable (2) completely unsupported. So, I suggest we set aside the question begging objection. Taking the argument just as Ayers has present ed it, (b) amounts to the objection that the conclusion simply does not follow from the premises. Given (1) and (2) it is also possible that Locke was simply confused about this issue and allowed an impossibility. While this is certainly
39 uncharitable, as A 29 That is, all of the imagists prior to Locke thought we had the kinds of ideas listed in (1) and that they also held (2) t o be true, while also denying non imagistic ideas. I believe I have already given an implicit response to (b). I agree that (3) does not follow deductively from (1) and (2), but I have already suggested that I think neither of those two premises are adequa tely framed. Moreover, I have argued at length in the foregoing, and will continue to argue in the following chapters, that Locke and we have good a more conce ptual understanding of ideas, so Locke has no good reason to think that there is an impossibility entailed by his theory and plenty of reason to think there is no such impossibility. Beyond that, and finally, if I can make a sufficient case for my own vers ions of (1) and (2) theory rather than think he was guilty of allowing an impossibility. I think what Ayers says to support the first part of (c) is right to so me degree. Jealousy and lie involve ideas of reflection, and Locke did carefully explain reflection as being very much reflection to be in effect a class of retain ed and revivable impressions which for the most part 30 Even granting the special (notoriously problematic for Locke) case of substance on a str ictly imagistic understanding. What exactly are we imagining, and how different, really, is this thing that is very much like a sense? There is perhaps the concern here that what is being 29 Ibid., 49. 30 Ibid.
40 considered to be images broadly understood is becoming so broad as t o suggest that there is no practical limit to what the imagination can be said to accommodate. All of this, I think, suggests caution in too quickly rushing to attribute such a puzzling doctrine to Locke. rstanding of ideas like unity being cause and power are accommodated by an imagistic reading. This involves an understanding of requires nothing more than partial consideration, as opp osed to strict separation, 31 but this does not entail that all nsidered as two in mind without having in mind the idea of some dual in particular, but considered barely as 32 in one move. This is simply fallacious. To move from the observation that we do occasionally have some particular thing in mind, as an image, properly construed, and partially consider it to derive some abstract idea to the claim that we always must do th at is, without more support, just a mistake. Moreover, existence is right we notice, perh aps, there is something the things around us have in common that other things we can think about, say, unicorns and fairies, do not, and we then partially consider something that does exist as just something that exists to come up with the abstract idea of existence (though, in this case, existent is the more appropriate general idea) but that does not mean that Locke thinks we have to have an image in our mind of some particular existent in mind every time we employ 31 See Chapter 5 (Â§3). 32 Ibid.
41 existence in our thinking. However, it is also clear that Locke thinks we do not have to observe a particular instance of something, like incest or patricide to acquire that idea. 33 position: I have suggested th at the argument is probably not how anyone has argued (i.e. asserting (2) so baldly) and that the charge of question begging seems premature; I have said that the list of ideas in (1) is too brief, and will come back to this in a moment; I agreed that the conclusion does not strictly follow from the two premises, but suggested that this was, perhaps, not a problem since the premises are inadequate anyway; I have suggested that employing an imagistic understanding as broadly as Ayers indicates might be probl ematic; and, finally, I have understanding of abstract ideas is fallacious. Taking myself to have shown that at least one of Ayers objections ((c) the one he develo ped most fully) does not go through, and having indicated that the other two depend upon the way Ayers himself has set up the argument, I will now present my own version of the argument. (1*) Locke believed that we have ideas of animal justice property adultery and incest (2*) It is wildly implausible that our ideas of these things could be (mere) images (3*) Without an explicit statement by Locke, or a solid argument, to the contrary, we should not think that all Lockean ideas are images. (4*) Locke m akes no such explicit statement, and Ayers fails to give a solid argument, against the general thesis or for the imagistic one. (5*) Therefore, not all Lockean ideas should be understood to be images. 33
42 The first thing to point out here is that the conclusion does actually follow, thereby accomplishment, I admit; I have actually tried to present a valid argument. I will now offer support for my premises. (1*) is clearly ideas that are not plausibly imagistic, and we have no clear reason to think that Locke is committed to the strong imagistic thesis Ayers is advancing, we should conclude that not all Lockean ideas are images. I am assuming here that we need good reason to think that Locke was simply confused or had a theory that runs counter to the intuition expressed in (2*). Of course, (3*) depends entirely on the truth of both (1*) and (2*). Th e first half of the conjunct expressed by (4*) is true, else this issue would have been settled long ago. Sections 2 through 4 of this chapter constitute my argument against the second half of that conjunct. This leaves (2*), which is clearly doing the hea vy li fting in this argument. I now turn to a brief defense of this premise. consideration reading of Locke, we use a mental image of some particular thing, partially considered, to represent for ourselves all such things that f all under a given general idea. So, we might imagine a particular ball and partially consider that ball as a sphere, or its circumference as a circle. This is all well and good for such things, and I have no quarrel with reading Locke as thinking that some of our more basic ideas of sensation and reflection are just like this. However, in the case of ideas such as animal democracy adultery and incest this is just not a plausible position. Yes, it is possible that I use some representative animal to stand for any animal whatever, but only the slightest reflection is needed to show that this is not at all what I do. Even when pressed to give some accounting of what I mean by an animal, it never once happens that I picture, say, an aardvark and use it as an i magistic
43 representation. I might start the process of forming the general idea of animal by abstracting from a particular animal, but it is clear now that whatever it is that I have in mind when I use or think of animal now it is not an image of an aardvar k, a raccoon, a grizzly bear, a chimpanzee, or anything else. The same can be said of justice property adultery and incest but with the added complication that in these cases I cannot even work out what sort of thing I am supposed to be using as a menta l image in the first place. Perhaps there are some people out there who have a pictographic mental life by which they imagine some representative case of adultery and use it to stand for all such cases, but that must be some fairly complicated image to hol d, and I can say that I do not do this. It is more likely that there are some people who have a more imagistic mental life than I do, but, given the kind of ideas under consideration, it seems quite counter intuitive to think that anyone has an entirely im agistic mental life like that required by the imagistic thesis, and, so, it is hard to believe that Locke did. 34 It certainly seems to be a far stretch to think when faced with ideas of this level of abstraction that Locke thought they must all be images when such images must be exceedingly rare, if they happen at all. Ergo, (2*). Now, it is quite possible that Locke did not think this through sufficiently. Maybe he is just wrong about this point. However, as I am arguing, the theory presented in the Essa y does not require that all ideas are images. The above considerations my revised argument suggest that since that theory is presumed, by Locke, to accommodate such ideas as animal justice property adultery and incest the theory is either severely flawe d or holds that not all ideas are mental images. With this dissertation, I hope to show that it is not severely flawed, at least in this regard. 34 There is an episode of Star Trek: The Next Generation in which Captain Picard encounters an alien whose way of thinking is based entirely upon allegory. So that when one wants to get across the idea of betrayal, for instance, one has to reference an arche typical case of that behavior. While I am not at all sure that this way of thinking could possibly get off the ground, it does seem to be what would be required to deal with an imagistic reading of ideas like adultery and justice Pointedly, the crew of th e Enterprise thought this way of thinking was bizarre and quite unnatural (at least for humans).
44 It is perhaps worthwhile here to summarize the discussion up to this point. I have considered the initial reaso ns that Ayers gives for thinking that Locke should be understood to be an imagist. I have argued that none of those considerations have sufficient weight to counter balance, let alone over ch supplies the points Ayers employs. I have also offered an explanation as to how Locke should be read which respects that claim of neutrality. his response is inadeq uate, largely because the argument he presented is itself rather weak. I then offered my own version of an argument against the position that all Lockean ideas are images, and showed that it does not run afoul of the objections Ayers raised to the initial version of the argument. I trust that this alternative argument emphasizes the positive thesis of this chapter, that Lockean ideas should be seen to include both images and more conceptual notions. I take myself to have shown that, so far, Ayers has given us insufficient reason to accept his imagist thesis. However, Ayers is not finished, and it remains to be seen if my own non imagistic reading of Locke will do the work it clearly needs to do. 2. 4 Imagism and the most fundamental and conclusive argument for taking ideas 35 and it is to that argument I now turn. It is important, though to first note how this argument is suppose d to be working. From the beginning of this argument, Ayers assumes that Locke is an is an imagist and his theory is imagist are sprinkled liberally through 35 Ibid.
45 knowledge. In effect, I take Ayers to be saying, This fits in well with the argumentative methodology Ayers established in his chapter. If Ayers were right th at he had adequately supported the position that Locke is best understood as an imagist and defended that position against the apparent argument against it, would clearly add support to thinking Ayers is correct to think of Lockean ideas as all images. However, I have given good reasons to think neither of the two parts of the antecedent have been established, which suggests that the most Ayers can hope for here is to show that an imagist adequate non and conclusive argument for taking ideas as own not strictly imagist position on the same theory. I will argue that my reading makes better sense of Lock thinking Locke an imagist actually work. Essay is above all an imagist theory of the a priori perception 36 Whether or not he is 36 Ibid.
46 uses that doctrine to try to come up with an explanation of such a priori 37 intuitions as the truth seventeenth century philosophers, geometry supplied Locke with his chief paradigms of a priori 38 Ayers again draws our attention to the d ebate between imagism and intellectualism and points out that when it came to geometry the point of contention intellect to form or make explicit its purely mental 39 Ayers goes on to particular ideas (e.g. th e equality of certain angles) literally with our eyes, and this knowledge becomes universal in so far as we employ those ideas as representative members of classes g 40 Ayers then quotes a passage of the Essay (4.17.8) which emphasizes that the legitimacy of our universal claims depends entirely on the relationships between our individual, particular ideas and whether or not they truly represent the thing s we take them to represent. I have no real problem with any of this line of thinking as it is spelled out, both here and reasoning seems to be that if Locke were not an 37 a priori started it. 38 Ibid., 50. 39 Ib id. 40 Ibid.
47 a priori knowledge and understandin himself to cases that are easily amenable to the imagistic correspondence to geometrical examples Ayers is suggesting, and in fact good reason to think that he did not. First, it hardly seems likely on the considerations that I mentioned in the previous section that Locke could have thought all cases of universal knowledge were so amenable, unless he wanted to rule out any universal claims about justice for instance 41 This would be a strange move for Locke to make, since it seems we can clearly work out the limits of that idea and compare it with relevant others, even though it does not seem to make sense to think of justice imagistically. More specifically, Locke says Every Man is an Animal or living Body, is worked out with certainty on the imagistic/geometrical model Ayers sugges ts. If my (imagistic) idea of animal happens to consist of a jellyfish, partially considered, it seems quite a stretch to suggest that someone would by an imagistic comparison see the appropriate correspondence with my representative image of man Even mor have knowledge of the existence of God, in 4.10. Here the main problem is the idea of God itself. Even if we manage to somehow have an image in our minds of infinity perfection and omnipotence which I of something that has these attributes if we are going to use that idea to make knowledge claims about it. 42 41 To do otherwise would seem to push Locke toward emotivist ethics, which seems implausible, and would be an injustice to Locke. 42 a ttributes to make meaningful claims about God, which, again, he does think we can do.
48 If, on the other hand, we are not restricted to imagism we avoid these proble ms, though of course there may be others. In the following chapters I will argue that more abstract, complex Lockean ideas, like justice and God are simple combinations such that we can effectively reach a priori universal knowledge if we have been carefu l in delimiting the sets of ideas included in those (complex) ideas. We compare what we mean by God with the other ideas we use in that there is a demonstrative proo working out the entailments of the ideas combined in the relevant terms, provided we were clear enough about exactly what we mean by men about which Locke is pessimistic to say the least. In t hese cases, the reasoning is indeed like the deductive reasoning involved in geometry (to asoning involves images. Certainly, there are cases beyond geometry where imagistic ideas are used to arrive at a universal knowledge claim, but, since Locke unquestionably wants us to include universal knowledge claims about ideas that seem implausible on an imagistic reading, 43 his theory had better not demand that all ideas are images. Fortunately, as I have shown here, and will continue to argue in not require that all ideas are images. So, since the combined simple combination and not exclusively imagistic reading that I am developing covers more of the universal knowledge claims that Locke, and we, would not make the best sense s story about those claims. As a result, Ayers has not given us any real reason to think 43 For instance, Locke emphatically says that we can have knowledge of moral truths, which involve a great many ideas that seem quite implausible on a strictly imagistic r eading.
49 that all reason to support such a strong claim. 2. 5 Images and Ideas To reiterate the position I suggested in Â§1, and to foreshadow the discussion of the following chapter, it makes the most sense to think that Locke was sincere in claiming that, for him and the E ssay whatsoever is the Object of the understanding (without supposing an external existence to that thing 44 ) is an idea. I have argued that it does not make sense to think of these ideas as consisting entirely of mental images. The positive reasons for taking Locke to have restricted himself in that way are unpersuasive; it is implausible that Locke considered all the ideas he mentioned to be images; and a not exclusively the more limited account. So, in the absence of some other reason to restrict Locke, despite his explicit clai m, it makes sense to take Locke at his word and assume that he is willing to grant the title idea to anything that can be an object of the understanding. In sum, because a strictly ore general reading is consistent with what Locke says in the Essay and does meet his needs, the more general reading of Locke on ideas is to be preferred. those who thought there must also be a non material intellectual faculty, why did he not say something about this in the Essay 44 See Chapter 1 (Â§3).
50 response to this, entirely fair, question that I favor is that he did make his agnostic pos ition clear and explicit, in 1.1.2 (quoted above) and several other places (most notably, in 4.3.6 45 ). Yes, he might have made the connection between his metaphysical agnosticism and the question of imagism clearer and more explicit, but it is certainly no presentation in the Essay is perfectly transparent. As a result of the foregoing discussion, it seems that there are at least two kinds of ideas available to Locke: ideas that are mental images of some sort, and ideas that a re not, at least not entirely. It seems worthwhile to consider at least some of the options available for combining images and ideas: ( i) An idea is nothing but an image (understood as an impression). ( ii) All ideas have images as necessary components. ( ii i) All ideas happen to have images as components, but this is not a necessity. ( iv) All ideas might have images associated with them, but some do not have them. ( v) Some ideas cannot have images associated with them. 46 Ayers seems to be defending the positi on that Locke held (i), though he is far from explicit about this. One might plausibly think that Ayers really means (ii), just to avoid some of the more apparent problems I have been raising for (i) throughout this chapter. If this is the case, 45 Ideas of Matter and Thinking but possibly shall never be able to know, whether any mere material Being thinks, or no; it being impossible for us, by the contemplation of our own Ideas without revelation, to discover, whether Omnipotency has not given to some Systems of Matter fitly disposed, a power to perceive and think, or else joined and fixed to Matter so disposed, a thinking immaterial Substance: It being, in respect of our Notions, not much more remote from our Comprehension to conceive, that GOD can, if he pleases, superadd t o Matter a Faculty of Thinking, than that he should superadd it to another Substance, with a Faculty of Thinking; since we know not wherein Thinking consists, nor to what sort of Substances the Almighty has been pleased to give that Power, which cannot be in any created Being, but merely by 46 I am indebted to Eugenio Zaldivar for the idea of this taxonomy and much of its content. The comments following are mine.
51 he does no t say as much and gives no clear indication that this is his intention. I believe I have said enough to indicate why I think attributing (i) to the Essay is a mistake. (ii) or (iii) seems more plausible. It does seem that more complicated ideas might be re ducible at least in part to simpler ideas. 47 If the idea of man includes things like walks on two legs has opposable thumbs and does not have a tail (in addition to many other things), it follows that each of those terms is a complex idea, which would be further reducible, or an simple idea might well be a mental image of some sort (including simple ideas of reflection). If, say, leg is not one of these simple ima ges, then each of its component ideas will similarly be either an image or be further analyzable. This would mean that, ultimately, our complex ideas are reducible to imagistic ideas. Does this reading entail, however, that our idea, say, of man includes t hose ideas, or has them as components? I am afraid that Locke says very little to allow an easy answer to this question, but in the next chapter I return to this topic. Adjudicating between (ii) and (iii) is tricky. Given that the scope of the Essay is sp ecifically and explicitly human understanding, it appears that the most Locke could say is that our ideas necessarily have images as components. That reading might be that it is a fact of human nature that all our ideas come from experience, and our experi ence is fundamentally reducible to simple ideas of perception and reflection. To be different in this regard would, arguably, make us something other than human. However, it hardly seems possible for Locke to make the claim that all ideas are so reducible. For all we know God or other spiritual beings might have any manner of non reducible ideas. 47 I return to this issue repeatedly in the next three chapters, but primarily in Chapter 3.
52 (iv) and (v) shift the consideration to what degree explicit images might be associated with our complex ideas. It seems that (iv) must be true, and (v) false. I t ake the notion of association that is operative here to be something like a tag attached to an idea. When I consider the idea cat an image of a cat just comes along with the idea, though that image is not constitutive of that idea. There seems to be no abs particular image with any given idea, even if that association might be odd and arbitrary. Anyone might associate in his or her mind the image of a particular kitten with the idea of democracy either from ment al derangement or decision, or both. This would, I think, be strange, but it is certainly possible. scales tip in favor of (v). For someone to have, for instance, a mental image associated with philosophy seems very hard to credit as the normal operation of a human mind. Note, however, that the fo llowing logical possibility is not on my list of plausible candidates. ( vi) None of our ideas are images. There is no doubt that some ideas are images; ideas of colors seem like undeniable cases of necessarily imagistic ideas. However, as I have argued, it seems that there is a very large number of our ideas that must be mental items that need not be imagistic in any important way. While there may be more kinds of ideas that can fit the overall requirements of serving as our ideas, these two seem sufficient and there is no indication in the Essay that Locke considered any others. In any case, as argued in the previous section of this chapter, these two will do the job Locke needs them to do.
53 One might think that an awful lot of our ideas are imagistic in a r obust sense (involving a strong sense of (ii)). However, I suggest that the reverse is true; the vast majority of our general ideas, for instance, must be of the non imagistic sort, if we understand them as involving images only in an inessential way (thou gh they may be reducible to images, in some sense). While red and sour are almost certainly robustly imagistic, even ideas of as low a level of abstraction as loud and rough seem unlikely candidates for being actual mental images. We might use images to co njure examples of loud sounds and rough textures, but we do not seem to actually carry around memories of loudness and roughness and call them up whenever we want to use the ideas of loud and rough The implausibility only increases as we move up further from the level of bare sensation and reflection, and it seems that the overwhelming majority of our ideas abide in the region between loud rough and color and lofty abstractions such as truth and b eauty Relatively speaking, then, only comparatively few of our ideas can plausibly be understood to be images in such a robust sense. Of course, the foregoing leans very heavily upon what I myself find plausible or implausible. I am not sure how to suppor t my position more securely beyond asking the reader to consider their own mental experience and intuitions. This all leaves us with the question, if some of our ideas are not images, what are they? Also, based on what Locke says in the Essay and accepting that we have these two types of ideas, what can we say about ideas as such, not just as images or non images (or as composites) but as ideas? In the next chapter, I develop a reading of Locke that attempts to answer these questions
54 CHAPTER 3 L On the basis of the reasoning in the last chapter, it seems to make sense to think of Locke a s committed to our having both imagistic and non imagistic ideas. This leaves us with two basic and important questions: if they are not all mental i mages, what exactly are our ideas; and how 1 This is perhaps an overstatement, but there is certainly some truth to it. However, it stands to reason that the nature of our ideas as such must have some bearing on what else can be said about them. In this chapter I will develop as much of a positiv e position as is possible, given what the Essay says. In regard to the second, Locke never makes the distinction between imagistic and non imagistic ideas explicit, let alone explains how the two kinds of ideas are related. Nowhere in the Essay does he s uggest that there are two independent systems of ideas operating in the understanding. Rather, Locke seems to have thought that all of our ideas, of whatever sort, can be accounted for by the theory he presents. In this chapter, I will offer a way to under stand how this works in the Essay I argue that we can make sense of these issues by adopting a fairly simple and straight forward reading of Locke. First, in Â§1, I will offer a preliminary discussion of what Locke says about our ideas as such. Then, in Â§ 2, I consider five commitments that Locke has regarding ideas: empiricism, both imagistic and non imagistic ideas, ideational atomism, the reducibility of all complex ideas to their simple ideas, and the distinction between particular and general ideas. Le aving the last 1 G.J. Warnock, Berkeley (Notre Dame, IN: University of Notre Dame Press, 1982), 39.
55 commitment to be dealt with in C hapter 5 in Â§Â§ 3 and 4 I consider what the Essay says about each of these commitments in regard to simple and complex ideas, respectively. I explain that what Locke says seems to leave us with several puzzles. In Â§5, I develop an account of how these puzzles can be resolved. Finally, in Â§6, I return for a concluding word about what Locke thinks our ideas are. 3. 1 Preliminary Distinctions As I mentioned at the beginning of the previous chapter, Locke says the following about : It being that Term, which, I think, serves best to stand for whatsoever is the Object of the Understanding when a Man thinks, I have used it to express whatever is meant by Phantasm, Notion, Species, or whatever it is, which the d not avoid frequently using it (1.1.8). This is almost all that he directly says about the nature of our ideas as such. He goes on to categorize our ideas as simple or complex, and the latte r further as modes (both simple and mixed), ideas of relations, and ideas of substance, but does not take the time to tell us what kinds of things these ideas are what it is, exactly, that the mind can be employed about in thinking. Instead, the next thing Ideas ce which I am going to call ideas, whatever those are, and we all know that we have them; nothing eds into his arguments against innatism. What Locke later says at the beginning of Book 2 on ideas, adds nothing to this. He repeats that we all know that we think, that we think about ideas, and therefore have ideas. The first question he then addresses
56 (2.1.1). This makes perfect sense, given the anti innatist, pro empiricist project Locke has set himself, but it does not help us with our current inquiry. Of course, Locke does spend a great dea l of time talking about ideas, including the aforementioned categorization, and it would seem to make sense that we are able to come to some basic understanding of the nature of these mental whatevers. As a first step, it is possible to make some claims in regard to some of the possible options. 3. 1.1 Private Things Lockean ideas are internal and private. Locke insists upon this point throughout the Essay adequately to our own. 2 The great challenge of communication is that we might inadvertently thei r ideas for our own. Chapters 9 and 10 of Book 3 are particularly concerned with this issue, and clearly emphasize the private nature of our his own Ideas to Ideas furnished to us by experience (which includes the operations of the mind upon the raw deliverances of said experience), which involves a great deal of commonality, but the ideas that result are all stored away in our own, private minds. 3. 1.2 Mental Objects Lockean ideas are mental objects. As the passages at the beginning of this section show, Locke is quite clear that ideas are among the things about which the mind thinks 3 ; they are objects of thought. Locke is consistent in treating ideas in this way. He speaks of having ideas 2 See Chapter 6 for quite a bit more on this. 3 As I argued in Chapter 1 (Â§3), Locke also thinks that we think about exte rnal things, like houses and horses.
57 and considering ideas throughout the Essay and in this regard at least is completely unequivocal. This does not tell us, however, exactly what these mental objects are One of the most fundamental questions to ask in this regard is if we are to understand these mental objects as representations of other things? And, then, are they merely intermediary entities in our minds, requiring the existence of something that is represented, or are they more like intentional objects th ings about which we think that do not necessarily require the existence of their objects. It has seemed to a number of commentators that understanding Lockean ideas as representative objects in either of the above senses is objectionable. To do so is to a dopt an i ndirect realism that invites skeptical worries to the party. Some, like John Yolton, have insisted that Lockean ideas are nothing over and above perceptions. This is apparently intended to distinguish them from objects with special ontological sta tus, and is generally employed to deny ideas, resulting in just such a merely mediated connection with the world. 4 While Yolton acknowledges that our perceptions c an be the objects of our thinking, he denies that ideas should Essay that Locke 5 6 Yolton takes this as a way to disarm the representationalist threat, which would invite either skepticism about our knowledge of the external world or Berkeleyan idealism. Our ideas do not sta nd in the way of our direct perception 4 John Yolton, (Cambridge: Cambridge University Press, 1970), 126 137. 5 Ibid., 134. 6 Ibid.
58 of the world, but are the way we perceive the world. Locke does say a number of things to at what time a Man has first any Ideas, is to ask, when he begins to perceive; havi ng Ideas seems to be clearly saying that to have ideas is just the same as perceiving. However, Locke also says quite a bit that lends support to the representational reading. Much of what he says in Books 2 and 4 of the Essay about our ideas and our knowledge of things external to us, can only be understood in this way. For one thing, he s ays that our minds Ideas out of our mental perception. It does seem odd to think that we do not continue to have ideas when we do n ot perceive them. There may be quite a long time between instances of my thinking of koala for instance, but it does not seem to make sense to think that I do not have the idea in those intervals, and Locke does not speak as if he thinks this is the case. This seems to suggest that Locke did indeed have a representationalist understanding of ideas, that our ideas should be understood to be mental objects, and this does seem to be the best, most direct reading of the Essay E.J. Lowe, however, raises pro objects or entities or things of a special kind, which indeed can stand in a genuine relationship to the subject or mind 7 reasoning is c haracteristically concise, so I will quote it in its entirety: 7 E.J. Lowe, Locke on Human Understanding (New York: Routledge, 1995), 41.
59 Now, things that can stand in a genuine relationship to one another are normally indeed, perhaps always and necessarily logically independent of one another, in the sense that either could, lo gically (even if not naturally), exist in the absence of the other. Thus, for example, given two human beings related as father to son, although it is not naturally possible for the son to have existed in the absence of the father, nonetheless it appears l ogically possible that he should have done so, because these two people are, as Hume would put not existenc e in the absence of that subject, either as a free floating and another subject. In the same way, there could not be a pain which was not the pain of a particular 8 Lowe takes this to show that there is a fundamental problem with thinking of ideas as objects at moment, but there are two things reification argument first. First, it is not at all clear why we should give so much weight to the claim made in the relationship, but Locke can be understood as specifying a special kind of relationship, a particular kind of ownership, which does not involve the kind of logical independence Lowe and Hume think holds to their owners in this particular way, and this ownership entails that if the idea or pain exists it must, necessarily, be so related to that particu lar owner. I cannot see why this would be problematic. Second, it is not clear that it is true that we cannot sensibly say that some idea could exist about the l ogical independence of the father and son is undermined. While it does seem to make 8 Ibid., 41 42.
60 much less clear that sense cannot be made of, say, my idea of monkey ex mind, even if I do not exist. There seems to be no conceptual difficulty in supposing that the and there seems to be no problem with imagining that the idea stands in the same representational relationship. Only the fact that my idea of monkey is my idea seems to be an issue, or to put it a bit differently that the one end of the representational relationship ( x represents y for S ) is t he only thing that changes. 9 However, this is no less the case for the father se the same individual. 10 So, it seems that if there is a problem with the logical independence of our ideas there must also be a problem with the logical independence of fathers and sons. 11 While I do not have a considered position on either of these two points, it seems clear that ing for rejecting the reification of ideas at least needs more support S S senses redl 12 9 There seems to be a type token issue here. The idea of monkey is the type that can be tokened in various minds. However, in this case the type can be specified (or, as Locke would say, determined) rather completely, with the only difference between tokens being in whose minds they are tokened. The problem seems to b e that the same thing son and can be specified completely, but is tokened of a different father. 10 Granting that the son is the same individual in this case seems to be gr anting a great deal, actually. 11 Alternatively, one might be thinking that it is logically possible that the son exist without being son, while it is not possible to be an idea or pain without being an idea or pain had by someone. This simply seem s to involve different understandings of what is included in the idea of a person. It seems perfectly reasonable to think that being a normal human being is a part of the idea of any particular individual, and that seems to logically require s son (or daughter). Locke certainly seems to think human being is included in ideas of particular humans, so this way of strengthening the objection does not seem to gain traction for Locke. 12 Ibid., 42 45.
61 to take Locke to have this position. 13 I find this puzzling to say the least, because Locke never says anything remotely like S ideas are mental objects. The position Lowe, Yolton, and others take on this issue seems to be a result of only worrying about occurrent perceptions of things (e.g. my idea of the compute r screen in front of me right now or the refreshment I have while drinking some cold water ). 14 The worries and arguments here are deep and tangled, but none of it will do for answering the question of what Lockean ideas as such come to, for those ideas clearly and explicitly include things other than occurrent perceptions. Locke thought we have par ticular ideas of memory and general ideas, like liberty and philosophy and it is very hard to see how anyone can think an adverbialist account will do for those ideas. Whether or not the adverbialist way of thinking about some ideas makes good philosophi cal sense, there seems to be no good reason to think Locke thought of ideas in general this way. Other than avoiding the reification issue mentioned above, Lowe seems to think that the adverbialist approach scores slightly better than an imagistic reading of Locke, but, as I have argued, it seems clear that we should not understand Locke as an imagist. If the imagistic reading is off the table, we can avoid the objections to treating mental objects as objects, and there is no positive reason to ascribe an adverbialist reading to Locke, 13 Ibid., 45 ss clearly adverbial than Lowe does. See also, the aforementioned book by Yolton, as well as Perceptual Acquaintance from Descartes to Reid (Minneapolis: University of Minnesota Press, 1984), 88 104. George S. Pappas also considers an adverbialist reading of Berkeley in (Ithaca, NY: Cornell University Press, 2000), 11, 128. I.C. Tipton gives a useful survey of this and related positions The Empiricists: Critical Essays on Locke, Berkeley, and Hume edited by Margaret Atherton (Lanham, MD: Rowman & Littlefield, 1999), 1 18. 14 For an illuminative discussion of the adverbialist position and some of its problems see Thomas M. Lennon, Pacific Philosophical Quarterly 85, 3 (2004), 322 360.
62 then it seems to me that the best way to read Locke is as thinking that our ideas are mental objects of some kind. In any case, rightly or wrongly, Locke does consistently treat ideas in this way. However, there are still the two options in this regard; either our ideas represent as intermediaries between our minds and their objects, which must stand apart from the mind, or they are better understood along the lines of intentional objects (i.e. things about which we can thi nk but which do not necessarily entail the existence of the things they represent). Here it Although, a s I have argued in Chapter 1 (Â§3 ), Locke clearly thinks that we can and do think about external things, his explicit disavowal of metaphysical commitments means that he is not really entitled to assume the existence of external bodies. In Book 4 of the Essay Locke addresses this concern, in part by emphasizin g the passivity of our reception of simple ideas (to which I return shortly) and the fact that we cannot help but suppose that external things exist, but none of this completely resolves the tension at hand. This strongly suggests that we should understand Lockean ideas as intentional objects in some sense. This still does not tell us exactly what kind of thing these intentional mental objects are. Locke seems to have been reluctant to take a stand on this question. He said, apparently as a complaint, th what kind of things you make these same ideas to be 15 Nowhere in the Essay does he give anything like a clear statement. If he had the foregoing dispute would have, perhaps, been settled by now. Instead, while he seems to have been aware of the worry, he did not address it directly. Perhaps this was because he did not feel up to the task, or was simply not particularly 15 The Locke Newsletter 2 (1971), cited in Tipton, 1 2.
63 interested. There is, I think, some reason to think that addre ssing the issue was outside the scope of the Essay even though that did not stop Locke from weighing in on other topics that were extraneous to the Essay reluctance stems from the me taphysical neutrality I have mentioned above and stressed in Chapter 2. This will become clearer in the following sub section. 3. 1.3 Imagination or Intellect One might also wonder if and how Locke acknowledges a Cartesian distinction between the intellect and the imagination. Certainly Locke was aware of Descartes, and it seems reasonable to suppose that he must have had some position on this issue. Unsurprisingly, this is closely related to the discussion in the previous chapter (Â§2) of the Cartesian doct rine of a such a memory available to his theory, and therefore must have only thought of memory as pertaining to that of imagination, which was restricted to imag istic content. Here, the distinction is between ideas that we receive from our sensory organs, which must thereby be imagistic, and are the basic constituents of ideas of imagination, and ideas we have which could not be so received, largely because they i nvolve some sense of perfection, which are inherently clearer and more distinct and are ideas of the intellect. The latter set of ideas includes perfection infinity mathematical and geometric ideas, and God 16 Although he does not make his position expl icit, Locke does not acknowledge this distinction. There is simply no division along these lines in the Essay Locke was quite fond of 16 It seems that Descartes would only classify the contents of the intellect as ideas, but neither Locke nor Ayers confines themselves in this way.
64 points to be some reason to think the absence of this distinction in the Essay is deliberate. There are at least two good reasons for Locke to ignore this distinction. The first is his empiricism and anti memory of the intellect and ideas thereof is that he thought it was impossible to explain our having those ideas by way of experience. Locke denies this premise, insisting instead that we derive all of our ideas from experience. (I return to this point in the following section.) This empiricism is incompatible with any sort of innate ideas, including those Descartes thinks we must have. For Locke, if it is true that we cannot get an idea by way of experience, then we do not have that idea. If Locke were to acknowledge a distinction between ideas of the intellect and of the imagination, he would need some non innatist story about how we get each of these sorts of ideas, but he is committed to one source of all ideas, experience. 17 agnosticism on just this point of a separate intellectual questions such as whether or not we have a separate, non physical intellectual memory and associat ed ideas. As I argued, that neutrality will push Locke toward simpler ontology. In this case, maintaining agnostic neutrality about separating intellect and imagination supports not making the distinction between two different sorts of ideas. What we know is that we have ideas, whatever those are, regardless of whether we are both physical and non physical or merely 17 Alternatively, he might acknowledge the distinction, but just deny that we have any ideas that do not come from experience. Perhaps a Lockean empiricist cou ld say, in principle there are both ideas of the intellect and of the imagination, but, because there are no innate ideas, we do not happen to have any ideas of the intellect. I am aware of nothing in the Essay that suggests Locke was thinking along these lines.
65 project. 3. 1.4 Summary agnosticism, offer a basis for understanding why it is that he does not give us a more substantive positive account of the ontological status of our ideas. To do so wou ld, perhaps, take him too far into speculations he has ruled out of bounds. Locke might plausibly think, given the state of the discussion at the time, that saying exactly what kinds of objects ideas are would depend on metaphysical issues that are themsel ves beyond the scope of the Essay At least, it seems reasonable to think that Locke might have thought this. Alternatively, it is possible that he was simply not interested in this particular question. At the very least, it does seem unlikely that Locke w as concerned about the same kinds of issues we are today, and unreasonable to expect him to respond to worries prevalent centuries after his time. This does not mean that we cannot say something our ideas. Recalling what has been said in the previous two chapters and the foregoing, we can see that for Locke ideas must be private mental (intentional) objects about which we can think. 18 We do not know what Locke thinks those mental objects are really, but then he either has no apparent interest in this question or he considers this to be outside the scope of his project, or both. Still, by carefully considering the Essay it is possible to get somewhat clearer on the nature of Lockean ideas, not in regard to their ontol ogical status perhaps, but in regard to how it 18 Again, as discussed in Chapter 1 (Â§3), Locke clearly thinks that we can think about external objects, but does not mean to include those as ideas. Such external things are (more or less) public non mental objects of thought.
66 is that Locke thinks they are supposed to work, and how it is possible for us to have both imagistic and non imagistic ideas. 3. 2 Locke has five commitments that shape and constrain what i t makes sense to say about our ideas: 1) we get all our ideas from experience; 2) we have both imagistic and non imagistic ideas; 3) we have both simple and complex ideas; 4) our complex ideas are ultimately composed of simple ideas; and, 5) we have both p articular and general ideas. In the following, I will say a bit more about each of these commitments. ultimate source of all our ideas is experience. Locke begins Book 2 of the Essay with the supposition that, since we lack any innate ideas, we must start with what is often called a tabula rasa Ideas 19 As I have mentioned above, Locke thinks t here can be no real doubt that we do have ideas (2.1.1), so in one word, From Experience : In that, all our Knowledge is founded; and from that it ultimately deri external sensible Objects the internal Operations of our Minds of Knowledge, from whence all of the Ideas must somehow be derived from experience, from the most basic ideas to the most abstract and complex. This is not to say that said derivatio n is a simple matter of observing external objects 19 Whether Locke has su ccessfully defended his anti innatism position is a question I hope to return to in the future. For now, it is beyond the scope of this dissertation.
67 and internal mental operations. Much of what Locke says in Book 2 involves explaining the various operations of the mind that we bring to bear upon the ideas we get directly, or immediately, from experienc e to all the rest of our ideas, whatever those might be. (2) As I argued in the previous chapter, and mentioned above it seems those ideas that are ultimately derived from experience must include both imagistic and non imagistic ideas. As noted in the fo object of thought must be an idea, and clearly we can and do think about m ental images. However, as I have argued, some of those things should not be thought of as images in any meaningful sense. So, Locke, or at least the theory of the Essay must be committed to both imagistic and non imagistic ideas. (3) Another basic distinction between types of ideas one that Locke does explicitly make is between those that are simple and those that are complex. Although we experience dista Ideas they produce in the Mind, enter (2.2.1). When we experience things, we receive many ideas all at once and in succession, but these are all conglomerations of wha t Locke calls simple ideas, nothing but one uniform Appearance or Conception in the mind, and is not distinguishable into different Ideas ortly see, not all of these simple ideas seem to be imagistic, and the distinction between appearance and conception in this passage suggests that perhaps physic al things in the world are experienced by us in everyday life as macro objects but on the
68 fundamental level are really just collections of more basic constituents, so are our ideas of those things. The most fundamental level is that of the simple ideas. Th is suggests a sort of ontological ranking. Just as a (non ideational) atomist might say that the atoms are the really real objects, Locke seems be giving some kind of privileged status to the simple ideas. He is saying that while we naturally receive what seem to be many complex ideas those are actually just combinations of simple ideas, they have a kind of mental ontological primacy. Whether or not this is the right thing to say, Locke is explicitly committed to this position, and the special status of sim ple ideas Chapter 4). Ideas into eceive all our simple ideas passively; they are thrust upon us by our experience of the inner and outer worlds, so to speak. As we have seen, what we get from that experience is a collection of simple ideas. However, our minds have the capacity to do a num ber of things with those simple ideas, including combining them into complex ideas. As Locke says: As simple Ideas are observed to exist in several Combinations united together; so the Mind has a power to consider several of them united together, as one Id ea ; and that not only as they are united in external Objects, but as it self has Ideas thus made up of several simple ones put together, I call Complex ; such as are Beauty Gratitude a Man an Army the Universe ; which though complicated of v arious simple Ideas or complex Ideas made up of simple ones, yet are when the Mind pleases, considered each by it self, as one entire thing, and signified by one name. (2.12.1) This is not to say that some of these ideas which we now combine as a complex idea did not occur together before we did so, as in man or monkey but that ultimately all we get from experience are simple ideas, which we then re unite as complex ideas.
69 In the following two sections, I will say more about each of these kinds of idea s, simple and complex, respectively. (4) The foregoing leads Robert Hanna to claim that thesis to the effect that every complex idea allows of a complete, finite, decompositional analysis into its simple sensory ideas tog ether with the simple ideas of various mental operations on the simple sensory 20 That is, given any non theory, be analyzable into relatively simpler ideas until everything has been reduced to simple ideas of either sensation or reflection. This Lockean commitment requires quite a bit of further consideration. For instance, we need to understand how the composition of our ideas is supposed to be analyzable in this way, particularly given the requireme nt that there be both imagistic and non imagistic ideas, which might both be involved in the analysis of a single complex idea. I will offer a reading of Locke that will attempt to explain this in Â§5. (5) Finally, Locke holds that, while only particular t hings exist, some ideas are capable of Other, non general ideas are only of particular things. So, my particular idea of this computer just stands for this particular computer in front of me; it does not refer to any other computers or anything else. My general idea of computer though, stands for any and all computers, provided my idea is adequate to that task. The consideration of how this is possible will be developed in the following two chapters, on representation and abstraction. To summarize, Locke is committed to two theses and three distinctions. The theses are those of empiricism and atomism, which we can express as one claim: all of our ideas are or are 20 The Review of M etaphysics vol. 44, no. 4 (June, 1991), 787.
70 reducible to, simple ideas that we only get from experience, either of sensation or reflection. complex ideas, imagistic and non imagistic ideas, an d particular and general ideas. It seems that each idea fits into one of each of those two options. That is an idea can be simple, imagistic, and particular; complex, non imagistic, and general; simple, non imagistic, and particular; etc. However, it is no theory. any positive understanding of ideas that theory offers must be compatible with the commitments introduced above. In the following I will attempt to develop such an understanding. First, though, it will be helpful to get clearer on exactly what Locke says about simple and complex ideas. 3. 3 Simple Ideas In the following, I will consider what Locke says about simple ideas and how this informs his position on each of the commitments discussed above, not including the distinction between particular and general ideas. It will be seen that Locke has no problem with the remaining three of those commitments in regard to simple ideas. 3. 3.1 Empiricism and Atomism As I have already mentioned, Locke holds that, although we experience the world in a great combination of ideas, those ideas are all simple ideas. That is, the basic, irreducible constituents of our experience of the world, both outer and inner, are simple ideas. Naturally, Locke distinguishes between simple ideas of sensation and simple ideas of reflection, which are
71 the only ways we could possibly get such ideas (2.1.2 5). 21 Simple ideas of sensation include both those received by just one sense, like those of individual colors, sounds, and tastes, and those received by more than one sense. For example, Locke says, The Ideas we get by more than one Sens e, are of Space or Extension Figure Rest and Motion : For these make perceivable impressions, both on the Eyes and Touch; and we can receive and convey into our Minds the Ideas of the Extension, Figure, Motion, and Rest of Bodies, both by seeing and fee ling (2.5). Perception or Thinking and Volition or Willing This grammatical structure clearly indicates synonyms. That is, Locke takes himself to be listing just two different simple ideas of reflection, not four, just perception and volition not those plus thinking and willing pleasure pain power existence and unity (2.7.1). Moreover, as sugges ted by the foregoing, the reception of these simple ideas is basically automatic. Locke tells us that as soon as we perceive we have ideas Ideas and Mi 22 We do not really choose which simple ideas to have, and we do not have any control over the content of those ideas. Locke emphasizes n the reception of simple ideas of sensation and reflection (2.12.1). There are two important restrictions on this point, though. experience. If there is a simple idea to be had, say that of a particular color, and one never comes 21 Not only is there no way for our minds to create new simple ideas, we also cannot willfully destroy those we have; see 2.2.2. 22 make their Approaches to our minds, and make ideas.
72 23 The second is as. As Locke puts it: [T]hough he that contemplates the Operations of his Mind, cannot but have plain and clear Ideas of them; yet unless he turn his Thoughts that way, and considers them attentively he will no more have clear and distinct Ideas of all th e Operations of his Mind and all that may be observed therein, than he will have all the particular Ideas of any Landscape, or of the Parts and Motions of a Clock, who will not turn his Eyes to it, and with attention heed all the Parts of it (2.1.7). So, although we will receive simple ideas from experience, the ideas we get depend upon how we direct our attention. To have clear and distinct ideas, we must consider our experience attentively. 24 It might be worried that we actually have to perform abst raction to get simple ideas like unity or existence white suggests an active process of the understanding (2.11.9). This would argue against what I have said about the automatic reception of our simple ideas. However, the point about the need for careful attention just made suggests a response to this problem. Experience gives us the ideas of white unity and existence but it requires some level of focus to discern these ideas well enough for them to count as clear and distinct. Once we focus our attention sufficiently, the reception of simple ideas is 23 Although it seems Locke would have considered this a good example there are at least two problems with it First, we can actually have an idea of that missing color. Second, it is not apparent to me that a color should be understood to be a simple idea at all. Each color seems to be further analyzable into shade, hue, etc., which would mean that it 24 simple Ideas are clear when they are such as the Objects themselves, from whence they were taken, did or might, in a well this seems to be guaranteed, as long as our sensations or perceptions are well ordered (whatever that comes to), but distinct Idea is that wherein the Mind enters the picture. If we are not attentive to the constituents of our experience we might not perceive that there are really, say, two distinct ideas presented to us, rather than just one.
73 automatic, even of ideas as abstract as unity and existence My phrasing is deliberate; the attention Locke is describing would seem to rise to the leve l of abstraction to arrive at such ideas. white referenced above, takes place in his initial treatment of abstraction to which I return in Chapter 5. e demands of his commitment to empiricism. Additionally, because they are the foundational elements of simple ideas. 3. 3.2 Imagistic and Non Imagistic Simple Ide as Some of the ideas Locke considers simple should clearly be understood to be imagistic. Locke seems to think of those of individual colors, sounds, and tastes, for instance, as images, and I can think of no good reason to think of them non imagistically 25 Even though an image as such must have parts, and therefore could not be simple, some ideas are imagistic in that they are just a basic sensory datum, like, for instance, a particular shade of a color, a particular tone, or scent. Any given instance of these things in experience will be, at least, bounded by a given extension or duration, which would establish relational parts of the complex idea that is experienced, but the simple idea is that of the shade, tone, or scent itself, which is just that and as such does not have parts but is only a particular qualia, to speak anachronistically. The simple sense to think of these simple ideas as images if by im age one is thinking of something like a picture, because, again, this would have to include things like some extension or duration, and so 25 However, as suggested in footnote 23, above, I do have reservat ions about the actual simplicity of at least some of these ideas.
74 could not be simple. As will become clearer shortly, and come to the fore in Chapter 6, this seems to have serious im On the other hand, some of the ideas Locke lists as simple are improbable candidates for imagistic content. The ideas of extension figure and motion for instance, cannot be both simple and images. Following the logic of the previous paragraph, if these ideas are images, they have to be images of some particular extension, figure, and motion, and as such they will be composed of parts and thus not simple enough to count as simple ideas. As I argue in Chapter 6, the best wa y to understand these ideas is as having extension (or being extended ), having figure ( being figured ), and having motion ( being in motion or moving ). These are simple enough, but it seems rather clear that they cannot be understood as images. The only way to have an imagistic idea of having extension would be to have a token image of an extension perhaps then using that one imagistic idea to stand f or all cases of having extension 26 and I have just pointed out that this would not be a simple idea, and this applies equally to the other ideas in this list So, either extension figure and motion are simple or they are images, not both. Locke explicitl y says they are simple ideas, therefore they must not be images. 27 Similar considerations suggest that pleasure and pain as such, should not be understood imagistically. While it might make sense to have some token pleasure or pain experience stand as a mental image to represent all instances of pleasure or pain, it seems that these will at least have some duration and therefore not be quite simple enough. If sense can be made of Locke thinking of these ideas non imagistically it seems to make sense to op t for that reading. Following the line 26 Berkeley seems to have something like this in mind in his Principles of Human Knowledge edited by Howard Robinson (Oxford: Oxford University Press, 1996), Introduction 12, 13 14. 27 See Chapter 6 for much more on this point.
75 of thought suggested above, thinking of these two ideas as being pleasurable and being painful seems to make sense. No imagism is required, and the ideas retain their simplicity. 28 Given the list of ideas we have been considering, the least likely candidates for imagistic simple ideas are existence and unity To put it mildly, it is very hard to see how these can be understood imagistically at all. Even if the previous paragraphs have been unpersuasive, here must be tw o simple ideas that need to be understood non imagistically. While considering unity Minds brings this Idea Locke cannot mean that every idea has unity as a part of it, because then there would at most be only one simple idea, unity and as we have seen Locke explicitly lists other simple ideas. So, unity We cannot have any thought (as a bit of thinking) without unity being somehow involved, or present. Clearly this explicitly include any reference to unity Beyond the negative characterization offered above (i.e. that it cannot mean inclusion), Locke does not make the involvement or presence he intends in this case any clearer. What is clear, though, is that on Lock unity 28 These considerations suggest that perhaps simple general ideas should be thought of as types. I hope to pursue this line of thought further in the very near future.
76 ev ery complex idea allows of a complete, finite, decompositional analysis into its simple sensory ideas together with the simple ideas of various mental operations on the simple sensory 29 eas are means that the ideas are simple components of sensory experiences, mental images, then Hanna is claiming that all complex ideas are analyzable into the simple ideas of various mental operations on imagistic simples not being involved in the analysis. Since the non imagistic simple ideas are simple, they cannot be further analyzed, so they would have to be present in the base level of the analysis. Therefore, if Hanna imagistic simple ideas cannot be constituents of complex ideas. This seems to fly in the face of things Locke clearly wan ts to say, extension is a simple idea involved in the complex idea of any particular geometrical 30 This suggests that we not take the thesis Hanna offers quite literally. Instead, we every complex idea allows of a complete, finite, So, the Essay includes both imagistic and non imagistic simple ideas, and consideration atomism. All of this will be important as I turn to the subject of complex ideas. 29 Hanna, 787. 30 deliberately vague at this point in the chapter.
77 3. 4 Complex Ideas Locke distinguishes several sorts of complex ideas: simple modes, mixed modes, ideas of relations, and ideas of substance. While the differences between t hese sorts will be important in understanding of these ideas poses some interpretive problems. It will be important to get clear on these issues before consider In the following sub sections, I focus on getting clear on exactly what Locke says about these ideas in regard to how we get them (empiricism), their composite nature (atomism), and the distinction between imagistic and non imagistic ideas, and then indicate some of the concerns raised by these points. I will offer a reading that attempts to respond to these concerns in Â§5. 3. 4.1 Empiricism and Atomism As I have already mentioned, according to Locke the ultimate sour ce of all our ideas, whether simple or complex, is experience. However, as the Mind is wholly Passive in the reception of all its simple Ideas so it exerts several acts of its own, whereby out of its simple Ideas as the Materials and Foundations of the rest, the other [the complex ideas] are framed (2.12.1). ideas, by actively performing operations, on those simple ideas. The simple ideas are both the basis on whi ch, and materials of which, these complex ideas are framed. Locke says, As simple Ideas are observed to exist in several Combinations united together; so the Mind has a power to consider several of them united together, as one Idea ; and that not only as they are united in external Objects, but as it self has So, while the mind has the liberty, and power, to create an unlimited number of complex ideas, both as they are observed to come together in experience (a s in our idea of a horse ) and are
78 combined without any such experience (as in the idea of a unicorn 31 ), all of these complex ideas Idea Because our complex ideas are all built on the foundati on of our simple ideas, which we receive exclusively from experience, the complex ideas are also derived from experience. Not only can we build up our own complex ideas from the simple ideas we have, but experience suggests a very great many of our complex ideas more directly. Our idea of horse is primarily suggested by seeing actual horses (which, strictly speaking, we take to be) in the world. We must remember, though, that for Locke that experience of actual horses is given to us as a complicated combina tion of simple ideas. So, c omplex automatically by way of how we receive their constituents, or material. Locke says that complex that by Sensation or Reflection furnished [us] can form by using simple ideas, saying: But all this still confined to those simple Ideas which it received from those two Sources, and which ar e the ultimate Materials of all its Compositions. For simple Ideas are all from things themselves; and of these the Mind can have no more, nor other than what are suggested to it. It can have no other Ideas of sensible Qualities, than what come from withou t by the Senses; nor any Ideas of other kind of Operations of a thinking Substance, than what it finds in it self (2.12.2). ideas are all formed exclusively of simple ideas, and experience is the only source of simple 31 Of course, this only applies to the earliest possessors of unicorn Everyone who has seen a picture of unicorn, for instance, has had the ideas of horse and horn combined in experience.
79 ideas, our complex ideas are derived from experience, despite having been put together by our minds using the relevant mental capacities (combination, comparison, and abstraction (2.12.1) 32 ). 3. 4.2 Imagisti c and Non Imagistic Complex Ideas While again, some of our complex ideas seem to rightly be thought of as imagistic in at least a limited sense, quite a large portion of them only seem sensible non imagistically. As an example of the former, consider the idea one might have of a particular purple polygon. Such an idea will, at least, have a mental image associated with it, which means that there are already at least two ideas involved in the complex idea since the mental image is itself an idea. That the i dea is just a mental image is problematic, though, for reasons I will consider in following chapters. For one thing, though, any given polygon, or an image of one, is extended in space, but it does not seem to make sense to think of ideas as being so exten ded, and as I have argued it does not make sense to think of is extended imagistically. On the other hand, the ideas Locke considers when first introducing the subject of complex ideas make poor candidates for images. Beauty Gratitud e a Man an Army the Universe possible to have some kind of mental image (broadly enough construed) associated with any of these, it is quite hard to see how any of them could be thought of as being made up entirely of images. For instance, to take the idea that seems the most likely candidate for a mental image, gratitude a readiness to acknowledge and return kindness received readiness, but the specific qualifications needed to specify gratitude rather than some other closely related readiness seem to require non imagistic ideas. 32 See also, 2.11.
80 The other ideas in the list are, I think, still more difficult, but there are other complex ideas which seem even less plausible to think of imagistically. For example, Locke tells us that theft the concealed change of the possession of any thing, without the consent of the proprietor might have some kind of pantomime based mental i mage of much of this, until one comes to proprietor which could have an image (say of a shopkeeper) associated with it, but having that image would be very far from having anything like a clear idea of proprietor which must include such ideas as ownershi p and property which in turn seem to be particularly poor candidates for imagistic ideas Other complex ideas that Locke clearly thought we have, or must have thought we have, include murder incest property and philosopher To think of any of these entirely imagistically seems to require considerable mental contortions, particularly if a clear and distinct idea is desired. While it is true that the system presented in Book 2 of the Essay requires a great deal of mental acti vity, Locke offers nothing in that long discussion that is plausibly intended to deal with such problems. While Locke never acknowledges any distinction between imagistic and non imagistic ideas, I take the foregoing to show that his commitment to having both sorts clearly applies to complex ideas as well as to simple ones. 3. 4.3 Concerns Locke explicitly says that complex ideas are in some sense made up of simple ones, but he is not at all clear about exactly how. He says that the complex ideas are frame d out of the y as strictly imagistic. In that case, any complex idea would be a composite mental image made up of more basic imagistic content, in the same sense that a picture of a horse is made up of simpler elements, like the bits that
81 constitute the tail, the mane, the hooves, etc., each of which can be reduced further, until, perhaps, we reach perhaps the perceptual minima, a smallest imaginable imagistic element, which, it should be noted, is not something Locke ever mentions. As I have argued, though, and will ar gue again in Chapter 6, it is not clear that this position will actually work out as neatly as Locke would like if he were an imagist (because any of these perceptual minima would not be simple enough). However, it is not at all clear how we are to under stand the composition of non imagistic complex ideas. Consider the idea of theft the concealed change of the possession of any thing, without the consent of the proprietor understand that the complex i dea theft just is the uniting together of concealed change possession anything without consent and proprietor plus of and the ? Certainly the way the Additi onally, it seems to make sense to think that some complex ideas will combine both imagistic and non imagistic ideas, like a particular color and is extended It is not easy to see how this might be done on the model of theft Ideally, there should be one c omprehensive understanding of how the understanding deals with imagistic and non imagistic ideas. Nothing in the Essay suggests separate theories in this regard. At the very least, w e should not have two completely distinct systems, but they must be relate d, and even interactive. If not, it seems In the following section, I attempt to deal with these issues by developing a reading of what Locke says in the Essay that makes his positio n at least somewhat clearer. 3. 5 The Simple Combination Reading My answer to the question asked above, is the combination of the constituent ideas of in a
82 suggests a surprisingly simple and straightforward understanding of how simple ideas are combined to form complex ideas. Locke explains that one of the basic operations we are able to perform on our ideas is that several of those simple ones it has received from the distinction between enlarging, which results in simple modes, and composition, which he seems to think results in all the rest of our complex ideas, this is all that Locke says about how we assemble our complex ideas. This adds nothing to our understanding of how to make sense of those ideas, except that it suggests that Locke did not think there was much more to say on this matter. This in turn suggests that what he had in mind was something fairly simple and straightforward. I submit that we should take this all at face value and return to what we have seen Locke has said about our ideas thus far. Our complex i deas are the results of combining our various simple ideas. To again take theft as our example, for Locke this just is the uniting together of ideas included in the concealed change of the possession of any thing, without the consent of the proprietor We can split the idea into two complex ideas, the concealed change of the possession of any thing and the consent of the proprietor joining them by way of without Taking just the second idea, we have the ideas of consent and proprietor plus the unstated bu t syntactically implied ideas of the relationships between that consent and proprietor and the proprietor and the thing he or she owns. Just as the syntax matters in language, it matters here, in what should be thought of as a basic, early modern, understa nding of mental discourse. Each of these ideas ( consent proprietor ownership of actions and ownership of property if those latter two ideas
83 are of different relations) is itself a complex idea which is made up of ideas, each of which is either a simple idea or a complex idea which can be further analyzed. This method of analysis is not only applicable to other ideas Locke discusses in the Essay but is generalized to cover all ideas. Any idea is either simple, and therefore unanalyzable, or complex. An y complex idea can be analyzed into ideas that are each either simple or complex. This analysis, or reduction, can be continued until every complex idea has been reduced to its simple idea components. The syntax matters, as above, at every level of the ana lysis before reaching simple ideas. 33 At any point in the analysis, the analysans will be or include non imagistic ideas, imagistic ideas, or a combination thereof. Most of the foregoing simply follows directly from, or is just a restatement of, what Locke explicitly says about our ideas in the Essay The emphasis on syntax is implicit in Locke, and seems to be needed if his account is to have any hope of making sense. The notion of mental discourse is something with which Locke certainly would have been fa miliar. 34 The distinction between imagistic and non incorporation of both types in his thinking. I submit this reading as the best explanation of Lockean compositionality. It is, in some sense a t least, deflationary, but it seems to be the best, most direct, reading of the Essay 33 This is true even for complex imagistic ideas. An image of a horse with a horn sticking out of its chest is not an image of a unicorn even though it has all the same components. 34 For more on this, see Ian Hacking, Why Does Language Matter to Philosophy? (Cambridge: Cambridge University Press, 1975), 15 42. The above discussion, particularly the emphasis on the importance of syntax, suggests a strong connection between my reading of Locke and contemporary language of thought theory. While I do not think i t is appropriate to say that Locke had a language of thought theory, particularly as we currently understand such a theory, it does seem that he very likely had some kind of similar thinking, perhaps operating in the background. I hope to return to explore The Stanford Encyclopedia of Philosophy ( https://plato.stanford.edu/entries/language thought/ 2010).
84 We have already seen what Locke says the idea of theft is. Consider two further examples: beauty elight in 35 and lead simple Idea of a certain dull whitish colour with certain degrees of Weight, Hardness, Ductility, a helpful way to think of them. Note that Locke tells us that t explainin g of one Word, by several others, so that the meaning, or Idea it sta nds for, may be certainly known (3.3.10). If we incorporate the notion of mental discourse we can adopt this s. I suggest that we think of Locke as explaining how we define our complex ideas for ourselves. If I ask myself, theft so I consult the idea, which is represented in my mind as a definitio n built up of simpler ideas. This allows me to know what the content of that idea is. The more careful I am, the greater my confidence in my understanding of the idea. The more time I have put into understanding the simpler ideas involved in the analysis o f the complex idea, the stronger my understanding of the complex actually is. Call this a simple combination understanding of complex ideas. This emphasizes both the direct reading of the text of the Essay and the fact that all our complex idea are ultima tely combinations of simple ideas. The complex idea itself can be thought of as serving as an explanation to ourselves of the meaning of the idea by way of mental discourse 36 35 Beautiful Beauty 36 In such a simple combination, the syntax of the inner explanation is critical. Otherwise, nonsense will quickly result. Even the imagistic ideas w ill have to fit the syntactic rules. This seems less strange if one remembers that what is involved is a kind of explanation involving the indication of certain phenomenal ideas, which need to be indicated at the right point in the explanation to make sens e.
85 It is important to bear in mind one point. As I will argue at several points in following chapters in regard to various ideas (most particularly that of triangle ), Locke does not seem to intend for the combinations he gives for these ideas to be complete. They are frequently an abbreviated explanation of the idea involved. This is na tural given the fact that Locke is committed to the privacy of our ideas. Whenever he explains what he means by an idea it is always with the implicit caveat that he is only explaining what he means by that idea, or what he takes to be the common understan ding of the idea. One might well take each of these with a incomplete ideas; particularly in regard to mixed modes, like triangle whatever idea we have is inherently complete. Rather, his explanation of the idea in the Essay is not intended to be complete. gold for instance, is less combination I have is simpler than his. Hopefully a This means that the metallurgist, Locke, and I have different ideas of gold though they all represent the same shiny yellow metal out in the world. 37 If the combination is different, the idea is different. Remember that what I am calling the simple combination of a complex idea just is that idea. As something of a corollary to the above, even simple ideas have a sort of simple combination. Imagine what someone would say to another person who wanted to know what 37 I return to this issue in Chapter 4.
86 t account it would be impossible to provide further words to explain the ideas; one would have to simply point and try to get the questioner to attend to the right aspects of the examples. So, simple ideas, while not analyzable, have simple combinations, but they are very basic (or as simple as a combination can get). With this reading, I think we can make sense of how to incorporate our imagistic ideas into our com plex ideas. If we think of ourselves as explaining to ourselves what each complex idea means, it is not hard to see that sometimes we are going to have to draw our attention to this some kind of imagistic content. Of course, this might not be a satisfying account of what our complex ideas are, but it does seem to make sense of what Locke actually presents in the Essay In the following chapters I consider how this simple combination reading can be understood in reference to representation and abstraction. I will argue that this reading makes each of these Lockean positions at least a bit clearer. 3. 6 In this chapter I have attempted to answer the two questions what exactly are our ideas, and how do imagistic and non In regard to the first question, I have argued that Locke deliberately avoids answering directly, but that we can say that Lockean ideas are whatever it is that can serve as the private objects of thought without any commitment to the external existence of those objects. (We can think about external things, but that does not thereby make them ideas.) This lack of specificity is perhaps frustrating, but I have also argued that it seems that Locke had reasonable theoretical
87 motivation (regarding his explicit neutrality on certain metaphysical issues) for being so vague. These objects of thought also constitute the elements of what can profitably be understood as our mental discourse. Thinking of ideas in this way allows us an answer to the second question. By adopting what I have called a simple combination reading of Locke on ideas, we can thin k of imagistic ideas as components of complex ideas in the same way we think of ostensive explanations in thinking follow more explicitly and less problematicall y from the text of the Essay I take the foregoing sections to show how Locke can be understood to deal with the two theses and two of the three commitments I discussed in Â§2. In the next chapter, I turn to the question of how our ideas represent that which they are ideas of, which will then allow me to discuss the third of those commitments we have both particular and general ideas and
88 CHAPTER 4 REPRESENTATION Having argued that not all Lockean ideas should be u nderstood to be images and for a simple combination reading of Locke on ideas, I turn now to the question of how to understand what Locke means by representation Not only do we need to understand how our simple ideas of sensation and reflection can repres ent what they do and how our ideas of particular things represent, but we need to get clear on how it is that general ideas are supposed to represent all the various things they need to how things like freedom horse and red represent instances of freedom horsiness, and redness. Moreover, we need a Lockean way of understanding how a non imagistic idea can represent an imagistic one. For instance, it seems that Locke must think the general idea odor will represent both any odor out in the world and our ide as of various odors, which includes our memories of odors. 1 After all, the experience of a smell is the receiving of an idea of a secondary quality, and that quality is nothing more than the power of certain things to cause us to have that idea. 2 So, for m ost of us at least, there is a strong sense in which all odor really represents are the various ideas of odors we have received, although, as Locke emphasizes, we do tacitly suppose that there is something in the world to which our ideas correspond and whi ch we naturally would think is also covered by the idea odor 3 Moreover, while my idea of a particular odor might very well be a kind of image stored in my memory, my idea of odor in general is clearly not an image (though perhaps it might be associated wi th one 1 It seems that we need to distinguish between, say, the smell of a rose and the aroma I recall from the rose I smelled yesterday The former is a general (abstract) idea and the latter is an idea of a particular One might reasonably think that odor really only covers the latter sort of ideas, particular smells, while something like ideas of odors is the right way to think of, say, a set of general ideas of odors. 2 I discuss this more fully in Â§1. 3 I return to this discussion in Chapter 6.
89 for some people). Somehow, we need to have an account of how it is that odor represents ideas like rose smell onion smell and fresh cut grass smell as well as the actual particular smells we e xperience Considering what Locke says in the Essay it seems that there are two questions to be asked: how do our simple ideas (particularly of sensation) represent, and how do other ideas, including general ideas, do so? I will argue that for Locke a simple idea represents what it does because it was caused by that thing and that a non simple (or complex) idea represents just because the idea is used by the understanding to do so. The respective kinds of relationship I am arguing for must each cover both imagistic and non imagistic i deas, to accommodate the two sorts of ideas and the requirements considered in Chapter 3. Otherwise, as a start, we would need representation that covers imagistic ideas (perhaps based entirely upon resemblance of some sort), representation that covers non imagistic ideas, and representation that somehow mediates between these two categories. Locke gives no indication that he is aware of any such multiplicity, so either there is a relationship that covers both sets of ideas that is compatible with the Essay problem. I believe the reading I will offer satisfies the need for a theory that covers both imagistic and non imagistic ideas. In the literature, there are two explanations of Lockean representation regarding sim ple ideas: resemblance and causal theories. 4 In Â§1, I consider resemblance theories of representation and argue that at the very best they offer a severely limited reading of Locke on this subject. In Â§2, I argue that a causal theory is the best reading of Locke on the representation done by simple 4 It actually seems that resemblance theorists often want to apply their position to all ideas, but, as will become clear in the following section, this only seems to make that position less plausible.
90 ideas, including the important one of thing In Â§3, I offer my interpretation of what Locke says in the Essay about representation more generally and develop the use theory of representation I mentioned above In Â§4, I tie this reading of Locke to what I have already said about simple combinations and what I will say next about general ideas (in Chapter 5), briefly showing how Â§5, I which my reading is clearly an example, and argue that he does not offer a reason to reject my position. 4. 1 Resemblance Theories of Representation The b asic idea behind a resemblance theory of representation is either (A) that x represents y (just) because x resembles y or (B) for x to represent y it is necessary that x resembles y It would seem that (A) entails (B). In either case, resemblance is neces sary, but in (A) the resemblance is at least involved in making it the case that the representation happens. As I will argue in the following, however, attributing a resemblance theory to Locke is deeply problematic. Ultimately, my position is that, for Lo cke, while (B) may be true for some sorts of ideas, (A) is never true; neither (A) nor (B) is true for all ideas First, why might one think Locke has a resemblance theory? Donald L.M. Baxter shows that attributing a resemblance theory of representation to Locke goes back at least as far as Berkeley. 5 B axter points out that Locke Ideas of primary Qualities of Bodies, are Resemblances of them (2.8.15), copies of the ideas in the mind (4.3.19). prima 5 Philosophy and Phenomenological Research vol. 57, no. 2 (June, 1997), 307 330.
91 facie 6 object, the idea must resemble it with respect to those q 7 He notes that Hume asked, 8 and also 9 Baxter gives reason to think that Berke ley agreed with Hume on this matter, and indicates that those two and Hume assume, and take Locke to agree, that an idea of something green is a green idea. An id 10 However, it is very hard to see how a resemblance view of either (A) or (B) variety can hold for all ideas unless Locke is committed to thorough imagism, as discussed in Chapter 2. In a footnote, Baxter a cknowledges that this is one of the two most common responses to the resemblance thesis: Locke should not be understood to accept the thesis because he is not 6 Baxter, 313. 7 Ibid., 312. 8 David Hume, A Treatise of Human Nature edited by David Fate Norton and Mary J. Norton (Oxford: Oxford University Press, 2000), I.iv.5.3, 153 (SB 233). 9 ambiguity in the next sentence, Hume has in mind something like the resemblance thesis Baxter is advocating. 10 Baxter, 313.
92 committed to the position that all our ideas are images. 11 I have said enough to indicate why I th ink we should agree with this assessment of the situation. Here I need only point out the limits of the ideas that Locke can plausibly think resemble what they are ideas of (in respect of the salient qualities). As I suggested in Chapter 2, most of our ide as seem to involve images only in a n on essential way; that is, most of our ideas do not necessarily involve images though we might associate images with some of them. Of course, many of our ideas may in part represent other ideas that are imagistic but th at does not mean that they must all include imagistic content as part of their immediate simple combination As I have argued, it is implausible that our ideas of even a fairly low level of abstraction and generality, like loud or rough are actually reme mbered instances of loud sounds and rough textures. 12 So, most of our abstract general ideas are not, or at least need not be, mental imag es. All of t his strongly suggests that a very small percentage of our ideas are sensibly thought to be images, and ther efore plausibly thought to resemble in the sense that, at least, Berkeley and Hume thought. As I have already argued, this is true even if we restrict our consideration to simple ideas; there are too many of those that must be non imagistic to support a ge neral resemblance position even just for simple ideas. While I think this is clearly the right position to have, it does seem, however, to leave us with questions about how we are to understand Locke on those ideas that are imagistic. Do (A) or (B) hold tr ue for those ideas? 11 Baxter, 309, footnote 3. Baxter does not really discuss this response. The other response is that there could be indeterminate images. I do not discuss this latter response because I think the non imagist response I have given in Chapter 2 is suffi cient and decisive. 12 Note that in contrast to Locke points out that we moral Ideas 4 3 .19).
93 For Locke, those imagistic ideas must be simple ideas or modes, 13 and the simple ones must be of primary or secondary qualities. 14 Of course, as Baxter points out, Locke does specifically indicate that some of those ideas do resemble wha t they are ideas of, for Locke says, Ideas of primary Qualities of Bodies, are Resemblances of them, and their Patterns do really the Ideas produced in us by these Secondary Qualities have no resemblance of them at all (2.8.15). So, our ideas of primary qualities do resemble what can be said to exist in bodies, i.e. things as they are in the world, but ideas of secondary qualities do not resemble anything that can properly be understood as existing in those bodies. Of course, this does not mean that our ideas of secondary qualities do n ot resemble the sensations we have when our sensory organs are stimulated by the primary qualities of said bodies, and Lo cke says nothing to suggest otherwise. It means that the simple ideas we get from the Objects themselves, but Powers to produce various Sensations in us by thei primary Qualities (2.8.10). The ideas of colors or tastes or odors for example, cannot resemble produced by those powers. So, when someone we are to think it also holds for the Essay kind of resemblanc e as indirect re sembla nce. 13 Ideas o f substance, by definition, include the idea of substance and if there is any idea Locke cannot think resembles it must be this one. 14 This assumes that these are ideas of sensation and not of reflection. The commentary on Locke seems to only concern itself with the former. This is probably because it is that set of ideas that is seen to be problematic. It is also for this reason that I do not discuss modes at any length.
94 Of course, as E.J. Lowe points out, idea could not be red in the way that a physical object is red [because] an idea has no microstructural parts with primary qualities; nor does an idea have a po wer to produce sensations in us rather, it is a sensation in us. 15 However, this should not be a problem on the present account because the imagistic idea of a secondary quality, such as it is, resembles not the secondary quality but the effect of the seconda ry quality upon us. So, a sensation (the idea of red for instance) resembles the sensation produced in us (the perception of redness), not the red of the physical object itself. It does represent the redness of the object, but that redness is not to be eq uated with the red we perceive when we look at the object. The redness of a rose is a power to produce the idea or sensation, of red in us, while the redness of the idea just is that sensation. Ideas of secondary qualities are all covered by this kind of thinking in the Essay What about a complex imagistic idea that combines ideas of primary and secondary qualities (a mixed mode), say that of a colored ball? It seems that Locke must say that such an idea both resembles and does not resemble the qualities of the ball. The idea resembles the primary qualities and does not resemble the secondary. In itself this seems to pose no difficulties; we are all acquainted with situations in which two things resemble each other in some respects but not in others (as i n siblings bearing striking, but not perfect, similarities, or in automobiles of the same make and model but that have different paint jobs). So, if we can make sense of our ideas of primary qualities (and their modes) resembling those qualities, it might make sense to say that (A) or (B) holds of our imagistic ideas (but only those ideas). 15 E.J. Lowe, Locke on Human Understanding (London: Routledge, 1995), 51.
95 is something of a sticking point. o difficulty seeing how ideas of secondary qualities might not resem but things outside the mind. 16 This puzzle goes back at le 17 Primary qualities of bodies are things like the extension, figure, number and motion of those bodies (2.8.12) and each of these things seem to require something situated in spa ce in a way that our ideas are not. The issue becomes still worse when we add the idea of solidity which Locke says is produced by those primary qualities (2.8.9) distinction between our ideas of primary and secondary qualities was groundless. 18 If ideas of primary qualities do not resemble those qualities after all, it follows that there is no distinction of the sort Locke identifies. 19 There are ways to try to res Although our ideas do not perfectly resemble, o ne might try to say that they resemble as a result of a recognizable pattern resemblance. An idea would not have to share all of the pertinent properties of what it represents, but only enough to establish enough of a resemblance that one 16 The Philosophical Review vol. 108, no. 4 (Oct., 1999), 463. 17 George Berkeley, Principles of Human Knowledge edited by Howard Robinson (Oxford: Oxford University Press, 1996), Part I.9, 27. 18 issue is only a step in that proj e ct. 19 Inter estingly, Leibniz goes the other way and says that ideas of secondary qualities also resemble in a manner sufficiently analogous to those of primary qualities, but thereby seems to have agreed with Berkeley that the distinction Locke makes is inappropriate See New Essays on Human Understanding translated and edited by Peter Remnant and Jonathan Bennett (Cambridge: Cambridge University Press, 1996), II.viii.13 15, 131 132.
96 can recognize what is intended to be represented. Jacovides uses the example of a plastic model of the Statue of Liberty, which he says resembles the real thing in the latter sense, which Jacovides calls a literal sense, though it does not in the former, perfect resemblance, sense 20 One might also think of a drawing or painting of someone, which certainly does not perfectly that one can recognize the intended subject without difficulty. However, as suggested above, the problem with this is that these examples require imagistic ideas, and such images seem to require extension in space, and Berkeley seems to rightly think that this is odd, to put it mildly. Certainly our ideas are not extended in this way. We might say that the ideas themselves are not extended, but the image s we conjure up wh en thinking of those ideas seem to us to be extended. However those images seem to be nothing more than the same kind of sense experiences we between ideas of primary and secondary qualities seems to collapse. Jacovides attempts t discussions at 2 8 .15 and 2 30 represent by exemplifying, which is to say that they represent in virtue of sharing or nearly sharing s 21 Emphasizing the entirely appropriate point that Locke was reacting to the historical background of Aristotle, Gassendi and Descartes, while observing that, of course, Locke shied away from considering whether or n ot the images [Locke] straightforwardly believes that ideas of primary qualities resembl 20 Jacovides, 468. 21 Ibid., 469.
97 22 For t hat to be true Locke must think it is 23 Locke believes that ment 24 Unfortunately, Jacovides really gives us no reason to think that this is a remotely plausible position theoretically Certainly, Locke may have thought it makes sense to think of mental images as extended and shaped, an d therefore that imagistic ideas have those properties, but it is very hard to see why we should think this was reasonable position for him to hold The principle of charity suggests that, given alternatives, we should opt for the reading that will saddle Locke with the more reasonable position. Again, we might think of these imagistic ideas as reflective sense perceptions, but then the distinction between ideas of primary and secondary qualities loses its force. Lowe offers another way of responding to t he objection at hand. He suggests that we physical events, su 25 So, days of the month resemble the sequence of natural numbers in the sense Lowe has in mind, as do the marks engraved on a compact disc the sound that will be produced when the disc is played. Lowe sums up this think structural isomorphisms obtain between our ideas of the primary 22 Ibid., 474. 23 Ibid. 24 Ibid., 475. 25 Lowe, 57.
98 26 This seems closely related to a suggestion offered by Jan Erik Jones: [I]f we adopt the idea of variables [ as a way to underst and how ideas are used to stand for things ] as part of how representation occurs, it seems possible to content of an idea and using the objects, or relata, as variables. The re semblance consists in the similar relationship, not the relata. The same can resemblance. 27 If we include specifications of relationships in our ideas we might say that those ideas r esemble included in the idea of brother and, understanding sibling as a relation between offspring of a single parent (or pair of parents) on e can say the idea resembles any such relation out in the world. This does seem to allow for a sense of resemblance that might be operating in the representation done by some non imagistic ideas, like brother or democracy 28 As will become clear, I do think there is an important sense in which these relationships are included in the content of an idea. However, while I suspect that one could develop this into a fairly robust sense of how this kind of resemblance could be understood to be involved in our ideas generally, I see no clear evidence that Locke was thinking along these lines or that this is what Berkeley or Hume thought. If this is what is meant by resemblance resemblance in perhaps its most abstrac t sense then it may very well be involved in the vast majority of our ideas, but, as I will argue in 26 Ib id. 27 Jan Erik Jones, personal communication, and in comments in response to my presentation of a n early and much abbreviated version of this chapter at the meeting of the American Philosophical Association, Eastern Division, December 28, 2013. 28 It seems to me that annoyance at least as far as its meaning as an emotion, might very well be an imagistic idea, broadly understood.
99 Â§3, when Locke says that x represents y he still does not mean to say that in so doing x resembles y. With the move(s) suggested by Lowe and Jones, to a very general understanding of resemblance, there is also a worry that anything can be said to resemble anything else. As Hume suggests in his Treatise anything that can be compared at all will have some d egree of individuals, it leads not the mind directly to any one of them; but by presenting at once too great a choice, does thereby prevent the imagination from fixing on 29 If the resemblance is supposed to set what it is that an idea represents at least in part by directing our minds to that thing when we think of the idea, it is unclear resemblance as such will do that reliably. The resemblance that is supposed to be doing the work in making an idea represent some particular thing or delimited set of things will fail to sufficiently specify if the resemblance is stretched too far. While I do not think that Jones at least, has anything so extreme in mind there is still a need to be careful not to abstract too much (so to speak) when establishing a resemblance, or run the risk of losing the very point of that resemblance. There is, I think, a remaining problem. None of the responses offered by Jacovides, Lowe or Jones seems to deal with the issue of the ideas of solidity bulk which Locke explicitly (2.8.23). 30 It seems to be quite clear that even an imagistic idea, however that is unders tood, can hardly resemble solidity or bulk in bodies There seem to be two ways to resolve this difficulty. First, if we consider what imagistic ideas of these two qualities might come to, perhaps one might think we have in mind the 29 Hume, I.i.5.3, 15. 30 The full list is Bulk Figure Number Situation and Motion or Rest
100 resistance we experienc e when we push against a solid object or when we try to move one that has bulk, but this suggests that these are more like ideas of secondary qualities than primary. Second, we might have the non imagistic (and therefore not resembling) idea of is solid or has bulk ( is bulky ) associated with our ideas of solid or bulky things. If there is a way to understand an isomorphic structural resemblance between our ideas of solidity and bulk and those qualities in bodies, I have no idea how that is supposed to go. C ertainly, there may be a logico mathematical way of characterizing a resemblance between these qualities and our ideas of them of the sort that Lowe seems to have in mind, but this hardly seems to be the kind of straightforward resemblance that Locke might have intended. Th e foregoing considerations suggest, I th ink, that Berkeley 31 was right; the distinction Locke makes between our ideas of primary qualities and our ideas of secondary ones is unfounded. It seems that the best thing to say is that, like ou r ideas of secondary qualities, to the extent that they do resemble the actual qualities in objects, our ideas of primary qualities resemble not the qualities themselves but the sensations caused in us by our interactions with those bodies. Perhaps the mos t that can be said is that ideas of primary qualities resemble more closely what we take to actually be in the bodies than our ideas of secondary qualities. However, none of these imagistic ideas perfectly or directly resemble the things of which they are supposed to be ideas. At the very least, we have good reason to be skeptical of the distinction Locke makes between ideas of primary and secondary qualities in regard to resemblance. At the beginning of this section, I identified two ways of understanding a resemblance theory of representation; either (A) x represents y (just) because x resembles y or (B) for x to represent y it is necessary that x resembles y Due to what Locke explicitly says about secondary 31 As well as Leibniz.
101 qualities, it is clear that he thinks neither (A) nor (B) can apply to all ideas. I have argued that such resemblance must be constrained to the relatively small set of ideas that are mental images, Our im agistic ideas of both primary and secondary qualities can only really resemble in the indirect sense that secondary qualities can be said to resemble. So, the best a resemblance theory of representation can get us is a partial and indirect understanding of Lockean representation. We cannot, therefore, rely upon resemblance to explain Lockean representation for non imagistic simple ideas and have no clear understanding of how non simple, complex ideas, particularly of mixed modes, represent In the following sections I will argue that a resemblance based representation is just not what Locke has in mind for either sort of ideas 32 4. 2 Causal Theories of Representation Contrary to the former reading, a causal theory of representation holds, basically, that an 33 So, my idea of a particular apple say, represents that apple ju st in case my idea was caused by that apple. In a parallel of (A) and (B) from the previous section, either (C) being so caused is what makes my (representational) ideas represent, or (D) being so caused is a necessary (but possibly not sufficient) conditi on of their representing. In the following, I argue for a causal reading of Lockean representation for simple ideas. 32 An alternative way to understand resemblance might be less problematic. If Locke is read as thinking of x resembling y in the sense that it accurately represents y one might tie a resemblance based theory of representation to what Locke says about the adequacy of our ideas. Unfortunately, I cannot pursue this line of thought further at this time. 33 British Jo urnal for the History of Philosophy (20:6) 2012, 1078.
102 The first thing to be said here is that this cannot be a clearly simple or direct sort of causation. If we are going to get any ideas beyo nd the particular ideas of sense experience, for instance, the various mental faculties Locke is at pains to elaborate must come into play in a sometimes complicated causal system. As a relatively simple example, according to Locke we do not acquire the ge neral idea of white simply by being exposed to many instances of white things they do not just cause us to have the idea of white but we must also apply abstraction to our ideas of those white things. Much more needs to be said to trace out and explicate h ow the causes are supposed to work in the Lockean picture. Perhaps it is sufficient at this point to suggest that the causal role is such that if we did not have the experience of x (in some, appropriate sense of experience) we would not have the idea that represents x However, Locke insists we can combine the ideas we are caused to have by experience into more or less novel new ideas (we cannot have an experience of such an x because there is no such thing), so it seems that still more needs to be said to explain such a result on a causal theory. The second thing to be said is that, as Ott observes, much in the Essay seems to suggest this causal view without endorsing it outright. 34 For instance, in Book 2, Locke consistently says 35 My idea of the yellow of a ball is true (in this sense) if it matches up with the yellow that I perceived are adequate Because be ing 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: And we are sure they certainly seems to think that our simple ideas 34 Ibid., 1079. 35 I return to the truth and reality (and adequacy) of ideas in C hapter 6
103 are what we are caused to have by our experience and our perceptual (and mental) capacities. Locke goes back to this point during his discussion of the so calle d veil of ideas worry in Book 4 using it to shor representation as such, there can be little doubt that the kind of causality that is needed for a simple ideas, which for Locke must be the fundamental level. To see how this works for Locke it is important to again emphasize the passi ve way in which we receive our simple ideas. As discussed in Chapter 3, 36 simple ideas are just what we are given by our experience of things (including our own mental contents and processes), though we may need to apply relati vely more attention to our experience to get at some of those things. The p assive and unavoidable way our simple ideas are tied to our experience of the qualities of things makes it the case that these ideas automatically represent those qualities. Our min ds cannot help but use red to represent instances of redness in the world, or in motion to represent things that are moving. On this level, that of simple ideas, the representation done by ideas is caused by our experience of the world. So, effectively, ou r experience and our nature make it the case that our simple ideas represent (at least) their causes. The primary representation done by simple ideas is, essentially, forced upon us. This may be well and good for simple ideas, but we have ideas of people and things, not to mention complex abstract and general ideas ideas of modes and substances and surely those ideas also represent those people and things. It is hard to see how representation beyond the level of simple ideas can get started if we cannot at least consider something in the world as a 36 I return to this subject in Chapter 6 as well.
104 discreet entity, as some thing ; we need to have the idea of thing or entity to begin the process of representation above the level of simple ideas. The worry is, h ow can we use an idea to stand for a thing, if we do not already have some way of representing the thing? This concern is closely related to a problem Lowe raises: how can we recognize something as falling under a given idea at all without falling into a regress, and how are we able to distinguish someth ing as an animal without having an idea of the kind of animal it is, without first having the idea of animal ? 37 To answer that question we must first answer the more fundamental question, how is it possible for this to get started at all without the pre exi sting, or innate, idea of thing external sensible Objects ; or about the internal Operations of our Minds granted that we are able to distinguish one thing from another. However, what he says about how a child gets his or her first ideas seems to emphasize the passivity of the chil d: He that attentively considers the state of a Child at his first coming into the World, will have little reason to think him stored with plenty of Ideas that are to with th em: And though the Ideas of obvious and familiar qualities, imprint themselves, before the Memory begins to keep a Register of Time and Order, yet few Men that cannot recolle But all that are born into the World being surrounded with Bodies, that perpetually and diversely affect them, variety of Ideas whether care be taken about it or no, are imprinted on the Minds of Childr en. Light and Colours are busie at hand every where, when the Eye is but open; Sounds and some tangible Qualities fail not to solicite their proper Senses, and force an entrance to the Mind (2.1.6). e Memory begins to keep a Register of Time 37 Lowe, 108.
105 on our minds. This does not explain how we are able to distinguish one thing from another in the way that Lowe seems to have in mind, but it does clearly indicate that Locke thought our ideational systems get started passively, by stimulus forcing itself upon our young minds. Perhaps it also makes sense to think that the basic level of representation is just something th at is done by our minds in some sense. The mind has the proto idea or category of mere something forced upon it, so to speak, by our experience of the things in the world. Using this most basic representation, as a thing of some sort, our representational system can get going. But is this adequately supported by the text of the Essay ? I believe this can be answered affirmatively, but to do so we must remember the following two points: we receive our simple ideas passively from our experience (though the a ction of experiencing is not passive) and those ideas come together with others of their kind. The line of thought is as follows. 38 1) As I will discuss further in Chapter 6, Locke says that our simple ideas are all real (i.e. they are ndation in Nature; such as have a Conformity with the real Being, and Existence of T hings, or with their Archetypes (2.30.1)) because they consistently produce rea lity lying in that steady correspondence, they have with the distinct Constitution of real Beings (2.30.2). Because we cannot simply create our simple ideas, they must be provided to us by things external to the mind, and Locke seems to take for granted t he connection between that understand how it is that representation can get started, by some kind of idea of thing perhaps, 38 Several of the following points are suggested by Epistemic Rol Pacific Philosophical Quarterly 85 (2004), 301 321.
1 06 we need to have some non innatist way of deriving this from our experience of the world, and project. 2) In discussing our knowledge of external things, Locke has the following to say: actual receiving of Ideas from without, that gives us notice of the Existence of other Things, and makes us know, that something doth exist at that time without us, which causes that Idea in us, though perhaps we neither know nor consider how it does it: F or it takes not from the certainty of our Senses, and the Ideas we receive from them, that we know not the manner wherein they are produced (4.11.2). So, according to Locke, it is the fact that we are receiving our ideas confirms that there are things ext ernal to us, and it does not actually matter how it is that those external things make us have those ideas just so long as they do. Because the ideas we are so receiving are all simple ideas on ernal things comes along with those particular ideas, which we have seen are gotten passively by the mind. 3) Locke also says: The Mind being, as I have declared, furnished with a great number of the simple Ideas conveyed in by the Senses as they are fo und in exteriour things, or by Reflection on its own Operations, takes notice also, that a certain number of these simple Ideas go constant together (2.23.1). Not only are the simple ideas, effectively, forced upon us, so long as we have the requisite exp eriences, but some of them come constantly joined together, and (from the passage in (2)) Existence of other Things 4) It seems not much of stretch here to connect these points and say that Locke might say, we begin with our simple ideas, which are thrust upon us by our experience of the world, and, because those ideas frequently come bundled together (as it were) in re gular, or even constant ways, these ideas come along with the idea of particular things An application of the mental
107 capacity of abstraction gives us the general idea of thing 39 So, there is a causal, non innatist story of how we come to have the basic id ea (or category) of thing with which we can begin to use representation. Perhaps we need a bit of attention to the fact that simple ideas come bundled in the way they do, but that is consistent with what we already have on the table. So, if it makes sense to think our simple ideas are automatically representative (at least of the things that cause us to have those ideas), this can be understood to include the basic, and simple idea, of thing or entity Alternatively, one might simply point out that thing or entity is a simple, though abstract, idea, and is therefore just given to us by experience like all the other simple ideas we have. Once we have the idea of thing or entity plus all the other simple ideas thrust upon us by experience, along with the re presentational relationships which we cannot help but set up between them and their causes, 40 we have the foundation upon which the rest of does seem to be the best re ading of what Locke actually says in the Essay There is much more to be said here about how best to understand these causal connections and how to understand the fundamental nature of Lockean ideas. To take just one worry, how should one understand Ayers representation with his blank effect reading of Lockean ideas? 41 Unf ortunately, I cannot now adequately pursue such questions However, because it is critical to the point of this chapter, I need to address the is sue of how the kind of causality is supposed to work for general ideas and function in an analysis of complex ideas. That is, on a causal theory of representation, how is it 39 I discuss abstraction in the next chapter. 40 I make this clearer in the following section. 41 Michael R. Ayers, Locke, Vol. I: Epistemology (London: Routledge, 1991).
108 that mammal represents ideas like cats dogs monkeys zebras and humans let alo ne the cats, dogs, etc. in the world? Here, I think, the causal theory of representation is not terribly problematic, at least on its own merits. It could be said that if causation (appropriately understood, as above) is what constitutes representation, and if we allow for the incorporation of our mental capacities for things like abstraction, we have a fairly standard Lockean understanding of how more abstract ideas can b e thought to represent less abstract ones, etc. The former ideas represent the latter just in case having the less abstract ideas is causally responsible for our having the more abstract ones. We would not have the idea of horse for instance, if we did no t have the ideas of individual horses. As I will suggest in Chapter 5, there might be a variety of causal paths by which we could arrive at many of our general ideas, but they would be nevertheless be causal relationships. However, as I will now argue, non e of it is quite what Locke has in mind by representation for non simple ideas. 4. 3 Having argued for a causal understanding of the representation done by our simple ideas, including that of thing it is now necessary to develop an understanding of Lockean representation that applies more generally; that is, to complex ideas. One option is to just say that all other representation is just derivative upon the representation done by simple ideas. An idea of a particular per son which is a mixed mode, just represents in so far as, and because, the simple ideas of which it is composed are caused by the simple qualities of that particular person. It is possible that more traditional proponents of causal theories of representati on have this in mind. In the following, though, I argue that this is not the best, most straight forward, reading of Locke on this matter. To support this position, I begin by shifting the focus to what Locke says
109 about the representation done by general i deas, and then generalize, so to speak. Moreover, I start with the representation, or, more accurately, signification, done by words. an Idea as is but a part of any Words, what Idea b eing only the explaining of one Word, by several others, so that the meaning, or Idea it stands stand for ideas, and definitions make it clear which ideas are stood for. Continuing on, Locke says in th e following paragraph: Words are general, as has been said, when used, for Signs of general Ideas ; and so are applicable indifferently to many particular Things; and Ideas are general, when they are set up, as the Representatives of many particular Things: but universality belongs not to things themselves, which are all of them particular in their Existence, even to those Words, and Ideas which in their signification, are general. When therefore we quit Particulars, the Generals that rest, are only Creatur es of our own making, their general Nature being nothing but the Capacity they are put into by the Understanding, of signifying or representing many particulars. For the signification they have, is nothing but a relation, that by the mind of Man is added t o them (3.3.11). or word and note that both are explicitly included here represents in any general sense is only a relation that is supplied by the mind. To be specific, the generality is made by our minds, but that generality consists only in the fact that our ideas and words are used to stand for many particulars. The general ity of our ideas comes directly and exclusively from the fact that we use them to stand for particulars. The representation (or signification) that an idea (or word) does is a relationship between the idea (or word) and what is represented, and that relati onship is something our minds establish, at least as far as general terms and ideas go. Locke goes on to repeatedly emphasize, over the course of the following paragraphs, that the association between
110 the essences of things and our ideas and then our words make our general ideas represent by using them to stand for things in the world (which must all be particulars), which we do by establishing a relationship between the ideas and the particulars. For instance, Locke says: I would not here be thought to forget, much less deny, that Nature in the Production of Things, makes several of them alike: there is nothing more obvious, especially in the Races of Animals, and all Things pr opagated by Seed. But yet, I think, we may say, the sorting of them under Names, is the Workmanship of the Understanding, taking occasion from the similitude it observes amongst them, to make abstract general Ideas and set them up in the mind, with Names annexed to them, as Patterns, or Forms, (for in that sence the word Form has a very proper signification,) to which as particular Things existing are found to agree, so they come to be of that Species, have that Denomination, or are put into that Classis ( 3.3.13). The sorting of things, especially types of animals and plants, is a workmanship of the understanding. The sorting into species being discussed by Locke is another aspect of establishing what shall be represented by a set of general term or idea, in this case species, or classis Stimulated by the similarity we observe, the understanding makes general ideas, which are patterns or forms, and it is to this pattern or form that we say things agree (or do not agree). The mind actively makes our general find in the world. 42 42 I n the above passage, note that Locke is explicitly talking about words and ideas. A couple of other passages that Adequate which perfectly represent 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); is usual for Men to make the Names of Substances, stand for Things, as supposed to have certai n real Essences, whereby they are of this or that Species: And Names standing for nothing but the Ideas they must consequently referr their Ideas to such real E ssences, as to their Archetypes (2.31.6 ).
111 This strongly suggests that general ideas represent particulars because we have made them do so, by using them to stand for things out in the world. This all echoes what Lo cke says in introducing abstraction in Book 2: Thus the Colour being observed to day in Chalk or Snow, which the Mind yesterday received from Milk, it considers that Appearance alone, makes it a representative of all of that kind; and having given it the name Whiteness it and thus Universals, whether Ideas or Terms, are made (2.11.9). I suggest that we take what Locke is saying seriously and apply it more broadly to Lockean id eas. Our non simple ideas do not represent because they resemble, but because we use them to represent to stand for other things, even though some of our ideas might happen to resemble what they represent. That resemblance, when it happens, is incidental to the handed people, and for all right handed people; the standing for comes from the way I use those letters in my discourse (about the handedness of people, say). Similarly, resemblance theorists like Jacovides (and by extension, Berkeley and Hume) are looking in the wrong place to identify where or how ideas represent. Primarily, ideas do not represent because they resemble what they stand for, but be cause we use them to stand for those us. This is not to say that setting up this relationship is always done voluntarily. There is an on that strongly (and rightly, I think) suggests, that some generalizations, and the representational relationships set up thereby, are quite natural and automatic like that he describes for coming up with the idea of white As I have argued, this is the c ase for simple ideas. But that does not mean that the represen tation is not something that is
112 introduced by us, by our understanding as it processes the information given to it. Recall that in that passage from 2.11, in which there clearly is some resembla nce involved, t he whiteness of milk, chalk or snow is isolated, or considered alone, and righ t there we have a Lockean idea a simple, imagistic one. Then it i s made a representative and becomes a universal. Yes, there i s an idea that resembles the instance s it represents, but the representation is something introduced by us. We do not decide to use the idea in that way, but our minds establish that relationship nevertheless. 43 Moreover, although the discussion so far has focused on general ideas (and Locke really focuses on general ideas for the vast majority of the Essay ), the reading I am advancing also applies to non simple ideas of particulars. While I am caused to have ideas of particular things, like the idea of the cat sitting in the window in the kit chen, the representational nature of that idea is something that is added by the mind. I am caused by a state of affairs in the world to have the idea of the cat, but the fact that my internal idea of that state of affairs represents the state of affairs i for in that way. The idea I have of the cat represents the actual animal just because my mind has established that relationship, and the idea need not resemble the cat in any real sense, though I m ight, and do, have various imagistic ideas associated with that idea. Is what I have just said about ideas of particulars in accord with the Essay ? What Locke says about how we get these ideas, the immediate objects of experience, is not particularly hel pful. As I have discussed in Chapter 3, in 2.1.1 5, Locke explains his familiar story of how we get our fundamental ideas via sensation and reflection, and there can be little question that Locke 43 Another way to understand the way representation is up to us is to say that without the involvement of a mind there would be no representation.
113 takes the mind to be largely passive in acquiring these idea s. However, it seems worthwhile to point out that none of the ideas he mentions in these paragraphs 44 are ideas of particular items; they are all abstract and/or general ideas. This seems to imply the active, if often automatic, operation of the faculties L ocke will go on to discuss (as in 2.9 11), particularly abstraction. In fact, as I have already pointed out, 45 Locke stresses the importance of our being attentive: For, though he that contemplates the Operations of his Mind, cannot but have plain and clea r Ideas of them; yet unless he turn his Thoughts that way and considers them attentively he will no more have clear and distinct Ideas of all the Operations of his Mind and all that may be observed therein, than he will have all the particular Ideas of a ny Landscape, or of the Parts and Motions of a Clock, who will not turn his Eyes to it, and with attention heed all the Parts of it (2.1.7). So, while it seems that we automatically get our ideas of particulars from experience, some attentive effort is re causal operation includes an application of our own faculties. This all suggests that we are still working with the causal account I began developing in the previous section. Our com plex ideas of particulars represent what causes us to have those ideas, but our minds are actively involved in that causal story, so that they represent what they do because our mind has set them up to do so (naturally and unavoidably). So, it is true that our ideas of particulars represent those particulars because we use them to do so; we just cannot really help doing that. The foregoing suggests that Locke is best understood as having something of a proto use theory of representation, or reference. 46 Whe ther or not, and to what extent, this is the case is 44 Whiteness, Hardness, Sweetness, Thinking, Motion, Man, Elephant, Army, Drunkeness (2.1.1); Yellow, White, Heat, Cold, Soft, Hard, Bi tter, Sweet (2.1.3); Perception, Thinking, Doubting, Believing, Reasoning, Knowing, Willing 2.1.4 ). 45 Again, in Chapter 3. 46 This should not be confused with a use theory of meaning, of course.
114 something I leave to a future effort. For now, it seems enough to point out that the connection is fairly clear. Without the clearly problematic and anachronistic assumption that Locke had a much later c onception of representation, it seems that he did have the kind of view I am advocating in mind, at least as a background view. Having established that my reading is supported by the text of the Essay and seems to have been what Locke generally had in min d, I must now address the worry that this reading will not adequately deal with the problems raised earlier for a resemblance understanding of representation. There is no requirement that ideas be imagistic, so we can set aside worries that the current re ading of representation will not apply to all our ideas, whether imagistic or not. Baxter 47 However, we are now in a position to say that Locke could just mean that the same abstract idea is used to represent the various particulars of a given sort. tion. We also have no problem with non imagistic images representing imagistic ones (or vice versa, for that matter). Moreover, because the idea of triangle need not be a triangle or triangular in any sense (though perhaps for some people it is) for it t o represent triangles, the force of Berkeleyan concerns is diminished. Representation works the same way for all our non simple ideas and the things they are ideas of. In each case, the thing that represents does so not because of a more or less nebulous r esemblance but because we use that thing in that way whether or not we have any choice in the 47 Baxter, 313.
115 matter Our understanding makes the relationship such that the representing thing stands for what is represented. An idea represents just by virtue of the fact t hat the mind uses it to stand in for considered earlier in this section strongly suggests this generalized understanding of the representational relationship The representation done by simple ideas, on the other hand, is best understood on the causal understanding covered in Â§2. In no case is it necessary for an idea to resemble what it represents for the idea to so represent. This is not to say that resembl ance does not play a role. As I shall discuss in the next section, resemblance is involved in the derivation of our ideas. This is quite far from saying that resemblance is required for all representation, which I have argued is simply not the case on Lock resemblance between things. This sort of resemblance clearly does seem necessary. We could hardly have general ideas without our noticing similarities between some of the particulars we whiteness there, in 2.11.9, explicitly relies upon resemblance, but it is the resemblance between different instances of white things. The representational role of ideas, in relation to things, as such does not rely upon that resemblance. On the other hand, my reading of Locke is clearly a modified, or extended, causal theory. Recall that I said there was no plausible way to understand Lockean representatio various mental faculties into the process by which our ideas (and words) come to represent. The ability to make something stand for something else in our thoughts is simply another mental faculty. So, in what seems to me to be a more indirect sense than other causal theorists seem to
116 have in mind, our ideas do represent due to the causal process by which we come to have those representing ideas While there clearly is a causal process that not only results in our ideas but is also responsible for the fact that those ideas represent what they do, that process should not be understood to be just a simple matter of our ideas representing their caus es because they were so caused. Instead, for our non simple ideas, the element in that process that causes the representation is introduced by our minds in establishing the representational relationship, even when the mind has no choice in the matter. I t ideas, and particularly our general ideas, without in some way thoroughly relying upon resemblance. I have discussed this in Chapter 3, but in the next section I will attemp t to make this linkage clearer in the current context. 4. 4 Simple Combinations and t he Limits of Representation Combining the foregoing theory of representation and the simple combination reading discussed in Chapter 3 makes it possible for us to systematically use representatives to stand for particulars (and relative particulars) in our thoughts and discourse. Eac h of our complex ideas (at least) just is a simple combination, which enables us to keep track of what things we intend to be stood for by that idea. In principle, the simple combination of man enables us to understand both 48 because the idea as simple combination includes living creature and mammal while excluding other mammals. These simple combinations, or, rather, their component ideas, ult imately reduce, by representational links, to simple ideas of sense and 48 Of course, Locke has serious misgivings about whether or not we can know the truth of propositions involving substance kinds, like man
117 reflection and it is on that most basic level that we rely upon resemblance to explain the content of some of those ideas the imagistic simple ones not their representative nature. It is the simple combinations of our terms and ideas that keep them properly connected to the world around us (and our other ideas and internal experience). If our ideas and the relations established between them map onto the world properly if they stand in t he right relations all around we simple abstract idea if the particular ha s all the things included in the simple combination of the idea and none of the things excluded by it. This is not an explanation of how our ideas represent things as such, but how it is that we understand the extension of our various complex ideas. This a lso applies to ideas of particulars, which must implicitly include something that ties the idea to just one instance or existent. If a more thoroughly worked out example is desired, take the general idea of chair On my reading of Locke, it is not necessa ry that we have an image of some chair or other (perhaps then using that image, a la Berkeley and Hume, to represent any chair in our internal or external discourse). Rather what is necessary is that we have a simple combination of chair which just is to etc.) and anything that seats just one but lacks a support for the back (e.g. stools). It includes, not be an image either, and which involves the general ideas of support and back By back here is intended, of course, th which in turn leads us to torso which represents torsos generally, and which we arrived at by abstracting from our experience of torsos
118 in the world. 49 simple ideas of sensation and reflection. Simple ideas are gotten from our experience; according to Locke there just is no other source for such ideas. 50 The foregoing entails that we are able to store a staggering number of inter related representations and ideas, which should come as little surprise to anyone who has dealt with a great credit that he appreciated the power of this faculty of the human mind to keep track of the myriad representations we do in fact use, and attempted to show how we can make sense of it all. It should be noted that these complicated simple combinations are also the workmanship of the understanding. Locke does not think that we are simply provided with such mental content by the world, but that this is something the mind does as a result of its interactions with and experience of the world. They are of human concernment in that this is how we, as the kinds of creatures we are, w ith the minds we happen to have, organize our ideas and thinking about the world, and if we do not properly attend to the meanings of the ideas we use we will be terribly disadvantaged. Of course, there is an arbitrary element to our choices of representa tives. However, there seems to be a limit to how far we can push this freedom. If some of our ideas are images in some sense, then the causal origin of those ideas would seem to determine both the idea and the class of things the idea stands for; because o f the representation that is being done there will be (perhaps indirect) resemblance, but the resemblance is not what is responsible for the representation. As I have also argued, we are unable to not use our simple ideas to represent the 49 We might, on the other hand, construct our mixed mode idea torso by combining ideas that do not normally come togethe r in our normal experience of the world. 50 Of course, most of this paragraph is a straight forward application of the material covered in Chapter 3.
119 causes of those i deas. Also, it would seem to be a very serious problem if we were to go around using man to stand for human beings, chimpanzees, cats and tables. The simple combination reading of Locke that I laid out in Chapter 3 seems open to the possibility of wildly d isjunctive combinations. As a result, care must be taken to delimit the meanings of our ideas, and Locke goes to great lengths to emphasize the importance of this. We might also worry about how we make sense of very general abstract claims like murder representing? Similarly, we might worry about mathematical claims. These concerns are about the proper meanings of our ideas though, and not about representation as such. As for how we can make sense of them, according to L ocke, I will return to this in Chapter 7. So, to review, the idea kitten which represents kittens in the same way that freedom represents, or stands for, particular instances of freedom. That relation of representation is one that is established by our mi nds in our interactions with the world and our ideas, and does not depend upon resemblance, though resemblance may sometimes be involved, and is, in fact, important if we are to have any real knowledge of the world. General ideas represent particular ideas in just the same sense; they are ideas used to stand for anything that matches up with the simple combinations of those ideas (which, again, just are those ideas). Simple ideas like say, red represent red things because our experience causes us to have t hose ideas which inherently represent their causes (red things). functional understanding of representation is that they were not specifically or primarily worried about distinction Locke makes between the ideas of primary and secondary qualities of bodies. Shifting
120 the perspective and focusing on how in the world our more abstract idea s are supposed to represent less abstract things seems to have been particularly helpful. 4. 5 because it effectively constitutes an objection to t he reading I have presented in the previous 51 Moreover, he gives some reasons for thinking that my own reading cannot be right. In this section I will reverse the order of development i objection does not offer any good reason to reject the position I have offered in the previous three sections. resemblance; since some ideas do not res emble, those ideas do not represent; rather those ideas signify 52 but his position is more nuanced than this identification would suggest. Along with other, related passages, Ott points to the beginning of 2.30, where Locke says: Our simple Ideas are all real all agree to the reality of things. Not that they are all of them the Images, or Representations of what does exist, the contrary whereof, in all but the primary Qualities of Bodies, hath been already shewed (2.30.2). 51 Ott, 1078. 52 Ibid., 1081.
121 Ott points out that these sentences seem to blatantly equate representation and being an image of what is represented. Based on this passage, Locke clearly seems to think that some of our ideas are not representations. As Ott says: This opens up a space for us to take ideas of primary qualities alone as representation. In such a case, it will be resemblance that grounds the representation relation. Ideas that do not resemble th eir objects are not representations, though they might still be signs. 53 should be understood to represent their objects. Other ideas, like ideas of secondary qualities, which do not resemble their objects, do not so represent. Those ideas might be signs. complicated; I can only very briefly summarize his point here. The s emiotic tradition in which 54 Ott indicates 55 we do have some understanding of what was meant by signification. There is a distinction between signs that are reminiscent and signs that are indicative. A sign that is reminiscent will, clearly, remind someone of something that was already experienced, while an indicative one will point to something that commit him to a theory of representation that is grounded in resemblance; 56 that is, our reminiscent signs require a basis in resemblance, because they are ultimatel y dependent upon our ideas of primary qualities. Our other ideas, which are not grounded in ideas of primary qualities 53 Ibid. 1089. 54 Ibid. 55 Ibid. 56 Note that this is not the identity claim quote d earlier.
122 and so cannot be grounded in resemblance, merely signify indicatively. So, all our ideas and words are signs, but ideas of primary qualit ies also represent, while everything else merely signifies. 57 interpreting the passage from 2.30.2 (above) is too quick. It seems rather plausible that sense that I have been considering in this chapter, and that of a picture or photograph that can be called a on now, but it seems to be this sense in which Locke is using the term in this passage. Note that Locke also Ideas (3.3.11). It is implausible that Locke is saying we have no general ideas that are not images, because it would require a very thorough imagism. It is particularly hard to take since Ott thinks 2.30.2 actually means that our ideas of secondary qualities do not represent the things they are ideas of, sayi denies that simple ideas represent their objects, and 58 As explained, Ott goes on to make much of the distinction he thinks Locke is making betwe en simple ideas of primary qualities, which, based on his reading of this passage, Ott thinks do represent what they are ideas of, and simple ideas of secondary qualities, which Ott thinks (merely) signify what they are ideas of. However, if we grant that 57 Ibid., 1089 1090. It is noteworthy that this is not the distinction Locke seems to be making in the Essay in which signification is most commonly used in reference to the standing for that is done by words and representation tends to be used in reference to ideas. 58 Ibid. 1081.
123 all simple ideas are images of the things they are ideas of, properly understood, and this is sim ply an unproblematic statement of his (non imagistic) position on primary and secondary qualities and their ideas. 59 Ott either thinks that there is sufficient reason to adopt his position on its own merits, that his position is where he is driven by the f ailure of externalist readings of Locke, or both. It is not secon objection to my position his claim According to Ott a view is externalist idea represents (what it is an idea of where i 60 According to Ott, what a Lockean idea represents must be fixed by only intrinsic features of that idea, like what it makes sense to say it resembles. The view Ott is explicitly arguing against is any causal understanding of representation, and Ott consistently equates externalism with causal theories. externalist readings, such as those of Michael Ayers and Martha Bolton, take ideas to represent their causes. Others, such as that of Sally Ferguson, take ideas to represent what God intended to 61 While I think my view is quite a bit different from these readings (which again 59 60 Ibid., 1078. 61 Ibid., 1077.
124 seem to lack a general application to Lockean ideas 62 ), it is also fairly clear from what I have said that on my view what a non simple idea represents is fixed by the connection the mind makes between the idea and whatever it is the mind takes or sets the idea to represent, and this is 63 So, it is important to emphasize that my view is an externalist one that is at least somewhat different from the causal accounts Ott has in mind, though, as I have said, mine can be considered a type of causal account, and clearly is causal in its position on the representation done by simple ideas. This certainly seems to be an externalist position that is different from those Ott has in mind. Rat her than a position based on a distinction between resembling and non resembling ideas, incorporating both representation and signification, which is what Ott advances, but Locke seems not to have done, it seems my reading of the Essay which is based on t he distinction between simple and non simple ideas, fits better with what Locke is saying. 64 The primary objection is that externalist positions sin the 65 That is that Locke repeatedly says that al l our mental content is 62 I hope to return to this more fully in a future paper. 63 Ott function of its causal history. On the teleological variant, an idea is an idea of that object which God intended to cause it. On either sort of vi resembles its object is a further question. Resemblance does not play the primary or even a role in fixing what an idea represents y the reading that I am putting forth in this chapter does qualify as externalist on these standards. (a) The establishment of the representational relationship is either caused or is something that the mind does with, or to, the idea, after (in some sense ) we have the idea, so it must be extrinsic to the idea, and (b) resemblance does not at all fix what an idea represents. 64 It seems telling that Locke does explicitly make the distinction between simple and non simple ideas, while we have seen that the di stinction between resembling and non resembling ideas is problematic and does not have the same general applicability to all of our ideas. 65 Ibid., 1078
125 treatment of the possibility of confusing one idea with another, in 2.29, in which he observes that no two ideas can be different without the m ind being able to perceive the difference between them. Without denying our ability to discern all the details of our ideas, Locke says the confusion that sometimes happens is due to our misapplication of the names to dissimilar ideas and to our having ill formed and indistinct, undetermined (or under determined) ideas. 66 As Ott points out, causal history is not one of its intrinsic features: causal relations are paradigmatic cases of extrinsic 67 The point is that we cannot simply examine our ideas and observe and compare their extrinsic features, like their causal relations. Naturally, this objection is closely tied to recent worries about how ext ernalist understandings of meaning can underwrite our own knowledge ascriptions. For one thing, if the meaning of water is established by facts out in the world say, that it is composed of H 2 O rather than XYZ how can we ever be confident that what we say a bout water refers to the H 2 O stuff and not the XYZ counterpart? While a detailed discussion of this lively debate would take me much too far afield, it should be noted that quite a few contemporary thinkers think that externalism is compatible with the kin d of self knowledge Ott seems worried about in his objection. 68 It seems reasonable to say that such a contested point should not be taken to be a robust objection to an externalist reading of Locke. Unless, that is, such a reading conflicts directly with w hat Locke says or his commitments. 66 Locke, II.xxix.5 9, 364 366. 67 Ott, 1078. 68 For more on this subject, see Knowing Our Own Minds Cris pin Wright, Barry C. Smith, and Cynthia Macdonald, editors (Oxford: Clarendon Press, 1998); Externalism and Self Knowledge Peter Ludlow and Norah Martin, editors (Stanford: CSLI Publications, 1998); and, Internalism and Externalism in Semantics and Episte mology Sanford C. Goldberg (Oxford: Oxford University Press, 2007).
126 Much of what Locke needs from his theory of ideas is compatible with the externalist account I have given. This includes what Locke needs for his theory of knowledge. As Ott points out, 69 the perception of the connexion and agreement, or disagreement and repugnancy of any of our Ideas [discussed above] are available 70 This is particularly clear in the teleosemantic casual stories, in which an idea represents what the idea is supposed but also applies to th e other causal readings Ott has considered. However, the content of all our ideas is not only fully available but is distinct from the representative connection our minds make between the ideas and what they stand for. entational content of an idea is its extrinsic relations, and if ideas are individuated by their contents (what they are ideas of ), then there is no guarantee that, from the first person perspective, I am in any way entitled to say that two ideas are reall y 71 On the one hand, this seems to be a hotly contested issue in semantics and epistemology. On the other, this seems a bit confused. The contents of our ideas are fixed by our various mental operations, like perception and abstraction, on our experiences and other ideas, which result in what I have been calling the simple combinations of our ideas. For our complex ideas, we then compare those simple combination contents when we compare our ideas. What the ideas are of what they represe nt, is a secondary issue, not necessary for said comparison. The representation that an idea does is an extrinsic (to the idea) relation established by the mind and 69 Ibid., 1086. 70 Ibid 71 Ibid., 1085.
127 us is imagistic, but then it hardly seems like the causal story behind such ideas is an issue. The other objection Ott makes to externalist readings is closel y related to the transparency worry I have been considering, and Ott seems to take it to just be part of that issue. 72 support of this objection rather hard to follow. However, if I understand him correctly, the worry is that on a traditional causal/externalist reading, since we have no access to the causal connection between our ideas and what they represent (presuming t hey are external objects and not objects of thought, though it is not clear that Ott thinks this distinction is relevant), we cannot know that our abstract ideas are actually mulate an abstract idea of, say, white that represents all and only things that have the power to produce that idea in which 73 This does not seem to be a problem for Locke, though. What we abstract from are the ideas delivered to us by experience, so that we are formulating an abstract idea of white that actually represents all and only the ideas of things that have the power to produc e the idea of white in us. Abstraction is performed upon the content of our ideas; we consider the content of our various ideas of white things and abstract from them the idea white After this, we use the resulting idea to represent that color. There is n othing in this story that is not available to our introspective consideration. 72 Ibid., 1087. 73 Ibid., 1088.
128 of Locke in general is not well considered. Moreover, if there is a problem with ext ernalist readings, we seem to be driven to a resemblance theory of representation, and we have already seen how problematic such a reading would be. I take these considerations to support the conclusion that Ott has not given us good reason to reject my ow n externalist reading of the Essay 4. 6 Summation formulating some version of causal or teleo semantics; it should not be a surprise that, as perhaps the firs t person to offer such a view in the modern period, he is deeply confused as to its 74 I will not quite opt for this, but I hope that the foregoing has been enough to suggest that perhaps Ott is overstating the depth of the confusion that Lock e may have had. 75 My two part position that the representational relationship is either causa l (for simple ideas, whether general or not) or the workmanship of the understanding (for complex ideas) is neither inconsistent with the rest of the Essay nor as p roblematic as Ott would perhaps have us think. Moreover, as I have argued, my reading is better than other standard theories of representation, because it is less problematic than resemblance theories, covers all of our ideas, and is more in accord with wh at Locke says. Importantly, at least for my purposes, it is consonant with what I have said about Lockean ideas and non imagism. 74 Ott, 1088. 75 Though Locke does seem to have been off track regarding the distinction he makes between ideas of primary and secondary qualities.
129 CHAPTER 5 LOCKEAN ABSTRACTION AND GENERAL IDEAS t of fire and inspired a great deal of confusion. In this chapter I will be primarily focusing on the confusion, trying to clarify what I consider to be two of the most serious issues in dealing with this subject: whether Locke actually had a single theory or two, and whether or not Lockean abstraction requires the mental separation of ideas (e.g., separation of the idea of a particular color from that of extension) rather than partial consideration of a given complex (colored and extended) idea. I will also address some of the objections that have been raised against Lockean abstraction. 2 and 3 of the Essay fferent forms of 1 In 2 Locke describes how we move from having a complex idea, say, of a glass of milk or a piece of chalk, to a simple idea, that of white, for instance, by ignor ing anything but that simple component of the given complex idea. In 3 however, we move from a complex idea to another complex idea, this time by removing only some components of the original idea while what is left becomes the new general idea. Following produced 2 simple ideas and complex ideas of substances. Certainly, there are two types of gen eral ideas here. E.J. Lowe offers a perhaps better way of thinking of the distinction when he 1 The Cambridge Companion to Locke edited by Vere Chappell (Cambridge: Cambridge University Press, 1994), 39. 2 Ibid., 41.
130 what are sometimes explicitly called abstract nouns sortal and mass terms 3 discussions of abstraction show two different applica tions of that mental operation, I will argue there is really only one general theory of abstraction at work only one way of abstraction which, when applied to different extents yields different results. While I will come back to address Ch directly in Â§4 my strategy in this chapter will be to primarily argue against his position by showing that Locke does have a single, unified theory of abstrac tion and general ideas (in Â§Â§1 4 ). I will address the particular question of ory requires the mental separation of ideas in a robust sense rather than merely partial consideration in Â§3. Having a, negative, answer to that question will help deal with some of the objections that have been raised against Lockean abstraction, most not ably by Berkeley and, by extension, Hume. Essay I will address a number of problems commentators have had in regard to what they take to be that theory. In Â§5 I will discuss the troubl esome 4 7 .9, famous for its assertion that the general idea of triangle Idea wherein some parts of several different and inconsistent Ideas are put together (4.7.9). If Locke is to be understood as offering a single viable theory, it is critical to show that he does not contradict himself later in the Essay In Â§6, I will explain and address two important problems for Lockean abstraction that have recently been raised by Matthew Stuart. Finally, in Â§7, I will consider three worries raised by 3 E.J. Lowe, Locke on Human Understanding (New York: Routledge, 1995), 154.
131 the arguments I have offered in Chapter 4. Ultimately, this chapter argues that the Essay has a single, unified theory of abs traction, which manages to avoid most of the objections commentators have raised against it. 5. 1 Essay 2 11 p rocess of abstraction in Book 2 of the Essay Though it achieves more than this, abstraction is ostensibly introduced to avoid the need for an infinite number of names for the endless supply of particular ideas: The Mind makes the particular Ideas received from particular Objects, to become general; which is done by considering them as they are in the Mind such Appearances, separate from all other Existences, and the circumstances of real Existence, as Time, Place, or any other concomitant Ideas This is called ABSTRACTION whereby Ideas taken from particular Beings, become general Representatives of all of the same kind; and their Names general Names, applicable to whatever exists conformable to such abstract Ideas Such precise, naked Appearances in the Mind, without considering, how, whence, or with what oth er they came there, the Understanding lays up (with Names commonly annexed to them) as the Standards to rank real Existences into sorts, as they agree with these Patterns, and to denominate them accordingly. Thus the same Colour being observed to day in Ch alk or Snow, which the Mind yesterday received from Milk, it considers that Appearance alone, makes it a representative of all of that kind; and having given it the name Whiteness it by that sound signifies the same quality wheresoever to be et with; and thus Universals, whether Ideas or Terms, are made (2.11.9). In order to avoid the need to store and use an unmanageably vast number of particular ideas, which is what is given to us by raw experience, Locke says we are able to use a particular to such abstract Ideas general idea. The process that results in such a representative, or general, idea Locke identifies
132 Ideas by considering ABSTRACTION brief explanation of how this process works. As I have already emphasized, 4 m ost, if not all, of the ideas given by experience come bundled with any number of other ideas. We do not have experience of just redness, for instance, but of red things: red balls, red wagons, red lipstick, etc. To avoid having to have an endless number of names and ideas (e.g. the color of that ball, the color of that w agon, the color of that lipstick, etc.), we use our faculty of abstraction, which consists in the consideration of a particular idea to the exclusion of any other ideas that may come with it, as part of a more complex idea or ideas, either necessarily or i ncidentally. We consider only the particular idea of the color of the ball, and ignore the size, shape and bounciness of it. i.e. things that are red), and the name we use to signify that these abstract ideas to help us organize the things we come across in the world into all the ki nds, or sorts, we find useful, as well as to give ourselves names for these kinds (i.e. things that are red things that are square things that are human etc.). As Locke says, these abstract ideas have to. Locke here seems to think that joining a name with the idea is a natural part of the process, but it does not seem to be necessary. One could, for instance, have a general idea of a particular color without ever giving it a name. Abstract ideas with names, though, have an 4 In Chapter 3.
133 important advantage in that referring to them by name makes communication possible which is an important element in the overall project of the Essay Naming our ideas also arguably, makes thinking itself much easier. If the name of an idea serves as a kind of tag, or abbreviation, for the idea a representative of the content of the idea it seems to make sense to think we can work with the idea more quickly and easily by reference to the nam e. In our own, private, mental discourse we can, perhaps, work with named ideas without calling up all the content of those ideas, with a resulting increase in efficiency. It also seems plausible to think that having too many unnamed abstract ideas would p rove a debilitating handicap because we would have to hold on to and call up the ideas without the benefit of those names 5 In any case, according to the end of the passage, any general idea s that we do have are made through this process of abstraction. say, a swatch of white color, in regard to which we consider only the color, which stands for any sufficiently similarly colored thing we might come across, and which we give the name This is, Locke says, how we make general ideas and terms. Note, however, that the other ideas that come along with it (e.g., a bstracting the redness from the bounciness, size and shape of the ball), but also seems to include the use of the resultant abstract idea as a representative for any other instance of, say, redness. That is, there are two elements to the process Locke is c 5 One might object that if the names of our general ideas make thinking more efficient, it ought to make our thinking more efficient still if we grouped our ideas, by their names, under other more general headings, with names. This might threaten a regress of sorts. Of course, we do just that kind of grouping and it does seem to help in our thinking. We do not seem to be affected by a regress, though, perhaps because there is a limit to the useful generalizations available to us. We quite sensibly, though, group our ideas according to the content of the ideas, not their names, which are merely arbitrary labels put on the ideas.
134 our use of the abstract idea as such a general idea which stands for more particular ideas and things. I n this way, it seems that we can draw a distinction between an abstract idea and a general idea; the first is merely one that is the result of the particular process of abstraction Locke describes, while the second is such an abstract idea that has been ma de general by being used as a representative idea. having a general idea includes using it in a certain way, as a representative idea. 6 The above passage leaves us with several important question s. Is this all there really is to as the set of examples in this part of the Essay indicates? How does the process he has suggested apply to the various sor ts of ideas Locke identifies? And, should Locke be understood as holding that abstraction always requires actually mentally separating away ideas from each other, which is problematic for a number of reasons (for example, it would seem to leave Locke open to one ires us to think of things that are impossible to think of), or is partial consideration enough? In the next sections, I will s presented in the rest of Book 2 ; in Â§2 I will discuss the application of abstraction to modes and complex ideas, in Â§3 I will consider the issue of separation versus partial consideration. 6 James Gibson enough that the mind should single out a certain content, as the object of its consideration. It must consider this content in a determinate manner, viz. as representing all the particular things in which the same quality may be James Gibson, Theory of Knowledge and its Historical Relations (Cambridge: Cambridge University Press, 1960 (1917)) 70
135 5.2 Of Modes and Substance Kinds So far, the example cases I have considered to any extent have mostly been what Locke calls simple ideas, particularly simple ideas of colors and of extension. However, we also have general ideas like dozen mile and day which Locke calls simple modes, as well as general ideas of mixed modes, things like murder and covetousness and ideas of substance kinds, e.g. man animal and table How are these simple general ideas and general ideas of simple and mixed modes and substance kinds related in regard to abstraction? According to Locke (2.13.1) of the same simple Idea (2.12.5). For instance, the idea of dozen Ideas (2.12.5) where Unite unit ) has already been shown to be a simple idea, and a simple general idea at that. We may get the idea of the simple mode from experience or cre atively come up with it, but it Ideas of several kinds, put together to make one complex one; v.g. Beauty consisting of a certain composition of Colour and Figure, causing delight in the Beholder (2.12.5), and so forth. Locke does not deny that these ideas might come Ideas so combined, as they are put together in t (2.22.2), The Mind often exercises an active Power in the making of these several Combinations (2.22.2). One can Ideas (2.22.2) or
136 Ideas which these words [ Sacrilege or Murther ] stand for, without ever seeing either of them committed (2 .22.3). 7 Of course, all of the ideas of modes Locke discusses are also general ideas, e.g. not this particular dozen but dozen and not a particular instance of beauty but the idea of beauty itself. When introducing his discussion of complex ideas, Locke reminds his readers that it is through abstraction, on his understanding of it, that all our general ideas are made (2.12.1). So, even though Locke does not directly refer to any of these complex ideas as abstract ideas, they nevertheless count as such acc ording to his theory. In any case, their constituent ideas are all general, which suggests that they must also be general. Ideas of particular sorts of Substances 8 mple Ideas as are by Experience and Observation particular internal Constitution, or un known Essence of that Substance (2.23.3). Locke is at great pains to ind icate that our ideas of these kinds, or sorts, are complex ideas derived from the etc. (2.23.3) ), which are Idea of something to which they belong, and in whic (2.23.3), (2.23.4). Therefore, when we have an idea of a kind of substance, say gold, we do not merely put together the various simple (or relatively simple) id eas of yellow color, malleability, density, etc., as we might if we were just thinking of a basic mixed mode, but we also think that there is 7 See also 2 22 .9. 8 Ideas simple ideas belong to some thing Substratum Substance Ideas of particular sorts of Substance s (2.23.1) This means that ideas of substance kinds really are a subset of mixed modes combinations of different kinds of simple ideas,
137 some thing that is yellow, some thing with a particular density and degree of malleability. So, while we can prett y much put whatever ideas we like together to form other complex modes, idea of an underlying substratum. It seems that even the idea of a unicorn would have to include the idea of substratum: the something in which the various unicorn attributes inhere, that which is like a horse and has a horn. For now, it will suffice if we consider how general ideas of modes in general, whether they are simple, mixed, or ideas of substances, can be seen to be the result of abstraction as Locke has described it. First, to take dozen as an example of a general simple mode, we either observe twelve units and decide this is a useful counting term, or we add up twelve units in our m ind and set this as the standard we will use. In either case, we focus our attention simply on the number of units involved, not on how the units are arranged or what they are units of (whether they are eggs, donuts, or numbers), so that we have an abstrac t idea of a dozen. We then say we will use this abstract idea to represent any grouping of twelve things whatever they may be whenever it is convenient and appropriate 9 to do so (perhaps only when they are things of the same sort). So, dozen is an idea tha t is made general by a process of selective consideration and being made appropriately representative in other words by Lockean abstraction. It does not matter if one starts with a group of twelve things or with one thing, abstracts to unit, and then multi plies that unit twelve times; either way, you have used the same general process of abstraction to arrive at dozen. 10 Of course, the foregoing assumes the generality of the idea unit Locke says that there is 9 certainly may have failed to consider all the relev ant options 10 interesting in its own right. I employ this flexibility in Â§6 of this chapter.
138 there is none more simple, than that of Unity Idea in our Understandings; every Thought of our Minds brings this Idea along with it (2.16.1). 11 So, we can start from any idea of anything at all and abstract away until we come down to u nity and thereby one unit (a single unity), which means that the general idea of unit represents any particular thing whatsoever, and you cannot get much more general than that. I take it to be a straight forward task to apply what was said above about do zen to other general ideas of simple modes. Second, to take murder as an example of a general mixed mode, we observe a deliberate, wrongful killing 12 or imagine such a case. We ignore things that we are not interested in for the purposes of focusing in on the general notion of like the identity of the murderer, the time of day or night, the weapon used, etc. Certainly, the who, what, when, where, why, and how of the matter are important facts, but they are not what you need t o know to decide whether something is a murder. Once we have done this selective consideration, we use this abstract idea to represent any such deliberate, wrongful killing, whatever the other, attendant circumstances may be. So, again, we have a general i dea that is the result of a process of Lockean abstraction. Third, in regard to substance kinds and taking gold as an example (it was apparently circumstantial differences (size, shape, ownership, etc.), and come to have an abstract idea which we use to represent in our minds any bit of gold we later come across. Because we cannot, as 11 Note that this includes even simple ideas. 12 Or whatever the best definition of murder is.
139 Locke says, imagine that those qualities should always be bundled together in just that way without some thing being the substratum in which they inhere, we cannot help but include the confused idea of substance along with the other ideas that make up this complex mode. This confused idea of substance may also be a result of abstraction, if we come up with this notoriously murky idea by ignoring all those qualities of an existent we can observe and arrive at the notion of that in which those qualities inhere, and we use that notion, as ill defined or c onfused as it may be, to represent any substratum whatever. Our general ideas of both simple and mixed modes and of substance, as well as simple ideas, then, are the result of Lockean abstraction. The only significant distinction, as far as abstraction go es, is the degree to which we abstract. In a sense, we focus in on a more or less abstract level : for a mixed mode, we selectively ignore a wide range of attributes (the perpetrator of the murder, the victim, the murder weapon, etc.), including the idea of a substratum for an idea of a substance. For a simple mode, we ignore an even wider range of attributes (until we have only something in focus that is one simple idea variously combined). And, for a simple idea we ignore everything but that simple idea. I n every case, we then use whatever we end up with as a representative of anything else that properly corresponds to that general idea, perhaps anything that could yield the same general idea if put through the process of abstraction (less the representatio n part), or, as I suggested in Chapter 3 whatever matches up with the simple combination of the idea. Of course, as Locke has already pointed out, we may also have these ideas explained or their representative words defined for us. In this case, we do no t perform the initial work of abstraction but have it done for us, though we presumably can reverse engineer the abstracting, as it were. There certainly seems to be a sense in which we have to engage in some level of
140 abstraction to actually have the idea; at some point our understanding of cue ball has to ground out in our understanding of such simple ideas as white round and solid which we can hardly understand without having abstracted to them earlier. Much of this was suggested by my earlier discussi ons of simple combinations and representation. Locke uses are already general ideas. He does not talk about a simple mode that is only this particular dozen, or an id ea of a substance that is the particular purple, fire breathing unicorn I have in mind right now, though it certainly seems that these are each ideas of the proper sort in their own right. Instead Locke talks about dozen itself and, things like, unicorns i n general. I and human cognition in general, to point out that what he was saying about the attributes of the ideas, if not how we got them in his current cont ext (which could not be free of abstraction), applies to both the particular and general ideas of modes. Certainly, it would be needlessly tedious to repeatedly refer the reader back to his earlier discussion of abstraction, though it might have helped if he had included, at the end of that earlier discussion, something along the lines of for the most part, apply to both those and any particular ideas we might emphasis on general ideas fits in well with Locke will later say (in Book 4 ) about knowledge, truth, propositions and maxims. 5. 3 Separation or Partial Consideration e level to which we have to be able to separate the elements of our experience to arrive at the desired abstract idea. For example, to arrive at the idea of green do we have to somehow mentally separate greenness from some thing that is green? Do we have t o pull apart the color and the fact that whatever colored
141 thing we are contemplating is extended? Do we have to be able to mentally consider just being pale skinned without, for instance, some person who is pale skinned to form the idea of pale skinned ? Hi storically, Locke is taken to offer one of two answers to those questions: yes, Locke is committed to strict separation in abstraction, a position I will refer to as separatism; and no, although there may be times when it is possible to separate the elemen ts of experience cleanly consideration of aspects of experience to arrive at the desired ideas. The latter involves having an idea of, say, a green ball, but only focus ing on the color and ignoring the ball and the fact that the greenness is extended to arrive at green As will be seen below, this issue is quite important for Essay requires strict separation, objections made by Berkeley and Hume seem unavoidable, and Lockean abstraction is deeply problematic. 13 In this section I argue that Locke avoids this pitfall because the partial consideration reading is the right one. On this matter I am in accord with Michael Ayers 14 who, ci ting 2 13 .11 and 13, says that 15 16 In particular, the ideas of unity and existence, 13 In discussing my dissertation with philosophers who are not so focused on early modern philosophy, when those philosophers thought there was a problem with Locke on general ideas it has been exactly that Lo cke is committed to separatism and so committed to impossible (or at least seriously implausible) ideas. 14 As well as J.L. Mackie, see Problems from Locke (Oxford: Clarendon Press, 1976), 110 115. 15 Michael Ayers, Locke, Volume I: Epistemology (London: Rou tledge, 1991), 251. 16 Ibid.
142 mind 17 would was separated away those two remain. Separating unity from existence still gives you both of them; there is no way to have an idea of something that has existence but not unit y and vice 18 On the other hand (and naturally), my position is at odds with that taken by Jonathan Walmsley 19 Citing 3 3 Ideas become general, by separating from them the circumstances of Time, and Place, and any other Ideas that may 3 3 .9, in which Locke talks about leaving out and retaining things in the formation of abstract ideas, and rejecting Aye account of abstraction is one of mental separation in a very full blooded way, and is not a partial consideration model as Ayers [suggests]. 20 Noting that Ayers focuses on 2 13 .11 and 13 to defend the partial consideration view, Walmsley takes the position that this is a misuse of these passages from the Essay Walmsley is certainly right to say that at this point in the Essay ed in the debate about the 17 Ibid. 18 Ibid., 252. 19 British Journal for the History of Philosophy (7: 1, 1999), 123 134. 20 Ibid. 130.
143 21 Beyond this role, though, of showing that the ideas of body and space are different, even if one cannot think of the one without the oth er, Walmsley thinks these passages merely show that any given complex idea is necessarily made up of simple ideas, and that it would be a mistake to therefore conclude that those simple ideas are somehow indistinguishable from the complex idea, and therefo re from each other. According to Walmsley, Locke should be understood here as being strictly interested in considering the idea of pure space as opposed to the idea of some particular region of space without also considering the boundaries (or superficies) we use to separate that region from the rest of space, which strictly speaking is a whole, inseparable thing. Walmsley writes, It should be noted that this solution of partial consideration is designed specifically to solve this particular problem in the conception of the measurement of pure space. Indeed it is, as far as I am aware, unique to this passage in the Essay Nor does it seem peculiarly applicable to the debate surrounding abstraction, as in abstraction we separate one idea from another, i.e. c olour from taste, or light from heat. In the above case we are not attempting to do this, rather we are trying to differentiate parts of the same idea: that of pure space. This is a unique problem and receives a unique solution in the form of partial consi deration. The problem of pure space is not a problem of abstraction. 22 committed to a strictly imagistic theory of ideas, 23 I believe there are good reasons to resist 21 Ibid. 130 131. 22 Ibid., 133. 23 See Chapter 2.
144 from that of abstraction. Before we consider those reasons, though, it is worthwhile to look at the passages in quest ion a bit more closely. In 2 13 Locke turns to the problem of how someone can disting uish between body and extension. As Walmsley notes, t understanding of space and the possibility of vacuum. Locke responds to the objection that body and extension cannot be distinguished because the two things are always together, showing therefore that they are really the same thing, in part, by saying: her can Scarlet Colour exist without Extension; but this hinders not, but that they are distinct Ideas Many Ideas require others as necessary to their Existence or Conception, which yet are very distinct Ideas Motion can neither be, nor be conceived with out Space; and yet Motion i s not Space, nor Space Motion (2.13.11). A bit later, he says, commensurate to a Foot, without considering the rest; which is indeed a partial Considera tion, but not so much as mental Separation, or Division; since a Man can no more mentally divide, without considering two Superficies, separate one from the other, than he can actually divide without a partial consideration is not separating (2.13.13). three claims Locke might be making, assuming something, x that is inherently both F and G: (1) it is impossib le for x to exist an d be F without its also being G say, being red without also being extended; (2) it is impossible to think about x being G; (3) the ideas of F and G are the same idea in some sense, so thinking a bout F and thinking about G are in fact thinking about the same idea in different ways. 24 24 There is a fourth option: it is impossible for us to think about x have some G must be extended). Howeve r, this seems to me to clearly be too strong. It certainly seems plausible to think that someone is able to think about a red thing without ever having given any thought to extension as such; perhaps
145 Readi ng just the first quote (from 2 13 .11) suggests that (1) is the claim Locke is making, while reading the second (from 2 13 .13) su ggests (2). (3 ) cannot be the cl aim Locke is making. In this part of 2 13 Locke is at pains to make clear that such things as color and extension, motion and space, are distinct ideas, in spite of the fact that it is impossible for them to be actually be separated even in thought. The w hole point of the above quotations, coupled with what Locke has said about abstraction in 2 11 .9, is that we are able to consider something that is inherently and necessarily both F and G in such a way as to form an idea of just F, or G. The simple idea of red is distinct from the simple idea of extension, even though we cannot actually ever have a n experience of anything that is just the former without the latter. We cannot ever have the experience of anything that is just the simple idea of Ideas (2.13.11). In a sense, it is impossible to thi nk of something that is just F, but we have to think of something that is FG. This suggests that, as far as we are thinking of some thing that is F (in, say, an imagistic sense) we must be, according to Locke, thinking of a thing that is both F and G (and possible H and I as well) but Given the above considerations, and taking the passages from 2 13 quoted above, in context with what Locke said in 2 11 I think the best reading of his claim a t the end of 2 13 is as a conjunction of (1) and (2): Given the appropriate x n ot only is it impossible for x to exist and be F without its also being G, but it is impossible to think about x about its being G, at least imp licitly, as far as thinking about something involves imaging that thing in some sense. It does seem that one could think of something round without consciously children do this kind of thing all the time. If this is true, then it cannot be right to say it is impossible to think about knowing that it must be G, let alone knowing that you know; this would require too much intellectual sophistication from anyone thinking about anything that is both F and G.
146 being aware of the fact that it is round, but you could not have the idea of that round thing wi thout including its roundness in the idea. Locke is also saying, with his talk of partial consideration, that such an impossible separation is not necessary for having simple ideas (or a comparatively simpler idea) if it is possible to partially consider the aspect in question say to partially consider the extended red thing so that what is being considered is just the color. This seems completely compatible with what he wrote in the official discussion of abstraction in 2 11 .9, quoted in Â§1. In the earli er considering how, whence, or with what that Appearance alone, makes it a representative of all of that ki do not actually, or even mentally, need to separate the nose from the face, but simply ignore the rest of the face when focus ing my attention upon the nose for the purposes of consideration. In partiality of the consideration is implied i Here, in 2 13 Locke is saying that such a (partial) consideration is possible even if the mind cannot actually separate the specific thing (say the color) from other things (say some size or shape of the colored thing). I can now return to the reasons to resist Walmsley First, Locke certainly seems to me to be saying that partial consideration can be applied to more than just the idea of pure space. After all, in the very same part of the Essay Locke writes, A Man may consider the L ight in the Sun, without its Heat; or Mobility in a Body without its Extension, without thinking of their separation. One is a partial Consideration, terminating in one alone; and the other is a Consideration of both, as existing separately (2.13.13)
147 It hardly seems to make sense to think that Locke was talking about a unique solution to a unique problem. Granted, the second of these is again about extension, but the sense of the passage clearly gestures toward the general applicability of what Locke is saying. Of course, the passages are actually about particular cases, so one may wonder how far the idea can be generalized. However, absent an argument for why we should think of these as particularly special cases, there seems to be no reason why one coul d not generalize them to cover any situation where some idea or quality is inseparable even in thought from some other idea or quality (or set thereof). There is nothing in the surrounding passages that suggest Locke was thinking that these were special ca ses, rather Locke seems to me to be applying his general thinking to particular cases that are problematic on other theories. Moreover, i t is not difficult at all to come up with countless other cases of necessarily FG things, so it would be almost unchari table to think that Locke would intend for his discussion here to apply only to these particular cases without some word from him as to why and what to do with the countless other cases. It seems possible, however, to read Locke here as talking about qual ities of things that cannot be separated, and not ideas. This may very well be true, but such a reading hardly suggests that Locke is somehow committed to a full separation when it comes to ideas. Moreover, if one thinks Locke is an imagist and both Walms ley and Ayers (and Berkeley and Hume) do the distinction between inseparable qualities of things and purportedly separable components of the ideas of those things, which are supposed to be images of those things, seems to me to become very hard to follow. Combining imagism with separatism seems to lead us right have mental images which are, in fact, impossible to have. Fortunately, Locke does not actually
148 seem t o be saying anything like full separation is necessary, even if it might be possible, though it is often not possible, as in the foregoing examples. Again, it is very hard to see how what Locke says about unity can make sense at all on a non partial consid Idea (2.16.1). I confess that I cannot see how it is at all possible we could have an idea of unity at all on a separatist re ading. theory as presented up to this point in the Essay and thereafter, this seems more likely to be the application of a general principle Locke had in mind to a particu lar problem, that of space and body. Specifically read in this way it is not at all difficult to see that what Locke is talking about is very much the same process he discussed under the name of abstraction in 2 11 .9 differentiating between the light and the heat of the self same idea, e.g. the Sun, and focusing on only one of them, and so forth. separate begging the question in this particular context (unless Walmsley is to be understood as having part). Finally, i t seems to me that Walmsley is simply wrong when he says: In the [2.13.13] case we are not attempting to [separate one idea from another] rather we are trying to differentiate parts of the same idea: that of pure space. This is a unique problem and recei ves a unique solution in the form of partial consideration. The problem of pure space is not a problem of abstraction. 25 25 Walmsley 133.
149 In the discussion of pure space that runs through (at least) this part of 2.13, Locke is dealing with difficulties in understanding how we are to think of pure space and particular bits of space, but for Locke we cannot think of those particular bits witho ut having ideas of them, and we cannot have ideas of those discrete parts of space unless we employ abstraction. In short, the problem of pure space clearly is a problem of abstraction, and it is one that raises the specific problem of separation versus pa rtial consideration. Moreover, the response that Locke gives to that problem is explicitly in favor of partial consideration. Given these considerations, it appears that the only ] a problem of because it does not involve separation and so could not be abstraction, which does require separation, but this would simply be a case of begging the question. Walmsley goes on to point out that the discussion in 2 13 .11 13 m akes it abundantly clear that Locke understood that there was a difference between mental separation and partial consideration, but Locke did not go through and specify that he meant partial consideration in the parts of the Essay where he is clearly talki ng about abstraction. This is particularly odd, Walmsley thinks, in the face of the fact that Locke often made corrections to the further editions of the Essay However, as we have already seen, Ayers has pointed out that in some cases, as in that of the i all the corrections he perhaps should have. In addition, I would point out in particular that Locke says that we make other Existences, and the circumstances of real Existence, as Time, Place or any other concomitant Ideas (2.11.9). We consider an idea as separate from extraneous other existences
150 and ideas. em in the mind from each other and this passage, I submit, suggests the former over the latter. Here, explicit discussion of partial consideration serves to reinforce this intuition. Now, i t must be acknowledged that Locke says color and extension are distinct ideas, which might seem to suggest that Locke does have separation in mind. As we have seen, for Locke, for two ideas to be distinct it must just be the case that the mind can readily distinguish them from one another So, distinct ideas must be separate ideas, and that might suggest that Locke is committed to separation of some sort. However, first, it is possible to think that our idea of a particular color somehow includes the fact t hat to be colored something must be extended. This would, perhaps, follow if what I have said about simple combinations is the best way to understand Locke. Our idea of color is distinct in the sense that the content of that idea differs from that of exten sion not that color does not at all involve extension 26 Second, the question at hand is not about whether ideas are distinct, but about what is involved in the process nd do, engage in a process that does not involve the mental separation of qualities in particular cases but results in distinct ideas that are distinct in their abstract nature. Third, and finally, this potential objection seems to implicitly presuppose im agism. The objection here requires us to be thinking that the ideas of color and extension are the (imagistic) ideas of colored and extended things, All of this does, however, lea ve what Locke says in III.iii to be dealt with. There, Locke Ideas : and Ideas become 26 Note that Ayers has also pointed out that all ideas will involve unity
151 general, by separating from them the circumstances of Time, and Place, and any other Ideas that may determine them to th is or that particular Existence (3.3.6). Obviously, and as Walmsley pointed out, Locke is using slightly different language here from his initial discussion of abstraction in 2 11 ; he explic While I believe I have shown why it makes sense to think that Locke was thinking of partial consideration when he was d iscussing abstraction in Book 2 (above), there seems to be something different goin g on here, in 3 On the basis of the foregoing discussion I think it would blooded sense all along, as Walmsley would have us believe. Rather, either (a) Locke is being inconsistent or (b) he is spe aking loosely here. Against (a), there is the fact that, as Walmsley pointed out, Locke was quite well aware of the difference between mental separation and partial consideration, and, presumably, of the dangers of relying too heavily upon the former. On my reading, Locke has said, in 2 13 .13 (above), that mental separation of ideas was not needed when partial consideration would do. If Locke slips into separatism here, his project is inconsistent or he has modified it in such a way as to make it vulnerabl e to the issues raised by Berkeley and company 27 to which I will return short ly. If the partial consideration option is still viable, and it certainly does seem to be, there is no reason to think that Locke is theoretically committed to separatism, as Walm sley thinks he is. though one might have wished he had chosen different words in the above passage. In particular, because of the efforts Locke made to edit and cla rify the material in the Essay over the course of 27 Principles is concerned.
152 subsequent editions it makes sense to think that Locke considered his overall theory consistent, so without good reason to think he was wrong we should not rush to think his project is inconsistent. In fav or of (b), recall that Locke has already laid out his official doctrine of abstraction in Book 2 He has said that in abstraction we consider ideas as separate from extraneous existences, and, later, made the point that full mental separation of idea from idea is not necessary when partial consideration is sufficient. To review this each time abstraction is discussed seems needlessly repetitive. Also, having given us the official doctrine in Book 2 here in 3 Locke is interested in moving on to other issues which he promptly does. Finally, there is no question that there are many cases of abstraction that actually do involve separation, in 3 Locke is primarily concerned with such cases, but this does not mean that Locke is com mitted to separatism across the board. For these reasons, I think it makes the most sense to conclud e that in the above passage (3 3 conflation is perhaps brought o n by the fact that prior to Book 3 Locke was largely concerned with imagistic ideas (those based upon sensible things), and he is now considering ideas more generally. One might reasonably ask, if Locke is sloppy about this distinction, how can we hope to be clear about what he thought about it? I can only say that if we read the passage ( 3 3 .6) as referring to an act of partial consideration what Locke says here fits in well with his overall theory, and if we do not we have an isolated bit of separatism th at neither fits with the theory or really makes good independent sense. He has said that ideas become general by considering them aside from their particular circumstances and using them as representatives, etc., and he has observed that sometimes it is im possible to conceive of them as being distinct from other ideas
153 but this is not necessary. So, regardless of what he says here, there seems to be no good reason to think Locke actually thought abstraction required separation in any full blooded sense. In s in Book 2 as a whole, I think the partial consideration understanding of abstraction is the best way to read Locke, notwithstanding what he seems to be saying in 3 3 As I said at the beginning of this section, t his is an important issue for Locke. S eparatism w ould leave his theory vulnerable to Berkeleyan objections. Donald L.M. Baxter has offered just such an objection that is well considered and insightful. 28 According to Baxter, Lockean 29 Explicitly following Berkeley, Baxter focuses upon the passage from 3 3 we have been considering. So, he concludes, 30 31 Indeed, this is how Locke talks, in this small part of the Essay Just reading the 3 3 discussion will, naturally enough, lead one to think th at Locke is committed to a full blooded separatism. Such a separatism then leads inexorably to a serious problem for Locke, as Baxter clearly shows. Even if we allow that Locke had a not strict ly imagistic understanding of ideas, and allowing for ideas as what Baxter calls Cartesian intentional objects 32 Locke is still in trouble. Baxter explains that 28 Philosophy and Phenomenological Research Vol. 57, No. 2 (June, 1997), 307 330. 29 Ibid., 307. 30 Ibid., 314. 31 Ibid. 32 Ibid., 315 318.
154 exemplified by the triangle, i.e. isosceleity, scaleneity, etc.) and therefore is (plausibly) identical with said specification, even if our idea of tri angle is that of something that need not exist as such, but only as an intentional object. As a result of this inseparability, Baxter argues, we are qual 33 Triangle is thought to be triangular but not some particular specification of triangularity, which is impossible because triangularity is inseparable, and therefore identical to, the specification; that is, it is impossible to have triangularity wit hout some specific kind of triangularity. 34 So, if Locke is committed to separatism, as Baxter thinks he is, Berkeley and Hume are right to reject Lockean abstraction. Baxter acknowledges that a variety of the partial consideration view, whereby, for instan considering the triangle, perhaps, but not as isosceles 35 will adequately respo nd to the challenge he has presented. This is precisely the sort of view I have been endeavoring to elaborate and support However, while he thinks 36 In several footnotes 37 Baxter considers the same points raised in favor of the sepa ratist position discussed above. Ultimately, he comes down in 33 Ibid., 328. 34 All of this presumes that the idea of triangle is a legitimately imagistic idea, because it is of primary qualities. While I do not quibble about this here, the discussion in Chapter 3 coupled with a simple combination for triangle suggest s that such an idea need not be a mental image at all. 35 Ibid., 329. 36 Ibid. 37 Ibid., 328 330.
155 favor of reading Locke as strictly committed to a literal reading of his own language, meaning partial consideration only in the few cases where he specifically says so, and (full blooded) separ ation where he speaks of separation. I have given my reasoning for taking Locke to not be so committed, above. I only add o Locke, a similar motivation suggests that we should think Locke had a partial consideration view. If we operate under the principle of charity, we are obliged to, wherever possible and wherever plausible options present themselves, read a y in the way that allows for the most coherence and least number of fatal flaws, provided that doing so does not engage in contradicting the text. I have argued for such a reading. Since the partial consideration reading is viable, we should not think Lock e is committed to separatism. To do so would, as Baxter so ably demonstrates, result in Lockean abstraction falling to a truly devastating objection. 5. 4 Summary the Theory With the foregoing discussion in mind, and incorporating the discussions of the pre vious chapters, ry of general ideas is possible: 1. General ideas are (originally) derived by engaging in a process of what I am calling Lockean abstraction, which has three components: 1a. The abstractive process by whic h the mind selectively considers certain qualities presented by experience and ignores others. 1b. The retention to some degree of what was ignored in the abstractive process (1a) (see (5)) excepting any truly irrelevant details (like the writing on a ba ll). This may sometimes involve separation of ideas, but need not (see (3)) 1c. The use of the abstract idea (the result of (1a)) to represent anything else that appropriately corresponds to that abstract idea, which makes that abstract idea a general id ea.
156 2. This process (1) applies to any general idea, whether simple, simple mode, mixed mode, or idea of a substance. The difference between these ideas, as far as their being general ideas, is only the degree to which (1a) is carried out, and if the idea inc ludes the idea of a substratum. 3. (1) does not require the separation (even just in thought) of qualities from one another, but only the selective focus upon one (or some set) while ignoring others. This is particularly important in considering the m ore fundamental ideas (like color and shape ) upon which our more complex general ideas are based, or from which our general ideas are made. Where separation poses no difficulty, we can separate, but it is not required 4. The result of (1) is 4a. some sort of image when the general idea is the right sort of simple idea (e.g. an idea derived more or less directly from sensation); 4b. probably not an image when the general idea is that of a mode. These more complicated general ideas are reducible, at least in part, to simpler ideas, each of which is derived, or derivable, from the same sort of process, i.e. (1). 5. The result of (1a) and (1b) is either a simple idea or a complex one The latter is best understood as a simple combination. 38 Clearly (1) is doing most of the work in this theory, while (2) (5) are merely providing some of the important details. A brief word about (1) seems in order. Given the way the theory is stated, one could follow the process to form an abstract idea from one parti cular idea. Say, to borrow a rose) for the very first time not that particular red thing, but the first time she sees any red thing. On the above theory, it seems possible that Mary co uld stop right there, before she sees any other red things and run the abstractive process to get an abstract idea of red The problem, perhaps, with this is that we actually seem to need several encounters with red 38 See Chapter 3.
157 things before we can form such an abstra ct idea. Not only is this suggested by his description of what we actually do in forming the idea of person in 3 3 same Colour being observed to day in Cha lk or Snow, which the Mind yesterday received from Milk, it considers that Appearance alone, makes it a representative of all of that kind; (2.11.9). On his explicit account we seem to be using chalk, snow and milk to come up with the general idea of white My response to this is that while it may be the case that we do in fact abstract on the basis of having noticed that some things of our experience can be grouped together, we do not strictly have to go through this route to form an abstract idea. Ma ry could form any number of abstract ideas from her encounter with that first red thing, but it is hard to see why she would bother. On the other hand, to use this abstract idea to stand for any corresponding thing, i.e. to make the abstract idea a general idea (1c), presupposes that there are, are likely to be, or could be, several things of the appropriate type. now explicit in the foregoing enumeration. Chappell says that in 2 11 .9 and 3 3 39 In the Book 2 version of abstraction, we move from complex idea to simple idea, while in Book 3 we move from a particular complex i dea to another complex idea. In the former, one starts with some particular complex idea, selects one component of that complex and focuses on it, ignoring everything else, so that a (comparatively) simple idea is reached. In the latter, 39 Chappell, 39.
158 by removing several components from this complex idea, say the the original idea, or what is left of it, which is now no more than the idea of a woman some woman or 40 The goals of the two processes are the very different kinds of ideas, simple and complex. Chappell is struck by the undeniable difference betwee n the ideas discussed in Book 2 and Book 3 He points out that in the case of simple ideas Locke seems nature, and that their application to particular individuals is determined by factors extraneous to them as such Time, and Pl 41 On the other hand, in Book 3 the general idea is that of man [individuals] are left out [by this process of abstraction] are not merely the extraneous circumstances of time and 42 Surely there is a distinction to be made here between the kinds of things that are being selectively cons idered away, and there is no doubt that Locke has two very different sorts of ideas in mind in these two different parts of the Essay Nevertheless, I am convinced that Chappell is wrong to think that Locke is describing two different forms of abstraction. (in fact, Locke recognizes four 43 40 Ibid., 39 40. 41 Ibid., 41 42. 42 Ibid., 42. 43 Simple general ideas, simple general modes, complex general modes, and general ideas of substance.
159 44 but to the degree to which the same pr ocess of abstraction is carried out. Chappell has failed to sufficiently generalize the process Locke has described. To begin with, the distinction between extraneous circumstances and proper qualities that Chappell makes is entirely arbitrary as far as Lo through abstraction. Then, whenever we consider some partial set of ideas that come with a given complex of ideas we are engaging in Lockean abstraction. If we push that p rocess until we arrive at a single simple idea and then make that idea representative of all appropriately similar simple ideas, we have a simple general idea, but if we stop short of a simple idea we will have a general idea of a mode. This is spelled out in (2) above, and has been argued for in Â§Â§ 1 and 2 and Chapter 3 Even in the case when we abstract to a simple general idea, we retain, as in (1b), the elements we are ignoring, such as extension, shape, particular item, etc., just as we do for the mor e complex general ideas Locke discusses in 3 3 though this is not so obvious 2 13 actually emphasizes the fact that we cannot help but retain those ignored elements. This is a big part of why someo ne who really has the simple general idea of red knows that it makes no sense to talk of a red murder or a red odor. Why, then, are the examples and discussion noticeably different in the two books? I believe the answer to this is the different projects of the two books. In 2 Locke is interested in showing how we get our ideas as such, and for expository (or pedagogic) reasons focuses primarily upon simple ideas. As I have shown (in Â§2), however, he does 44 Ibid., 41.
160 furnish the tools for moving beyond simple ideas in Book 2 In 3 Locke is working on the philosophy of language and thought, and is therefore more focused on complex modes and ideas of substance. The fact that our more complicated ideas are handled somewhat differently, by virtue of their complexity, does not mean, however, that they are not grounded in simpler general ideas as discussed in 2 Furthermore, as Locke makes quite clear at the beginning of 3 he has the plan of moving on to the topic of knowledge in 4 and for that purpose he needs to make sure that we understand how his theory about general ideas applies to such terms as man sacrilege and murder Now that I have laid out Lockean abstraction as such, I can turn to dealing with some of the objections that have been raised against it, both histo rically and more recently. 5. 5 Essay 4 7 .9: Those Damn Triangles! The passage of the Essay on abstraction that has historically caused the most consternation is no doubt 4 7 .9, in which Locke seems to contradict himself and leave himself open to serious difficulty. The passage read s: For when we nicely reflect upon them we shall find, that general Ideas are Fictions and Contrivances of the Mind, that carry difficulty with them, and do not so easily offer themselves, as we are apt to imagine. For example, Does it not require some pains and skill to form the general Idea of a Triangle (which is yet none of the most abstract, comprehensive, and difficult,) for it must be neither Oblique, nor Rectangle, neither Equilateral, Equicrural, nor Scalenon; but all a nd none of these at once. In effect, it is something imperfect, that cannot exist; an Idea wherein some parts of several different and inconsistent Ideas are put together (4.7.9) This passage has had a powerful effect on quite a few commentators. In the Introduction to his Principles of Human Understanding Berkeley refers to this passage as clear evidence that what Locke has in mind when he refers to general ideas is the result of exactly the kind of abstraction Berkeley rejects that is, a process that results in ideas that somehow combine
161 contradictory elements or are a kind of unparticularized image. 45 Both Vere Chappell 46 and Richard Aaron 47 think Berkeley and many others are guilt y of misreading this passage, and so of 48 In reference to a triangle being dramatic exaggeration of the 49 formation of general ideas 50 While Lo cke is discussing the psychology of the formation of general ideas in 4 7 .9, the passage quoted above should not be seen as primarily a part of an maxims something of a coda to his discussion in Book 1 As I will show, below, working through Book 4 up to this point, it becomes clear that the point Locke is making with the above quote is that if we had innate maxims, as Locke understands them, we would also have to ha ve innate general ideas 51 and trying to work out what kind of sense we could possibly make of such 45 George Berkeley, Princ iples of Human Knowledge edited by Howard Robinson (Oxford: Oxford University Press, 1996), Introduction Â§13, 15. 46 Chappell, 43 44. 47 Richard I. Aaron, John Locke (Clarendon: Oxford, 1971), 195 197. 48 Chappell, 43. 49 Mackie, 116. 50 John Yolton, A Locke D ictionary (Oxford: Blackwell, 1993), 8. 51 Because the maxims would have to involve general ideas, else they would not be maxims.
162 notions is extremely difficult. While, if we understand them to have been formed as a result of the processes Locke has described in Books 2 and 3 the diffic ulties are much less severe. 52 In his Introduction (1 1 ), Locke identifies two things that he will be discussing in Book 4 endeavour to shew, what Knowledg e the Understanding hath by those Ideas [which he discussed in II]; and the Certainty, Evidence, and Extent of it (1.1.3). He begins Book 4 with the, which are the immediate objects of our thinking and reasoning (4.11) Locke goes on from there the perception of the connexion and agreement, or disagreement and repugnancy of any of our Ideas (4.1.2). The first chapter o f Book 4 is concerned with working out the details of how we have this perception of agreement or disagreement, and in what ways it applies to ideas. Chapters 2 3 and 4 are spent working out the of knowledge, and its extent and reality that is, whether our knowledge is about things as they really are, as opposed to merely some sort of mind game, since knowledge is supposed to only be conversant about our ideas. Having laid out his position on kn owledge as such, Locke then, in chapter 5 turns to the the joining or separating of Signs, as the Things signified by them, do agr ee or disagree one with another (4.5.2). Thi s general correspondence theory of truth provides the link between truth and the reality of knowledge, discussed in the Essay previous chapter. If our ideas properly correspond to what they are supposed to be ideas of then our propositions involving those ideas, if they 52 future.
163 correctly relate the ideas to one another, may properly be thought to be true. This is why Locke emphasizes the need for having clear and well defined ideas and discusses at length his worry about the confusion of words and ideas. He distinguishes mental and verbal propositions, the former of which involve putting together ideas in our minds and seeing their agreement or disagreement without the use of words. For instance, if I am looking for my keys an d see another set of keys, I might compare the idea of this present set with my idea of my own keys and see that they do not agree, and no words need to be involved in this process. Verbal propositions, on the other hand, involve the use of words, which ar e needed for complex propositions. Locke considers the worry that truth might then be thought to break down into mental and verbal varieties, but reminds the reader of what he has said involving the correspondence of our ideas with reality, and the appropr iate signification of our words, which he thinks will safeguard the reality of the truth of our true propositions. He is, however, explicitly wary (to put it mildly) of claiming truth for our propositions concerning real essences, at least as far as those real essences include any reference to an underlying substance Chapter 6 what he has said prior to this, Locke emphasizes the need for clarity in our general ideas, which is often not a real problem (particularly when they concern simple ideas), but sometimes means we cannot claim certain knowledge, for instance when our propositio ns are about the real essences of things that involve substance To consider our two examples, we do not have assurance that our idea of man corresponds perfectly with the real essence of man, so we cannot true, whereas our idea of triangle is perfectly inclusive of the true attributes of triangles, so we can be sure that the second universal
164 proposition is true. There may also be problems with general claims about substances that are not directly about the ir real essences, if that real essence has some bearing on the issue. As Locke General Propositions of what kind soever, are then only capable of Certainty when the Terms used in them, stand for such Ideas whose agreement or disagree ment, as there expressed, is capable to be discovered by us (4.6.16). This shows, Locke says, that what certainty we have in regard to general things is only to be found in the adequacy of our ideas; anything having to do with things other than those must involve some degree of doubt, because our experience is always about particulars, about which we can never have perfect knowledge. It is the fact that we are able to frame general ideas, which we may understand adequately, that enables us to have general Ideas that alone is able to afford us general Knowledge (4.6.16). Chapter 7 under the name of Maxims and Axioms have passed (4.7.1) and 4 8 consider more particular sorts of knowledge, e.g. knowledge of the existence of God and things other than ourselves. So, we can see that chapter 7 knowledge in general and our truth claims in particular. He has staked out a correspondence theory of truth, and argued for a limited range of propositions that can be made with certain knowledge. Prior to discussing maxims, he has established that general propositions, which include maxims of the sort he wants to discuss, have what certainty they have as a result of the adequacy of our general, or abstract, ideas. So, in su mmary, we start out with the particulars furnished to us by experience and only arrive at workable general ideas by an application of
165 mental effort; if we came equipped wit those general ideas would, he thinks, cause more tro uble than help. 53 To pursue these issues any further is beyond the scope of this chapter. Relevant to may be in its consequences, is not a contradiction of what he wrot e in Books 2 and 3 but on abstract or general ideas in a problematic way. The possibility that Locke might be contradicting his earlier discussion of abstract id g imperfect, of the general idea of a triangle. To understand the general idea of a triangle prope rly, one has to understand that it is a general idea that covers all kinds of particular triangles, even though some of the attributes of one sort of triangle are incompatible with those of another, e.g. an equilateral triangle has angles and a ratio betwe en the lengths of its sides unlike those of a right triangle. When thought of in this way, as including inconsistent and sometimes contradictory elements, what the general idea of triangle is cannot exist. That is, nothing that contains contradictory aspec ts or features can exist. Can Locke possibly mean that the general idea itself cannot exist? He has taken great pai ns (in Books 2 and 3 ) to explain how the mind forms, or frames, general ideas, and talks about general ideas all over the place in the Essay Surely he thinks that not only can general ideas exist, but that we have lots of them. So, how should one make sense of what Locke is saying here in 4 7 ? 53 For more on this, see Aaron, 195 196.
166 As we have seen (Â§1), in 2 11 .9, Locke describes the mental process of abstraction as something invol ving the consideration of particular ideas apart from the other ideas they are associated with in experience, say of the idea of red apart from the spherical shape and bounciness that come with our experience of a particular red ball. This account is modif ied, to 2 13 .11 13, that it is impossible to separate, even just in thought, some things from others, like color from some degree of extension, but that this is not needed for abstraction when it is possible that the mind can merely partially consider things so that the desired idea is focused upon (Â§3). Also in 3 3 .7 9, Locke expands upon his discussion of general terms and supplements these considerations by explaining how, after we have abstracte d to get at whatever it is that we want to generalize, we idea and leaving out what is not. We need to move toward having a simple combination of the general idea. understanding of the idea or a definition of the term in the proper sense, it is no accident that the very next section of 3 3 Â§10, is concerned with definitions. It is an understanding o f th e simple combination of the general idea that we have when we understand a general idea. I believe these considerations shed light on the best way to read what Locke is saying here in 4 7 .9. It is what a general idea is of that cannot exist, in its generality; only particular ideas and things that answer to the meaning of the general idea can exist. A general triangle cannot exist; annot correspond directly with anything that has what Locke calls real existence. You cannot look at Equicrural, nor Scalenon; bu t all and
167 none of these at once (4.7.9). and inconsistent Idea s are put together (4.7.9). is by the process of Lockean abstraction, resulting in a simple combination Understood in this way, it is clear that Locke is not contradicting himself, though he does certainl y seem to be saying that it is actually the general idea of a triangle which cannot exist. This is particularly odd, given that, as I have pointed out, he does seem to actually think that we can and do have a lot of such general ideas. Here we have to reme mber that Locke is saying that no idea, or actually anything, can exist if it has contradictory features. A mental image of a triangle that is both acute and obtuse is impossible and cannot exist just as a horse that is simultaneously three legged and five legged cannot. Perhaps it is also true that Locke is speaking more loosely than we might like. Said looseness would stem, I believe, from the fact that Locke is engaged in a specific project in 4 7 that of showing that it does not make sense to think our general ideas might be had innately (as summarized above), which is somewhat tangential to his earlier discussions of abstraction as such. Moreover, he has already laid out his theory of abstraction and general ideas, which as I have argued is actually fa irly clear and univocal. Rather than contradicting himself in 4 7 Locke is applying what he has said in 2 and 3 to a particular issue. If commentators have thought he was contradicting himself it is because they have not understood what he said in the fir st place. 54 54 E.J. Lowe also considers the difficulties arising from this passage, in Locke on Human Understanding (New York: Routledge, 1995), 158 161. His discussion is a briefer than mine and differs in the details, but is, I think, compatible with my reading. He does agree with me that the problems commentators have had in this regard are based upon
168 5. 6 Matthew Stuart raises two problems for theory of abstraction that he thinks are significant 55 As a first, over ly simple, statement, they are (1) the idea of color (and by extensi on sound involves him in making conceptual truth clai ms that are implausible at best; and (2) t here are no observable characteristics that characterize all and only things like males, philosophers, heroes, and timepieces. Taking these problems in order, (1) do es seem to be a problem based discussion of color As Stuart says: A general idea performs its function by representing one or more features that the members of a kind have in common, and by not representing the features that distinguish those members from one another. The trouble is that there are cases in which a general term is clearly meaningful, and yet the operation that Locke describes cannot account for the creation of the idea that w ould supply its meaning. 56 And then: some red thing and then isolate just one proper part of it, the part that is also a component of the ideas of other red things. We cannot perform this manoeuvre 55 Stuart actually considers four problems, two of which he takes himself to have dealt with adequately The first is that there are actually two theories in the Essay about which enough has been said in this chapter. (Stuart agrees with me that there is a single theory.) The second is that abstraction to a simple idea just gets you an idea t hat you already role of abstraction may be just to make the simple ideas into general ones. I would add that this latter role is, in part, explicitly what the role of abstraction is, as explained above. However, I think Stuart is wrong that we do not need be the other part of agreed, a conglomeration of simple ideas, but those ideas are not yet the clear simple ideas Locke is talking about. Rather we have confused and indistinc t ideas furnished by experience until we perform abstraction upon our experience to arrive at legitimately simple Lockean ideas. 56 British Journal for the History of Philosophy Vol. 16, No. 3 (2008), 524.
169 in common, because the i parts to isolate. 57 3 4 .16, where Locke also offers something of an ad hoc only the way they get into the Mind (3.4.16). That is, color just signifies (a set of) those things which come to us by way of vision, and sound those things that come by hearing. However, as Stuart White Red and Yellow are all comprehended under the Genus or name Colour it signifies no more, but such Ideas as are produced in the Mind only by the Sight, and have entrance only through the Eyes (3.4.16). colours have in common is just that they are produced in us only by the stimulation of cer tain of colours are produced by the stimulation of our eyes 58 but this seems to be entirely contingent. Stuart offers the example of an individual who se eyes are defective and have never given that person any ideas at all, but who has photo receptive cells in her right palm which do give her a subjective experience indistinguishable from that of normally sighted individuals (say, as from someone observi ng the world by way of a camera held in the right hand). Stuart correctly 59 57 Ibid. 58 S tuart, 525. It is not at all clear to me that Locke is committed, or need be committed, to the conceptual truth of 59 Ibid.
170 Stuart suggests that Locke might respond to this by saying that tho se cells are an eye, but then points out that this would involv e Locke in circular reasoning. of colour is its being produced only by sight; yet he would also be presuming that the criterion for we recognize as ideas of colours. 60 Stuart correctly, I think, insists that it is the systematic phenomenological experience of things that makes the idea of color possible, not the (conting ent) fact of the physical organ that makes that experience possible. Although I think some of these problems can be addressed by emphasizing a difference between vision or sight and whatever sense is dependent upon the eyes However I want to color and sound without the move Locke himself makes in 3 4 .16. Rather than attempt to move from simple ideas of, sa y, red and green to the idea of color which would be problematic for all the reasons Locke and Stuart suggest, we should move as follows: ( a) Ideas of things that are red, green, blue, purple, etc. ( b) The general idea of colored things ( c) The genera l idea of color The moves from (a) to (c) are straightforward applications of the Essay laid out in the previous sections of this chapter. The content of colored things in (b) would seem to be, at least for Locke, those things w hich have the power to produce in us the ideas of the various colors. While color would be whatever those ideas produced in us actually are. This is a 60 Ibid.
171 rather disjunctive simple combination, but it has still been abstracted from our experience in a way that can be seen to be an application of the process explained in Â§4, above. The simple combination theory developed in Chapter 3 helps to keep such disjunctive ideas straight. Note, also, that the set is actually limited by the spectrum of light actually visi ble to our eyes. 61 To get from (b) to (c) we need to abstract away (by selective attention) color from the things that are colored. The process is very similar to the way we get the ideas of red green and blue except in that a different direction, in a manner of speaking, is taken from (a) to (b). Moreover the general ideas of color and the various colors are connected at (a). This last point is also implicit in the (albeit very brief) simple combination of each color. Recall that even simple ideas have (limited) simple combinations. It is not difficult to tell a similar story for sound : ( d) Experiences of particular sounds. ( e) The general idea of sounds ( f) The general idea of sound Here, the meaning of sounds hose events which have the sound Again, this is a disjunctive idea, but abstract and general nevertheless. I cannot say why this approach, from particulars to gener al ideas rather than from simple (general) ideas to the general ideas, does not seem to have occurred to Locke. It is clear that color is problematic, as Stuart says, but it is also clear that Locke did not need that discussion to pro perly ground that idea in the theory of abstraction that he developed in 61 On this reading, al iens might have a slightly larger, smaller, or just different extension of their color position on secondary qualities seems entirely compatible with this result.
172 the Essay Essay has a problem, but that which can and does accommodate the issue. In regard to objection (2), Stuart writes: of abstraction fails to explain our possession of many other general ideas. For him, comp lex general ideas, like complex ideas generally, are ultimately composed of simple ideas acquired through sense and introspection; yet many general terms stand for kinds that cannot be defined in terms of shared clusters of sensible or introspectible prope rties. 62 male but he makes the same objection to timepiece philosopher and hero To follow the line of thought developed in regard to male which is the locus of his exposition, features which serve to distingui sh the males of one species are quite different from those of another species, and it is often even quite difficult to sort male and female within the same species (as in hummingbirds). So, there seems to be no basis in experience (either by observation or f you gather a bunch of philosophers together and examine them you will not find any set of features they all share which is not shared by a number of non philosophers. This just seems false. If by philosopher certain set, or type, of fundamental questions and a concern with the method of approaching est, and concern for the method) which are focused upon through abstraction, with any other characteristics being set aside (e.g. gender, race, economic status, length of hair, level of education). There are features shared by all and only philosophers; wh at those features are exactly depends on what is meant by 62 Ibid., 526.
173 philosopher and any confusion about the meaning of the term is no problem for Locke. Similarly, if timepiece uses on those traits (broadly, the function of the artifact or device, as well as its artificiality) while ignoring such things as size, material composition, etc. Hero is no more problematic; whatever specific meaning one attaches to this idea, one focuse s on those abstraction. So long as we are able to deconstruct (or reverse engineer) each of the terms in the above simple combinations I can think of t hree reasons why Stuart might have thought this was an issue. First, h e might think that Locke did not in tend to include behavior or function (broadly construed) as appropriate traits by which to group kinds, but I can find no reason at all to think this is the case. observable or introspectible am. In fact, his discussion of hummingbird sexes seems to suggest this, as he does not consider features must include how things ar e used and the testimony of those who identify themselves as, for instance, philosophers. Some of the introspectible features must include the contents, or simple combinations, of our ideas. Each of the items on his list of problematic candidates are mixed modes, so they are at least to some extent arbitrary put together by the mind so we must be able to inspect their contents, which seems to count as introspection. Finally Stuart might think that, since there is no consensus on the meanings of philosopher timepiece and hero there is no set of properties accepted as common to all and only members of those kinds, but this is no problem for Locke at all. Ideas for Locke are notoriously private things and he goes on at
174 great length about the problem of diffe ring meanings from person to person, but this does not mean that the general idea of philosopher that a person has cannot be gotten by way of abstraction. As to male I confess I lack a sufficient grasp of comparative anatomy or biology in general to offe r anything like an adequate meaning, however tentative, but it seems likely that some sort of functional definition I have mentioned above is what biologists (or whomever) have in mind and so should be no real problem for Locke. However, even if male turns out to be hopelessly muddled all that would show is that this particular general idea is confused and in need of clarification. Perhaps we have been mistaken in thinking there is just one kind of male, while what we really need is a different taxonomy (e .g. hummingbird male mammal male lizard male and so forth). In any case, it hardly seems that there is any problem here for Locke. difficulties, by directing ou r attention to the need for just such a set of observable properties as Stuart mentions. if the response the Essay actually gives does not work. 5. 7 Three Problems from Lowe recognition individuation ms of resemblance 63 These are potentially significant issues for Locke. I will discuss each in turn. 63 Lowe, 158. Lowe does not list, or deal with, these in the order I have given, but nothing is lost by changing the order of consideration in this case.
175 5. 7.1 Recognition As Lowe points out, and I have mentioned above, Locke holds that all of our ideas, and in fact everything else, are particulars ; there a re no actual universals 64 Instead, we set up particular ideas to stand for, or represent, things that conform to those ideas, and this standing for or representation is what ma kes an idea general for Locke. The basic idea, as discussed in Chapter 4, is that we recognize something as a zebra because it conforms to our zebra idea. Locke refers to our hich particular general ideas functioning rather like patterns in a wallpaper pattern book: to know what sort of thing we are confronted with, we compare it with 65 On the face of it, what Locke says does suggest something just like this. However, as Lowe points out, themselves particulars so how do we recognise them as being of a cer 66 That is, how is it that we are supposed to identify the idea to which we are favorably comparing that thing out in the world (our putative zebra) as one of a zebra, rather than some other thing? We seem to need our abstract general ide as to organize particulars, but those ideas are themselves particulars, so it looks like Locke needs to say that we need general ideas to organize the first set of ideas. This seems to pretty clearly lead to a vicious regress. 64 Ibid., 162. 65 Ibid. 164. 66 Ibid.
176 Lowe offers a response to thi s problem, which seems to me to be on the right track. He suggests that while the pattern book model follows naturally from the text of the Essay 67 Lowe sugg recognize and classify 68 to recognize things we experience. He then offers a second analogy, as a way of r eplacing that of the wallpaper pattern book: the coin acceptance mechanism on a vending machine. Such a further the slots 69 As Lowe puts 70 So, we are able to recognize that striped horse thing ov er there as a zebra because it just matches up, in the right way, with zebra. world as being represented by our general ideas as the result of an automatic process, 71 which process is itself a question of As I have said, this response seems to me to be on the right track. I would only add that the simple combination reading I deve loped in Chapter 3 seems to make the nature of our mental combination of a general idea supplies the criteria 67 Ibid. 68 Ibid. original emphasis. 69 Ibid., 165, original emphasis. 70 Ibid. 71 Ibid.
177 by which we recognize this as a zebra and that as a painted horse. In effect, the simple combinatio encounter in the world. 5. 7.2 Individuation Locke supposes that we classify objects as being of this or that sort or kind by noting their agreement or disagreement with certain abstract general ideas which we have formed through our experience of particular objects. But this presupposes that we can notice particular objects single them out perceptually from other object s altogether independently of being able to recognise them as being objects of any general sort or kind And that seems highly questionable. 72 While we seem to be able to distinguish something as an animal without having an idea of the kind of animal it i s, it seems that we would need to at least have the idea of animal to do so. as a mere something 73 Lowe concludes empiricist thesis, that we have no innate ideas but get them all from experience (often properly modified by the operations of our minds, like abstraction), must be false; some of our ideas must be innate. However, as we have seen in Chapter Essay does give us the material to understand how we get the idea of thing or entity something thing is effectively forced upon us by every interaction with the worl d. This provides a non innatist way of getting our sortal classification systems going. 72 Ibid., 161, original emphasis. 73 Ibid., original emphasis.
178 This reading seems to be consistent with what Locke says and has in mind, and does not strike me as utterly implausible. Granted, going back at least to Kant and includ ing such recent philosophers as Noam Chomsky, thinkers have thought that we must have some kind of innate mental categories to get our coherent thinking going in the first place. I have never found that line of thought particularly persuasive, and given th e argument from Chapter 4, and his general position on innate ideas, it seems that Locke would also remain unconvinced. Rather, Locke must want to say, and we should read him as thinking, that our earliest experience of the world something 74 5. 7.3 Resemblance Â§7.1). This kind of particularism has a potential problem with explaining how we are supposed to apply the same general idea to the wide range of things it is supposed to cover. How do we manage to group all the different instances of red we encounter under the same idea, red ? And, conversely, how is it that we can use the one, particular, idea to represent all of those instances? ivable resemblances 75 to explain how we group things together. 76 77 74 Another move Locke, or a Lockean, might make is to say that the basic distinguishing of things is a natural faculty of the human mind. Much like we do not need to understand the slot mechanism from the previous sub section to use it, we do not need to hav something another thing, we just use our innate faculty to do the base level distinguishing. Of course, Locke does not at all deny that we have such faculties. I find the reading developed in Chapter 4 to be more robust and persuasive, but that does not mean that the faculty reading is not also available. 75 Lowe, 162. 76 Although he raises the problem of how our general ideas represent the various things we take them to represent, L owe does not include that part of the issue when he is considering his problems of resemblance. 77 Ibid., 163. Lowe considers just three of these problems, and seems to suggest this exhausts the set.
179 oblivious to these problems conc erning resemblance, but together they seem to imply that his 78 Let us look at each of the problems in turn. it is always necessary to specify in what respect they do so for instance, in respect of colour or shape or size and this threatens to reintroduce what appears to be talk of universals 79 This is the extent rent universals would seem to be those of color, shape, size, etc. However, if each of those is a properly formed general idea, and I have argued that there is no reason to think that they should not be, it is not clear what the problem is exactly. There m ay be a problem with resemblance as a basis for representation (and I have argued that there clearly is 80 this latter issue is dealt with by the simple combinations of our ideas, which at least implicitly include the resemblance in which respect component(s) of our resemblance groupings. nsistently to never be numerically the same resemblance as any resemblance which that second particular has to a third so that a problem arises as to why we grou p some resemblances a problem which cannot be resolved simply by repeating the original strategy of the resemblance 78 Ibid. 79 Ibid. 80 In Chapter 4 (Â§1).
180 81 First, it is not clear to me, to say the least, how anyone, let alone a resemblance nominalist, could reasonably think that one resemblance is not numerically distinct from another resemblance. Second, what seems to be desired here is definition, or proper simple combination, of resemblance that is not circular. I do not see any reason why resemblance should not be a perfectly legitimate Lockean general idea, arrived at by the process laid out in Â§4, but I do not at present have even the beginning of a simple combination to offer. The most I can say is that if such a n on circular meaning can be presented this problem seems to lose its force. If, however, no such meaning is available, this problem resemblance. The third, and final, pr oblem is that any two things resemble, and do not resemble, each other in an indefinite number of ways, 82 but some of those ways we determine to be trivial and others significant, but there is no explanation of how we come to grasp that significance (or tri innate component in our cognitive 83 First, Locke certainly does not inks that we have innate natural, built in faculties, and one of those might well be involved in the way we assign significance (and triviality) to some resemblances over others. Second, what we count as an appropriate resemblance is closely tied up with t he nature of representation Locke thinks is involved between our ideas and what they represent. Moreover, as suggested in response to the 81 Ibid. 82 163. 83 Ibid.
181 issue depends at least t representational relationship actually depends on resemblance. To this I need only say that I have argued that resemblance simply is not required for representation (though it may be involv ed in some representation). 5. 8 Summation understood as a single unified theory that is the basis of all his general ideas, whether simple ideas or simple and c actually has two kinds of abstraction in mind. I have also argued that it does not require the separation of ideas as Walmsley and others would have us believe, but merely partial conside notoriously difficult section of the Essay is concerned; that it does not run afoul of the p roblems Stuart raises for it; and that it is at least not clear that the problems L owe conside rs are fatal to but I have argued that it fares better than it is often supposed.
182 CHAPTER 6 REAL, ADEQUATE AND TRUE IDEAS, AND PERCEPTUAL ERROR According to Locke, some of our ideas are real, adequate, and true. This seems odd and puzzling to the contemporary ear. By real we normally mean something like actual, as in Pinocchio wanted to become a real, actual boy, not just a puppet copy of one. But then, certainly, all of our ideas must be real: real ideas. Surely Locke is not trying to say something as obvious, and pointless, as that. On the other hand, it seems easier to apply adequate to ideas, but then the question is adequate for what? Finally, we normally think that truth or falsity only applies to propositions, and actually Locke seems to agree with this, so how can it make sense to say that closel y related to real and actual, but is that all Locke is trying to say? If so, this at best only compounds the difficulty with the reality of our ideas and seems to add very little to our understanding. Unfortunately, there is very little in the secondary li terature to help us understand this Lockean notion of real, adequate, and true ideas. In the following section, I will work through what the Essay has to say about real, adequate, and true ideas, trying to understand what it is that Locke is saying. In ea ch of the three cases, I will try to explicate what Locke means by the term and how it is he thinks ideas of various sorts qualify for that description. Then, in sections 2 and 3, I will consider a problem, recently raised by Antonia LoLordo, in regard to perceptual error. The basic issue is that we seem to be all too this. Say that we see a square tower off in the distance, but it is so far away that it appears as a round dot aga inst the horizon or landscape. Locke says that all of our simple ideas are real, adequate and true, and that ideas of figure are simple ideas. So, on the one hand we are getting an idea of roundness from seeing the tower in the distance, but the tower is s quare. It hardly seems
183 to make sense to think that the idea we receive is real, adequate, and true. On the other hand, the Essay we can respond to this issue on Loc interesting, and potentially seriously problematic, issue for Locke becomes clear: some of the ideas he lists as simple are not what most have taken them to be, and this means that it is actually very h 6. 1 Real, Adequate and True Ideas Ideas other Considerations belong to them, in reference to things from whence they are taken, or which they (2.30.1). Before getting to this distinction, it is worthwhile to note that the distinction follows from considerations involving two things : the source of the ideas and what the ideas are of, or what they represent. So, when Locke says that an idea is real, he is saying something about either how we got that idea or what it is an idea of, or both; he is not saying that it is a real idea, as o pposed to an unreal idea. Locke then spells out the threefold distinction, saying an idea may b something of what Locke means by the first distinc tion, but we have nothing useful to add to our initial understanding of adequate and true ideas. Fortunately, Locke offers a chapter on each of these distinctions, and I now turn to considering these in turn. 6. 1.1 Real Ideas (2.30) Locke tells us fairly Nature; such as have a Conformity with the real Being, and Existence of Things, or with their
184 Foundat correspond to the reality of things or to archetypes, and those qual correspond. To get clearer on this, it makes sense to work through what Locke says about various sorts of ideas and how they stand in relation to this particular distinction, but first it is important to get clearer on what Locke means by archetype in this context. The most useful explanation of this term happens to appear in the set of chapters currently under consideration. In the ch apter on the adequacy of our ideas, Locke says: Our complex Ideas of Modes being voluntary Collections of simple Ideas which the Mind puts together, without reference to any real Archetypes, or standing Patterns, existing any where are and cannot but b e adequate Ideas Because they not being intended for Copies of Things really existing, but for Archetypes made by the Mind, to rank and denominate Things by (2.31.3). se can be pains, here and particularly in Book 3, to emphasize that the latter set, which includes most notably all our mixed mode ideas, are the workmansh ip of the understanding, put together by us for our use in organizing our experience of the world. As he says in the foregoing passage, these need to organize. L standards against which individual things are compared to assess whether or not they should be classed together. For instance, we compare particular acts of people against the arche type we have to see if those acts count as adultery ; and, in this case, the archetype is just the mixed mode idea we have of adultery (3.5.3 16). Moreover, in Book 4, Locke stresses the importance of our
185 ave knowledge (4.4.8). As he says in 4.5, the our ideas must appropriately, or adequately, represent that to which they refer. (I will return to this adequac y in the next sub section.) So, for Locke, an archetype is something that serves as a standard against which things are compared. The archetypes of our ideas, by being the referents of those ideas, set the standard for those ideas, establishing the criteri a and limits of what falls under those ideas. These archetypes are things in the world, particularly for our ideas of substances, or the ideas themselves, particularly for our ideas of mixed modes. As we will see, this distinction carries a lot of weight i simple Ideas are all real that this does not mean that they are images, 1 What makes it the case that all of these simple ideas conform to the reality of things outside us is tha Ideas are all real and true, because they answer and agree to those Powers of Things, which produce them in our Minds, that being the passivity of our reception of simple ideas (which I have emphasized in Chapters 3 and 4). We cannot create our own simpl e ideas, but are restricted to only those given to us by experience, 1 the ideas that are not images do not represent things that exist, because, as discussed in Chapter 4, Locke seems clearly committed to the fact that they do so represent.
186 which reinforces the thesis that those ideas must conform sufficiently to the reality of things (at nothing better we could possibly do, so simple ideas are as real (in the current sense) as ideas can get. This is not to say that all of our compl ex ideas are therefore fantastical. Mixed modes and Ideas to make them real (2.30.4 ). As mentioned above, these ideas are archetypes that we use to organize our experience of the world. They are not supposed to capture the way the world is, per se, but we look to see if the world matches up with our idea. We do not, according to Locke, p rimarily get the idea of, say, justice from experience from the world. Instead, we put together various ideas to form this mixed mode idea and then look at things in the world to see if they fall under it. As Locke says later in the Essay : To know whether his Idea of Adultery or Incest be right, will a Man seek it any where amongst Things existing? Or is it true, because any one has been Witness to such an Action? No: but it suffices here, that Men have put together such a Collection into one complex Idea that makes the Archetype and specifick Idea whether ever any such Action were committed in rerum natura or no (3.5.3). So, it does not really make sense to worry whether or not our ideas of adultery incest obligation drunkenness a nd liberty conform to the world. While we often ask whether the world conforms to our mixed modes and relations, those ideas, as archetypes, are real just so long as they conform to themselves, which of course they cannot help but do. Finally, complex ide as of substances (as in kinds of substances, not the idea of substance as substratum which Locke leaves unmentioned in these chapters), are real to the extent that
187 Ideas as are really united, and co exist in Things w ithout 2 man for instance, can be said to be a real idea if the ideas we have put together to come up with this mixed mode are those that actually conform to the aspects of men that actually do go together in the external wo rld and produce those simple ideas in us. However, once we add anything to that complex idea, like, say, the idea of horns the idea becomes fantastical. Of course, as Locke acknowledges, such a horned man is possible, so this particular complex idea is le both winged and wingless (2.30.5). Here it is worthwhile to consider what Locke says about real and nominal essences. He the inner constitution of thing s that explains the various properties and qualities of them; it is what makes it the case that these qualities always come together, for instance. It is an understanding of what this is for substances in the world that we lack, as Locke repeatedly reminds us. Nominal essences, on the other hand, are the complex ideas by which we organize things in the world. Since we cannot ever grasp the real essences of substances, we organize those substances by their nominal essences their genus and species for instan ce, or by what is simple combinations This is a necessary and understandable thing to do given our practical needs and ignorance of the relevant real essences, but we go astray when we confuse nominal essences for real, or if we pre sume to mistake the clarity of our nominal 2 Here Locke seems to be talking loosel y, as he surely does not mean to say that the ideas co exist in things outside our minds.
188 essences for a true understanding of real essences (3.3.15). If we are not considering ideas of substances, though, but mixed modes, the essences involved are essences of ideas, and then the nominal essences are t he real essences. The real essence of, say, liberty just is what is included in that mixed mode, its nominal essence. So, to tie this distinction in to the current discussion, if we are considering a nominal essence of a substance, it may very well conform to what actually exists in the world. However, we can have no assurance at all that our ideas conform to the real essences of such things. Ideas of mixed modes and relations, however, just are nominal essences and there is no related problem for them. Fr reality is that it properly conforms to the appropriate object, which is either the way things are in the world or the ideational archetype in our minds. If an idea confor ms to something that really exists in the world or is a mental archetype, the idea is real. Simple ideas, which we get passively from experience, conform to that which causes us to have them, and are all thereby real. Complex ideas may or may not be real. In the case of mixed modes and relations, these conform appropriately to themselves. That is, they are the things to which other, external things conform, and the mixed modes and relations are themselves inherently real. Finally, ideas of substance kinds conform to the kinds of substances actually or possibly present in the world. This is the kind can be real, if it does conform to actual kinds of things in the wo rld, barely imaginary, if it conforms to a kind of thing that could possibly exist, and entirely fantastical, if it conforms to something that could not actually exist. If the substance idea, however, refers to the real essence, rather than the nominal, we have no way of knowing that our idea really conforms with reality at all.
189 The notion of conformity that Locke has in mind is, however, a bit unclear, to say the least. As we will see, conformity also plays a key role in the truth of ideas, and I will dis cuss it further in Â§1.3. 6. 1.2 Adequate Ideas (2 31 ) which the Mind supposes them taken from; which it intends them to stand for, and to which it lete representation ground at all; either an idea is adequate or it is inadequate. The modifier doing the heavy lifting lation of adequacy. What does it mean for an idea to perfectly represent its archetype, though? From the foregoing it can be assumed that Locke means that such a representation is complete, but this just leads to the question, what does it mean for an idea to completely represent its archetype? all our simple Ideas are adequate fitted and ordained by GOD, t o produce such Sensations in us, they cannot but be correspondent, it seems that the simplicity of the ideas, and our passivity in receiving them, ensures tha t they one uniform Appearance or Conception in the mind, and is not distinguishable into different Ideas can hardly be partial or incomplete. To thi s is added the fact that a simple idea just is the idea we get from its cause. Its adequate correspondence is, effectively, a given. Whatever simple ideas you get from experiencing, say, a pineapple, just are what it is that you get, and they
190 are automatic ally the right, and adequate, ideas. For Locke this is just a matter of entailment. So, all our simple ideas are adequate. 3 Our ideas of mixed modes and relations are also all adequate, and Locke again employs the notion of completeness to secure this ade Things really existing, but for Archetypes made by the Mind, to rank and denominate Things by, essences, both no minal and real. These ideas are combinations of ideas we have put together to answer our needs, and as such include everything we need to use them appropriately; they are inherently adequate in the sense Locke has in mind. Moreover, as simple ideas are cau sed by measure up to the standards set by the appropriate authori ty, God or the human mind, triangle Idea of three Sides, and three Angles: in which is contained all that is, or can be essential to it, or necessary to complete it, where Locke was apparently presuming, the idea of plane figure there is nothing more to be added to, or wanted from, the idea of triangle That idea is adequate in that it perfectly represent s the thing it stands for, which is the archetype idea itself, as the mind intended. This is not to say, however, 3 In this connection, in regard to what it is that our secondary qualities are of, Locke makes a surprising claim. Because our ideas of secondary qualities are merely the result of our senses being appropriately affected by the primary qualities of external things, Locke concludes that there really would not be anything that is, say, orange or sour if there were no one with the proper senses. However, Locke extends thi s thinking to light and heat. He says those Organs to receive the Ideas of Light and Heat, by the those impressions from the Fire, or the Sun, there would yet be no more Light, or Heat in the World, than there would be Pain if there were no sensible Creature to feel it, though the Sun should continue just as it is now, and Mou nt tna Essay was written before the notions of light as wave particle and heat as the mean of m olecular motion were accepted. Locke is (at least probably) mistaken here, but in this matter so were all the best minds of his time.
19 1 that what is included in the idea is all there is to be known about triangles. Locke emphasizes a all the Properties of that one, no very compound Figure, a Triangle though it be no small numbers, that are already by Mathematicians all of their prope rties, but just what is needed to make them adequate to represent whatever it is they stand for, which is themselves, as the archetypes against which we compare things we encounter. In this sense, all of our complex ideas of modes and relations are adequat e. 4 On the other hand, ideas of substances are all inadequate. Locke says that these ideas of the substance kind man is intended to refer to something that is essential to that type, the essence of man However, Locke argues that we do not have any adequate idea of such essences at all. We have an assortment of ideas that go together under the i dea of man but we have no understanding of anything that makes their going together necessary, as the notion of essence demands. In reference to gold, Idea of its Essence, which is the cause that it has that particular shini an essential cause of the coexistence of the various properties o f the substance in question they cannot be adequate; they do not perfectly represent what they are supposed to represent by the mind. Since all of these ideas lack this idea, all of them are inadequate (2.31.7). 4 the rel ationship may very well be rather seriously in adequate. The more complex the idea in question, the more something rather different by courage than I do. Locke points out, though, that this is more a problem with communication than with thinking or ideas themselves (2.31.4 5).
192 In the second case, sometimes our ideas of Pictures and Representations in the Mind, of Things that do exist, by Ideas of those qualities that man is only supposed to include the qualities, or p roperties, of men, or humans, that we can see and of which we can, presumably, have adequate ideas, but does not include the notion of an essence of man It is, effectively, only intended to be that of the nominal essence. However, those who think of subst ances in this way, arrive not at perfectly adequate Ideas Ideas of Substances, do not contain in them all the simple Ideas we gro up together a relatively small number of the relevant ideas, which is considerably short of the kind of completeness Locke requires for adequacy. Moreover, we can never be sure that we have adequately understood the relations that undergird our ideas of th e secondary qualities of the things we are trying to group under a substance kind idea (2.31.8 10). all our complex Ideas of Substances are imperfect and inadequate hink we are including an idea of an essential quality of a kind in our idea of its substance, we are sure to be mistaken because no such idea is available. If we do not include such an idea we can never be sure we have included all of the required ideas. Therefore, according to Locke, while our simple ideas and complex ideas of modes and relations are all adequate, our ideas of substances are necessarily inadequate.
193 From the foregoing, it is clear what Locke thinks it means for an idea to completely repr all that is, or can be essential to it, or necessary to complete it, where left out of the idea, because it is already simple. Comp lex ideas of modes and relations are their own archetypes and inherently include all the things we have or have not included in them, so there really is no sense to their being incomplete. 5 Ideas of substances, though, can never be sure to be complete repr esentations of the things they are supposed to represent. Finally, each type of idea has a kind of authority over it which Locke seems to think sets the standard for the perfect adequacy of that type of idea. The relation between simple ideas and their c auses is set up by God, and the relation between the other sorts of ideas and what they are thought to represent (mental archetypes or things in themselves) is set up by the understanding, successfully in the case of mixed modes and relations, and unsucces sfully in the case of substances. 6. 1.3 True Ideas ( 2 32 ) Locke opens this chapter of the Essay by conceding just the point that was made earlier: strictly speaking, truth or falsity only applies to propositions, not ideas. 6 However, he thinks Ideas themselves are termed true or false, there is still some secret or tacit It is the supposition that is really true or false. So, if I have an idea of an elephant, one could say 5 That is, unless we are trying to compare ideas across minds, so to speak. 6 firm, and he repeats this point so frequently in this chapter that one might well wonder why he did not make things clearer when he introduced the three fold distinction in the first place.
194 7 is true. This leaves the question of ho w to understand the conformity involved. Locke gives three sorts of cases that he seems to think will illustrate what he has in mind: when we think our ideas conform to the ideas in the minds of others which have the same names (as in when I think you and I share the same notion of justice ); when we think our ideas conform to something that really Ideas of a man, and a Centaur, supposed to be the Ideas of real Substances, are the one true and the other false refers any of its Ideas to that real Constitution, and Essence of any thing, whereon all of its Properties not all our Ideas of Subst supposedly, that something is dependent on the context of usage. While each of these uses of tru e or false ideas still sounds odd, the conformity involved is still unclear, and one would wish that Locke gave further clarification, he instead moves on to a discussion of the truth or falsity of our abstract complex ideas. Locke says that the suppositi on that our ideas are true (that is that they appropriately Ideas so, Locke explains, because of our concern for having knowledge about the world. He reviews his pos ition on abstraction and explains that this is necessary in organizing our thinking and engaging with the world (2.32.6 something in the Mind between the thing that exists, and the Name that is given 7 Here I am assuming that my idea was intended to represent a par ticular elephant and was not the general idea elephant
195 such we are concerned that there is a conformity both between the ideas and the way things are in the world and between the ideas and the way we use their names in communication with each other (2.32.8). Locke then discusses each part of th double Conformity In regard to the conformity involved between our ideas and those of others, as mediated by the use of the same names for those ideas, Locke observes that any of those ideas may be simple Ideas are least of all liable to be so mistaken is easy enough to verify the referents of each of these names in the minds of other people, and we are rather unlikely to get sufficiently confused about such ideas as to result in our mai ntaining Complex Ideas are much more liable to be false in this respect ; and the complex Ideas of mixed Modes, much more than those of Substances This is because our ideas of substance kinds are all inten ded to refer to things in the world. We have at least a way of pointing out the kinds of things we mean by our words, and the sensible qualities of those things place some reasonable boundaries on our ideas of substance kinds. This is not the case with ide Justice or Gratitude or Glory constituents of these ideas at our own discretion, and they are intended to be their own archetypes, so there really is no external standard against which to compare our various, po tentially conflicting ideas. This is why we run into such difficulties trying to decide whether someone has a true or false idea of, say, justice (2.32.12). real Existence of Ideas and must be suitable to those Powers, he has placed in external Objects, or else they could not be produced in us: And thus answering those Powers, they are what they should be, true Ideas
196 (2.32.14). Locke considers two objections to this position, 8 but the foregoing is his positive reasoning for the truth of simple ideas. Essentially, all there is to our simple ideas is the 9 primary qualities and our sensory apparatus, so there is no room for error to creep in. Note that this does not really help with our understanding of conformity, but only tells us that our simple ideas must conform appropriately to their objects; there is no way they could do otherwise. N either can our complex Ideas of Modes, in reference to the Essence of any Thing really existing, be false by the mind; they cannot fail to have the constituent ideas that they happen to have, and they are the standard of representation, so that relation must be adequate. Again, this does not really explain the conformity Locke has in mind, because it seems that this class of ideas must conform, because they conform to themselves However, when we come to ideas of substances (or substance kinds), these ideas may false when looked upon as the 8 The objections Locke considers: that some might confuse the ideas with things that are in the objects themselves (2.32.14); and that it might be the case that different people have different phenome nal experiences and thus ideas that differ in content when presented with the same object (2.32.15). In regard to the latter, according to Locke, for my idea of green to be true it only has to be such as is produced in me by my looking at green things, or what I phenomenal experience is like, or even my own phenomenal experience. Green just means the idea produced in me by green external obje green will be true things, and it cannot help but do that (2.32. 14 15). 9 my own position, that he is thinking of all simple ideas imagistically. However, I think there are two reasons to think Locke is spe aking loosely here. First, there are good reasons (discussed in previous chapters, particularly 2) to think that this is not a plausible position and that Locke should not be accepting that commitment. Second, in this part of the Essay (2.32.14 16, at leas t), Locke is generally speaking a bit less carefully than he does in other parts of the Essay For instance, in this chapter he also does not consider any distinction between ideas of primary qualities and ideas of secondary. Resemblance, or the lack there of, does not seem to be on his mind, so, I suggest, he slips into just thinking about ideas which are plausibly thought of imagistically. This makes sense, as he is concerned here to deal with objections having to do with phenomenal experiences of objects (see above).
197 unknown essence that somehow makes it necessary that the other qualities grouped together under a substance kind idea always come together is unavaila ble to us. We can have no such idea. I think it makes sense for Locke to say that where we cannot have an idea we cannot have an idea that conforms to things as they are in the world. 10 On the other hand, when we mean just Ideas i n the Mind, taken from Combinations of simple Ideas existing substance will be false either when the ideas grouped together do not actually appear together in n ature (as in centaurs and unicorns), or when, by intention, the idea leaves out something that is always conjoined with the other (retained) elements of the substance kind (as in the idea of a .32.18). However, Locke says that if the latter leaving Locke offers the following as something of a summary: Any Idea then which we have in our Minds, whether conformable, or not, to the existence of Things, or to any Ideas in the Minds of other Men, cannot properly for this alone be called false For these Representations, if they have nothing in them, but what is really existing in Things witho ut, cannot be thought false being exact representations of something: nor yet if they have any thing in them, differing from the reality of Things, can they properly be said to be false Representations, or Ideas of Things, they do not represent (2.32.20). 10 That is, unless there is no such thing out there in the world. It seems to make sense to think that our lack of an idea does conform with the reality of things if there is no such thing. However, Locke seems to think there must be some kind of unknown essence that ties various qualities and properties together as substance kinds, but has ruled any further, more substantive, considerations of the matter out of bounds. As a result, he is left with the claim that any idea of such an unknown esse nce of a substance kind is chimerical.
198 Rather, the mistake, which results in this sort of falsehood, is when we make a judgment that our a collection of things that never exist in reality, when w e deliberately leave out things that do co exist in reality, or when we think that our ideas of substance kinds contain some real, hidden essence (2.32.21 24). These judgments, all fitting into the form my idea conforms to this other thing are capable of being true or false. 6. 1.4 Conformity, and the Real, Adequate, and True ideas, it is necessary to try to come to some understanding of what Locke means by conformity I quoted the above passage at such length because it seems to give us the clearest advance on this issue. It will be noted that, according to what has been covered above, both the reality of our ideas and their truth depend very much on this notion of co nformity. Indeed it seems that the notions of real ideas and true ideas are co extensive, as long as we recall the qualifiers upon our 11 and we set aside ideas of substance for the moment. It is only in the above pas sage, though, that Locke links conformity with representation in a clear way, and this link happens to also bring in the adequacy of our ideas. As we have seen, our ideas are false, and also fantastical, if they lack conformity to the way things are in the world (or in conformity involves what it is that is included, or not, in our ideas, which are representatives of the things to which we want them to conform. Th erefore, it is no surprise that simple ideas are both real and true, because there is nothing that could be left out or added to them. Moreover, 11 In the rest of this chapter I will proceed with these qualifiers (i.e., primarily, that the truth or falsity only strictly applies to a proposition affirming a conformity of the appropriate sort) in mind, only menti oning it if and when it seems important to emphasize this distinction.
199 since adequacy was explained as our ideas including all of the necessary components (all of the simpler ideas), it is now clear why it should be the case that the reality, adequacy, and truth of our ideas should naturally and generally work together. Locke is effectively considering the very same aspect of our ideas from slightly different perspectives. The aspect of our ideas Locke is considering is the relation between an idea and what it represents. The reality of our ideas depends on whether or not those ideas conform to something that is real (or, a bit more clearly, properly represent something that is real), whether that something is an external existent, an internal experience, or an idea that serves as an archetype. x conforms to the appropriate referent y e. In either case, the issue depends upon the adequacy of our ideas whether or not the idea is complete enough to sufficiently represent its sort of thing is as a representative of that object. Simple ideas are all real, adequate, a nd true, because they just are the basic ideational elements given to us by experience, as so ordained by our creator. They could not be other than they are, there is nothing to be left out of them or added to them (without making them something other than simple ideas), and they automatically conform to that which they represent, the experience we have of their objects. Mixed modes and relations are all real, adequate, and true (so long as we are not considering their conformity with the ideas in other m inds), because they are created by the mind as archetypes against which we compare reality. They are inherently complete, since they are
200 their own archetypes (each is a complete idea of itself, so to speak), are as the mind intended, and, so, automatically conform to what they represent (themselves). Ideas of substance, however, are problematic. Such ideas that are supposed to include the unknown essence of the substance kind are not real, adequate, or true, because that idea of essence is unavailable to us. Substance ideas that are merely collections of the ideas that we experience as always going together can be real, adequate, and true, if we are careful to include everything that we should in the ideas. It is, however, very hard to see how we could eve r be sure that we have met this demanding standard, as Locke acknowledges (2.31.13). However, it is possible for a substance idea to be real, because it refers to a real, existing substance in the world, but not adequate, because our idea is (perhaps neces sarily) incomplete. The inadequacy of the idea then seems to mean that the idea cannot be true. So, we can see that where the reality, adequacy, and truth of an idea come apart, it is the inadequacy of the idea that explains that coming apart, and Locke se ems to think that this will only happen with our ideas of substance. Finally, if we are concerned with the correspondence of our ideas with those of other th at my idea of x match up. At the very least, for all practical purposes, we have very good reason to think that these ideas sufficient conform to each other. With, often grea t, patience and care, it is at least possible that we can become confident that our ideas of mixed modes and relations correspond to those of others. 12 Locke seems to think this standard is hardest to meet in regard to this class of 12 The amount of work that would be required to be sufficiently confident would, it seems, depend largely on the context involved and the complexity of the ideas involved. Arguably, math ematical and geometrical ideas are associated with a higher level of confidence (of the relative truth of shared ideas) than those involved in, say, political discourse.
201 ideas, because there is no external source for these ideas. On the other hand, as long as we are not including the unknown and unknowable idea of substance as substratum real essence, we can more easily be sure of the truth of our ideas of substance, when we a re concerned with our ideas corresponding to those of others, because we are able to work with the existing substances to which we are communally referring. Of course, this is not to say that we can suppose either ourselves or others to have adequate ideas of these substances. So, again, it seems that ideas of substance cannot be thought to track reality, adequacy, and truth in anything as simple a way as simple ideas and ideas of modes and relations. With this understanding of what Locke means when he spe aks of real, true, and adequate 6. 2 The Problem of Perceptual Error In a fairly recent paper, Antonia LoLordo raises what seems to be a substantial problem for Locke. 13 On t he one hand, it seems clear that we are capable of perceptual error, for instance account of perceptual knowledge apparently makes no allowance for such error. In the Essay Locke includes figure as one of our simple ideas ( at 2 5 ), and as we have seen says that simple ideas are all real, adequate, and true. So it would seem that although we are clearly capable of such error, we are simultaneously, according to the Essay not capable of it. This is indeed a serious worry for Locke (or a Lockean position); as ground all knowledge and probability in the ideas received through sense should, surely, be 14 Locke is committed to t he 13 Philosophy and Pheno menological Research Vol. 77, No. 3 (November, 2008), 705 724 14 Ibid. 707 708.
202 empiricist thesis that all of the things we can and do think about (our ideas) and thereby any knowledge that we might have is ultimately derived from experience. According to Locke, that experience is given to us by ways of simple ideas received from i nteracting with the world and our own mental operations. So, if there is a problem with how Locke deals with the possibility of the first set of those ideas being mistaken, it would seem that we have good reason to have doubts about his project at the outs et. To make the problem clearer, I will consider the two opposing points. To take the second point first, we have seen in the previous section that Locke takes all simple ideas (and several other sorts) to be real, adequate, and true, where this means t hat the simple ideas we get from our experience all conform to what they represent. So, when we receive a simple idea from viewing a tower in the distance that idea should conform to the relevant aspect of that tower. At least, that would seem to be what i However, to consider the first point, it seems undeniable that we can experience perceptual error, as in the round seeming square tower in the distance, and Locke must have been aware of this fact. At the very least, it seems possibility. I t hardly seems plausible to think Locke was unaware of perceptual error. The example of a square tower that looks round from a sufficient distance is a traditional case of perceptual error. This is a well known issue, both among philosophers and in everyday life. Sextus Empiricus discusses a variety of such examples in his Outlines of Scepticism including the square tower in three places, 15 and Descartes mentions this kind of error in his First M editation. While it is tricky business to assert with any measure of confidence exactly what 15 Sextus Empiricus, Outlines of Skepticism edited and translated by Julia Annas and Jonathan Barnes (Cambridge: Cambridge University Press, 2000). See particularly I 32 (page 11), I 118 (31), and II 55 (81)
203 Locke had read or was even aware of, there can be no doubt that cases of perceptual error were between primary and secondary qualities seems to suggest that he was alive to the kinds of issues that might, and often do, lead to cases of perceptual error. So, it seems a bit unlikely that Locke was not aware of this issue and therefore, it seems that we cannot simply disregard the conflict LoLordo points out. Additionally, and I think more importantly, while it would surely be understandable that Locke did not offer a response to a problem of which he was unaware, this hardly seems to be a robust res cannot be accommodated by the theory presented in the Essay then there seems to clearly be a problem for the account. If we grant, as it seems we ought, that such cases do in fact occur, That last point is worth brief consider ation Locke might not owe us a response to the problem of perceptual error if the problem itself is more or less ir relevant. However, as mentioned above the importance of the reality, adequacy, and truth of our simple ideas 16 is critical to what Locke wants to say about the possibility of our having adequate knowledge in the possibility of perceptual error is anything but irrelevant. So, it seems that Lock e, or at least his theory, owe s us some kind of response to the problem of perceptual error. However, following an Epicurean tactic, one could deny that a putative perc eptual error is about our ideas. The basic Epicurean point, revived by Gassendi, is that our senses cannot deceive us. The appearance of the tower is round; it is only our judgment 16 at this point.
204 that the tower itself is round that is in error. The tower does look round at this distance, so our If this were true, then characterizing perceptual error as a problem with the ideas we have (particularly the simple ones) would be a mis take. T his does not appear to be a Lockean option. What is at issue is not the proposition that the tower is round, but the idea of roundness we get from our particular experience of the tower. While it seems likely that if there is a problem with our si mple ideas there will be a problem with our propositions involving those ideas (and, plausibly, our less simple ideas), the issue raised by the argument happens at the more basic level of those simple ideas. According to the explicit theory of the Essay introduced above, it seems that we should not be able to get an idea of roundness from an experience of a tower that is actually square. Perhaps one might say that the error in judgment lies in thinking the idea is of roundness but this hardly seems to make sense since the whole point is that the visual experience is of something round. These considerations show that the Epicurean move is not available to Locke. However, as we have seen, Locke does explicitly say that there is a tacit judgment made in r egard to the truth of our simple ideas. This judgment involves the proposition, for example seem that we might easily say this judgment is in error. Unfortunat ely, as we have also seen, Locke has also explicitly said that all of such judgments involving simple ideas are true. So, on the face of things, it seems that we cannot sidestep this issue by resorting to blaming our judgments for the error. 6. 3 Responding to the P roblem There are, I think, three ways to respond to the foregoing problem of perceptual error worth considering: deny that the idea of the shape of the tower is a simple one; argue that,
205 actually, even that round appearing idea we get from viewin g the distant tower is appropriately real, adequate, and true; and employ the distinction between imagistic and non imagistic ideas. I think each of these moves has something to recommend it. I will discuss each in turn. 6. 3.1 The Round Idea is not Simple It seems clear to me that an idea of a shape can hardly qualify as appropriately simple enough to count as a simple idea. To put it bluntly, shapes clearly have parts, so they cannot be simple in the sense that Locke needs (for them to be the basic consti tue nts of our ideational system). As mentioned earlier, thus] contains in it nothing but one uniform Appearance, or Conception in the mind, and is not disti nguishable into different Ideas (2.2.1). I take it to be plausible that the idea of a circle can be distinguished into ideas including curved line interior. The general idea of circle on the other hand, I take to include ideas like plane figure In either case, the putative simple ide a is bodies Bulk Figure Number Situation and Motion or Rest of their solid parts (2.8.23). I submit that in this context figure is not shape, but the quality of having a shape. Figure language in 2.5 is similar. Here, as the kinds of ideas we get by mo re than one sense we have Space or Extension Figure Rest and Motion (2.5). forward. If Locke means that our idea that something in the world has a shape is a s imple idea that is one thing. If, on the other hand, he means that the idea of figure includes the actual shape (or figuration), then Locke seems to be clearly mistaken. It might be plausible that some very figure produces a simple idea in us no matter how complicated that shape is is wildly implausible. With the possible exception of rest this
206 carries over to each of the other ideas in the list from 2.5 It is reasonable for Locke to think that the motion o f x prompts a simple idea of motion in us, but it is not reasonable to think that the idea of is a simple idea no matter how complicated that motion turns out to be. In regard to number it makes best sense to think that this means something akin to there are a number of this thing where that means there are at least one. For Locke, unity and unit (or unite s of unit (2.16.1 2) so Locke is explicitly committed to denying that the number of items perceived is a simple idea, at least if there is some number other than one. He certainly cannot be claiming that the number is a simple idea no matter what that num ber is. I t seems to me that the way Locke treats his discussion of these ideas in the rest of Book II strongly supports this reading. In fact, as I said above, it seems clear that even the simplest shapes should not be considered simple, even though they depend upon primary qualities of things. That truth depends upon a upon the appropriate completeness of the idea in representing its object. Simple ideas are real, adequate, and true largely because of their simplicity; they are inherently complete rep resentations because there could be nothing left out of a simple idea. In the case of the square tower, the idea received clearly does have something left out the corners of the square shape. This all suggests that, if we consider the possibility that Lock e did not think of a particular shape
207 as a simple idea, the problem LoLordo is raising actually supports thinking that those ideas must not be Lockean simple ideas. However, LoLordo raises three considerations against the move to deny the simplicity of id eas of shapes. First, perceptual error can apply to other cases, not just simple ideas gotten by sight, so it would seem that the above approach would have to be generalized to cover an array pparent perceptual error can occur 17 This would suggest that real, adequate, and true However, if we recall tha t simple ideas of secondary qualities are only supposed to represent the experience we have of them, and not what is in bodies themselves, it is not clear that this is a substantive problem for Locke 18 I have just dealt with putative perceptual errors rega rding the primary quality of figure. LoLordo does not actually give further examples of other kinds of perceptual error, and it seems that all the standard examples fall to either the distinction between primary and secondary qualities or the kind of reaso ning I have just developed. That is, is not a simple idea. 19 However, it is worthwhile to pause and consider other possible cases. One might look in the dis tance and mistakenly think there are two towers when there is only one, or think there is 17 LoLordo, 723. 18 For more on this, see Pacific Philosophical Quarterly 91 (2010), 554 586. 19 Additionally, it is important to recall that in Chapter 4 (Â§1) serious doubts were raised about the disti nction Locke makes between ideas of primary and secondary qualities. If what I said in that earlier discussion has merit, then ideas of primary qualities are really much more like those of secondary qualities in that they too more closely resemble the expe rience we have of those qualities than they resemble the qualities in the bodies themselves. This seems to diminish the strength of the problem being raised.
208 just one when there are two (perhaps because perspective and distance blends them into one). However, as mentioned above, the actual number of towers is not automa tically a simple idea. Rather, being a number seems to be what Locke had in mind. Now, one or, more accurately unite is a simple idea, but it seems that we do get that idea from two or more towers in the thinking there are a number of somethings when there are not. Take the typica l case of a mirage: an appearance of something very much like water just below the horizon, given the appropriate atmospheric conditions and so forth. Here it seems that we see something from which we get mistaken simple ideas of both number and existence even on my reading. There is not even one really existing watery thing out there to perceive. However, there is! There is a shimmering, watery looking illusion out there just below the horizon; there is one of them and it really exists. In this case we ca n, I think, invoke the distinction between ideas and judgments that was not available in Â§2. Our idea of a shimmering, watery something is real, adequate, and true, but our judgment that that something is water is too hasty. As will become clear at the b eginning of Â§3.3 I do think there is at least one case of perceptual error that seems immune to the kind of moves I making here. What I take to be the point of the foregoing considerations though, is that it is not as easy as LoLordo suggests to generali econdary qualities as well as what the former actually come to. In any case, we must distinguish between actual cases of perceptual error that involve actual simple ideas and those that do not, for it is only the former cases that are a potential problem f or Locke.
209 against the move to deny the simplicity of ideas of shapes is that the examples the Essay gives of real, adequate and true simple ideas would no longer be good examples. 20 But it actually seems that Locke has given us good examples of what he means by real, adequate and true simple ideas. His examples of those simple ideas in 2 30 32 are of secondary qualities whiteness, swe etness, cold, pain, et cetera and there is no problem there. It seems that Locke thought there was no related worry about simple ideas of primary qualities because those ideas appropriately resemble the qualities possessed by things. There is nothing wrong with his examples of ideas of secondary qualities, an d the list of primary qualities bulk, fi gure, motion, et cetera seems to be fine as far as it goes, even if it might not be what we were expecting. 21 So, it seems reasonable to think that Locke did not intend this to be a general principle and we have no reason to think that this is a primary qualities that we receive fro m more than one sense are sometimes (or often) shaped by our judgment. What a blind man has learned by touch to be a sphere later, when he has become sighted, now appears to be a circle variously shaded. The formerly blind man is able to reconcile the app arent discrepancy through the application of experiment and judgment (2.9.8) This all visual ideas of size and shape that normal adult human beings possess are altered by j udgment in this 20 LoLordo, 723 21 Ibid.
210 22 One might even generalize this over other sense modalities so that any ideas that can run afoul of the kind of error we have been considering are properly und erstood as non that the response to the Molyneux problem seems to only apply to ideas of primary qualities capable of being gotten by multiple sense modalities. It does not seem to help at all with ideas of secondary qua lities only received by one such modality, like those of taste or color. My initial response to this is to insist that the ideas considered above, and in the Molyneux case, are not simple, so it is not clear that this response even counts as a legitimate consideration against the move to deny the simplicity of ideas of shapes. That is, the Molyneux case does not involve simple ideas, so the fact that Locke did not offer further Molyneux style responses is not to the point. Moreover, while it is clear that Locke does not give us anything analogous to the Molyneux solution for tastes and colors, it is not at all clear why this would be necessary. Locke already has available his distinction between primary and secondary qualities. The problem Molyneux posed w as intended as a puzzle about primary qualities, after all, and the Essay ha s the resources to deal with potential cases of perceptual error in regard to ideas of secondary qualities. So, I do not think LoLordo has given us good enough reasons to resist the move to deny that the idea of shape we get from seeing a square tower at a distance is a simple idea. However, which I will return in Â§4. 22 LoLordo, 723.
211 6. 3.2 The Round Ide a is Real, Adequate, and True If the idea we receive from the distant square tower is not simple, what is it? It seems plain that it is a mixed mode. An idea of a circle combines the ideas of curve plane figure extension etc. Recalling the discussion in Â§1, though, suggests that the idea must after all be real, adequate, and true, at least as far as considering the idea itself is concerned. The idea of round shape is an archetype against which we compare things we enco unter, Ideas to make them real The idea of roundness is also complete (what could po ssibly be missing from it?), so it is also adequate. Finally, the idea itself, as an archetype, is inherently true; it conforms to itself. Only the supposition that the idea we receive conforms to the actual shape of the tower is false, but this is the o nly one of the ways that the case offered by LoLordo actually seems to be discussed above in Â§1.3. In fact, given the detail Locke went into in 2.30 32, this response seems perfectly generalizable over all mixed mode and relational ideas. Unless we can find an actual problem covered. There is such a case, but there is also another response. 6. 3.3 The Distinction between Imagistic and Non Imagistic Ideas First, the problem case: imagine you see a stationary tower in the distance, but due to some kind of atmospheric disturbance the tower seems to be moving a bit. Here we hav e a case of perceptual error that involves a clearly simple idea of a primary quality on the reading I offered above (Â§3.1) We are receiving the idea of motion from a motionless object. This is not the idea of some particular motion(s) but being in motion or having motion So, with this example we seem to be right back in the quandary with which we began finding a way out of
212 the original inconsistency between perceptual error and what Locke says In this case, though, because it involves an actually simple idea of a primary quality, we cannot sidestep the difficulty by way of the considerations of the foregoing sections. It seems to me that there are two ways out of this particular problem. The first is to say that the atmospheric disturbance is running in terference in this case and so the case of error does not count. Generalizing this, though, would seem to have unwelcome skeptical results, as it is hard to see how we can be sure in any particular case that there is not some similar interference in play. One might say that the disturbance is such that we are getting the idea of motion from the atmosphere, or the somewhat mirage like image of the tower, and not the tower itself. Again, this seems to intro duce a level of complexity Locke would find uncongeni al, as it makes it very hard to understand how we can be sure that actual ideas received from experience are real, adequate and true which leaves the door open enough for skeptical worries to creep in. Fortunately, the second way to respond to this proble m seems to do the trick. As I have emphasized, Locke has a deliberately broad understanding of the term idea He the Object of the Understanding when a Man th inks (1.1.9). I have argued that this include s both particular perception s, or percepts, and what we think of as concepts ; we have both imagistic and non imagistic ideas We can think about a beautiful sunset or a terrible tragedy, so those can be ideas i shaped motion and white are. At least some of the former set of ideas are plausibly thought of as some kind of images which are perceived mentally when we think of them, while the vast majority of the latter set seems to not involve images at all (certainly not in any necessary sense). If I recall a beautiful
213 sunset, the object of my thinking is a recalled image. However, if I contemplate the nature of justice, I need not think imagistically at all. So, it seems one could res pond to the apparent problem of perceptual error by a rguing that it conflates percepts (or imagistic ideas) with concepts (or non imagistic ideas). When we is false, because the percept does not conform to the actual shap e of the thing in question. Similarly, our erroneous idea of motion is a perception of motion where the tower is not actually moving. However, our simple ideas of shape 23 and motion are not percept s but are simple general ideas, or concepts, and those are, according to Locke, real, adequate and true, as specified in Â§1 24 On this reading there seems to be no real problem of perceptual error for Locke at all, and this does seem to generalize over any possible case of perceptual error. Whether it is of a prim ary or secondary quality, the mistake will involve a particular percept, or set of percepts, and not the simple ideas we have as such. Again, and finally, it seems that Locke does not actually have a as been refuted. 6. 4 Troubling Consequences While the reasoning in the previous section seems to deal with the putative problem of perceptual error, the argument presented in Â§ 3 .1 seems to have troubling consequences for d about our simple ideas of primary qualities is correct, and it seems to me that it is, Locke, or anyone developing a Lockean theory, faces two difficulties. The first is that we seem to be owed a story about those ideas of primary qualities that might ha ve 23 For the sake of the argument, I am momentarily granting shape the status of simple idea, even though it really shaped 24 I hasten to add that there seems to be no good way to think of these simple ideas ( shape and motion ) as both simple and general images without running directly afoul of Berkeleyan objections about their general applicability and inconceivability.
214 seemed simple (ideas like circle and triangle ) but are not. The second is that it is not clear how theory. In response to the first, part of such a story w ould certainly be to say as suggested in Â§3.2, that they are simply complex ideas of primary qualities, either simple or mixed modes. Because they can be analyzed into parts, our ideas of a shape, or a motion, for instance, should be understood as either Idea Ideas (2.12.5) While it is perhaps un clear where exactly a parti cular idea of, say, a shape (like square or circle both of which are general ideas) or a particular shaped idea (like the idea of the shape of that particular square tower) falls in the mode taxonomy, 25 if they are not simple ideas, they must be modes. 26 T he response to the second difficulty, though, is quite a bit less clear. Whether simple or mixed, modes are supposed to be the combinations (and/or variations) of simple ideas. In the case of simple ideas of primary qualities, what Locke has to work with a re bulk figure number situation motion rest space and extension where I have argued this must mean having bulk being figured having some number being situated etc. How are we supposed to combine and vary being figured (or having some figure ) and having extension to come up with square or 25 British Journal for the History of Philosophy 10 (3) 2002, 359 378 (and particularly 361 should be considered simple or mixed modes. She thinks Locke takes them to be simple, but they are actually mixed, whil counts them as mixed. For now, I offer the evidence of 2.30 32, in which Locke clearly, and fairly explicitly, includes triangle as an idea of a mixed mode. I return to this discussion in the following chapter. 26 12 .4), as we might think the idea of the square tower itself does, which would make them ideas of substances.
215 circle ? It is not difficult to see how we might abstract from our ideas of various shapes, or various shaped things, to come to the simple idea of shaped but it is extremely hard to see how we can work the other way around to come up with the idea of a particular shape from such simple ideas. While the simple ideas Locke has listed do seem to be the simplest ideas of primary qualities, they do not seem to be sufficient to serve as the building blocks of the modes we have. What we seem to need, in the case of our ideas of shapes, for instance, is some at least relatively simple idea of an extension not just having extension (or being extended ). Perhaps from that idea, and maybe combined with being shaped we might be able to combine and vary our way to modes of shapes. 27 It seems to me that such ideas would also have to, at least implicitly, include something like plane figure but the point remains that we need these other ideas to have ideas of things like sh apes and motions, if those things are not simple ideas themselves and if Locke is right about modes. One might think that what Locke wanted to say was that modes were the result of combining and varying relatively simple ideas, like those more traditionally thought of as Lockean simples. These would include some kind of extension and space, perhaps, as well as basic shapes, and so forth. While these might not satisfy the simplicity criteria Locke specified, they do seem to meet the standards of his real, adequate and true ideas. For one thing, i t seems that Locke would want to say that our idea of squareness does accurately conform to (in some appropriately general sense) the actual squareness of bodies and drawings 28 Unfortunately, 27 Note that this way of putting the process suggests that the resultant ideas would be mixed modes. 28 Of course, things that we take to be square in the real world are never perfectly so, and Locke acknowledges this.
216 Locke does not say this, and the taxonomy of ideas that he gives us is not apparently compatible with any middle ground between truly simple ideas and modes. 29 ideas and clear that while we can set aside the supposed problem of perceptual error, the consideration of that issue and a close reading of the Essay have l That is, what Locke says about the compositional simplicity of simple ideas leaves his account of other i deas crippled. This might be a relatively easy fix for a generally Lockean theory, but it seems to be a serious problem for the account provided in the Essay 6. 5 Results In the foregoing sections, I have endeavored to get clearer on what Locke means when he amounts to considering the completeness and representational relationship of various types of ideas from slightly different perspectives. Employing a close reading of the text of the Essay on this issue enabled a response to the problem of perceptual err or raised by Antonia LoLordo. Unfortunately, though, the discussion of that response unearthed a problem at the very bedrock ideas in the way Locke requires fr om the simple ideas he actually allows for in the Essay I am unfortunately unable to supply a truly Lockean response to this problem, and I am not aware of anyone else who has noticed this issue. 29 Alternatively, perhaps a Lockean might say that the list of simple ideas we have been working with is not intended to be complete, and that there is r oom to include some fundamental idea of an extension perhaps a Hobbesian perceptual minima. However, I do not know of anything in the Essay that would support such a move.
217 shaped and in moti on Essay Assuming for now that such a response is pos ition on general ideas and demonstra tive knowledge, and in particular his understanding of mathematics and geometry.
218 CHAPTER 7 IDEAS, MATHEMATICS AND GEOMETRY According to Locke one of the most secure kinds of knowledge we can have is Disagreement of any Ideas without intermediate steps involved in the perception. Instead, the mind proceeds by the comparison of intervening ideas to form proofs which, in principle, yield certain knowledge (4.2.2 4). Locke identifies two subjects that should count as at least potential systems of demonstrative knowledge, or science: mathemati cs (including geometry) and ethics. 1 Unfortunately, what Locke says about th e latter is tantalizingly brief and primarily for that reason, I will be foregoing a consideration of that subject, focusing instead on just mathematics, and geometry. Perhaps u particularly geometry. Early on, both Leibniz and Berkeley raised problems for Locke on this score. 2 More recently, Emily Carson has developed a sustained critique of Locke on mat hematics. Ultimately, as I will shortly argue, Locke takes the standing of demonstrative knowledge to depend largely upon the nature of the ideas involved in the propositions used. The high standard of confidence we have regarding claims made in mathemat ics and geometry, for instance, results from the fact that they involve ideas of modes and relations, which, as we have 1 Locke primarily talks about demonstrative knowledge, but he does say that this kn instance, at 4 2 4 2 .10). 2 theory of mathematics is generally which was taken to be problematic for (at least some of) the reasons suggested by earlier critiques. However, one n right. Of course, Mill offered his own empiricist theory. This history is worth further study, but lies outside the bounds of my current project.
219 see n (in Chapter 6), are adequate. That is, they are complete representations of their objects, which are themselves. Such an idea is, a s Locke says, its own archetype. We do not primarily use the idea of circle for instance, to represent things in the world but to represent the idea of something that is a closed plane figure equidistant from its center at every point along its perimeter. It is against that idea that we compare things in the world to determine if they count as circles. Locke takes the same to be true of the ideas and propositions involved in a putative science of morality. On the other hand, Locke tells us that knowledge of substances does not amount to much. Because the archetypes of these ideas are external to us, we cannot be sure that substance ideas are adequate (4.4.11 12). 3 Locke seems to think that the most we can be sure of is that when we have experienced some si mple ideas to come together, united by a single substance, they may do so again (and therefore be combined in the idea of that substance had an union in Nature, may be united again (4.4.12). Locke does not say much more than this about knowledge depending on ideas of substance, and I will not discuss it further. understanding of demonstrative systems of knowledge. First, in Â§1, I will explicate more fully what the Essay says about such knowledge, and develop an understanding of how the adequacy of the ideas involved support the resulting knowledge claims. Then, in Â§2, I will discuss several nd geometric project. In Â§3, I will consider at length the critique offered by Carson. In Â§4, I will consider a Leibnizian objection developed by Carson, which stands somewhat apart from her main argument. Throughout, I will argue that these objections and 3 S ee Chapter 6, Â§ 1.2 for more on this.
220 7. 1 Locke on Demonstrative Knowledge To be worthy of the name science at least in principle, and conclusions that are not merely trifl claims that are proven by a chain of intuitively certain steps. Locke thinks mathematics satisfies these conditions. It is not difficult to show that mathematics does, and morality might, meet the standard of demonstrative knowledge, but meeting the slightly higher standard of scientific kno wledge is clearly desirable, and Locke does talk as if he thinks this standard is met. Much of involved in these putative sciences, which has been developed i n the previous chapters, particularly Chapter 6. Here I will merely briefly assemble those considerations before considering some potential problems for the account. Considering mathematics 4 as the paradigm case of a science in the foregoing sense, we can be sure of our claims in this area primarily because they involve only simple ideas and relatively simple modes, though they are very often mixed gone to considerable trouble, detailed in Chapter 6, to explain his understanding of our simple ideas and ideas of modes as real, adequate, and true. Basically, simple ideas meet these criteria as the fundamental ideas received from experience. As the given, most basic elements of that experien ce, t hey c ould not be other than they are; as simple ideas, the y are necessarily complete; and as the most immediate product of our interaction with the world, they automatically conform to that which they represent, the experience we have of their objects Mixed modes and relations 4 Hereafter, unless otherwise specified, all references to mathematics should also be understood to include geometry.
221 are all real, adequate, and true because they are the archetypes created by the mind to which reality is compared That is, the idea of square is not adequate because it matches up with squares in the world, but those latter sha pes count as squares because they conform to our idea of squareness Such ideas of mixed modes and relations are inherently complete, since they are their own arche types, are as the mind intended, and, so, automatically conform to what they represent ( that is, themselves). 5 Because of this status (as real, adequate, and true), these ideas are capable of being such that we completely understand their essences. That is, we can grasp what it is about these ideas that explains why their constituent simpler idea s must go always together, and why objects that conform to these ideas have the properties they do. This is because the real essences of these things just are their nominal essences, of which we can have complete, adequate knowledge. Moreover, because ou r mathematical ideas are so (relatively) simple, it is possible for us to use these ideas without an undue likelihood of error, as long as we are careful. Also, as James Gibson observed in 1896, 6 the ideas of mathematics are also particularly distinct. We might be inclined to confuse, for instance, different ideas of shades of green because they are so similar, but the ideas of mathematics are easily distinguished from each other. Even if it is well nigh impossible to distinguish one million pebbles or peop le from one million and one, with the use of mathematical notation and reference to the ideas involved, there is no trouble at all in seeing that they are perfectly distinct. The same is true of geometrical ideas; we cannot possibly confuse the 5 On the other hand, for reasons covered in Chapter 6, ideas of substance are problematic. They are most often not real, adequate, or true. 6 Mind Vol. 5, No. 17 (Jan., 1896), 46.
222 ideas of tr iangle and square 7 It does seem particularly difficult in these cases to distinguish between the idea and the notation of that idea, and therefore what exactly is being distinguished in comparing one million and one million and one. It seems to me, howeve r, that the idea and notation are distinct, but the idea is primary. It also seems that one of the advantages mathematics has over other areas of inquiry is that the notation we have developed for its ideas is much more precise than any other notation pres ently available. the Mind in all its Thoughts and Reasonings, hath no other immediate Object but its own Ideas Knowledge then seems to me to be nothing but the perception of the connexion and agree ment, or disagreement and repugnancy of any of our Ideas called way of ideas. The only unmediated objects the mind can consider are its ideas, so, Locke thinks, the only thing we can really claim to have knowledge about (or to think knowledge ranges over) are those ideas. So, if we can only have knowledge of propositions, which Locke seems to have assumed, those propositions, in affirming or denying anything, must be stating a perceived agreement or disagreement between our ideas. Unsurprisingly, the degree of the confidence in those knowledge claims depends upon the adequacy of the ideas we are comparing. The highest degree of knowledge, that attended by the most certainty, Locke calls intuitive knowledge. This involve or Disagreement of two Ideas White is not Black 7 Note that if we thought these ideas were strictly imagistic there would clearly be a problem with distinguishing thou sand sided figure (the notorious chiliagon ) from a thousand and one sided figure. The reading of Locke I am advancing in this dissertation allows for two such ideas to be perfectly distinct because of their specification (the number of sides). Locke is qui te clear that he thinks all of these ideas of mode are perfectly distinct, which again lends support to the not strictly imagistic thesis.
223 Circle is not a Triangle Three are more than Two and equal to One and Two next degree of knowledge, which may still be certain, is demonstrative knowledge, in which the Thus the M ind being willing to know the Agreement or Disagreement in bigness, between the three Angles of a Triangle, and two right ones, cannot by an immediate view and comparing them, do it: Because the three Angles of a Triangle cannot be brought at once, and be compared with any other one, or two Angles; and so of this the Mind has no immediate, no intuitive Knowledge. In this Case the Mind is fain to find out some other Angles, to which the three Angles of a Triangle have an Equality; and finding those equal to two right ones, come to know their Equality to two right ones (4.2.2). We cannot simply compare the ideas of the three angles of the triangle and the two right angles, but need to use a method of comparing the two ideas by way of intermediate angles. In other words, we need to construct a geometrical proof. This kind of demonstration depends upon our adequate grasp of the ideas involved. If, say, we did not know all of the constituent ideas of triangle, we could not be sure that our comparisons of that id ea to other ideas were taking everything relevant into account. We could hardly think the resulting perception of agreement or disagreement was certain. To put the matter positively, whenever we can be sure of our perception of the agreement or disagreemen t between two ideas, whether intuitively or demonstratively, we can have knowledge that is certain (4.2.9). We can be so sure when we compare simple ideas and ideas of modes and relations. It is with these that, say, mathematical propositions are concerned Given that we can (in principle) be certain of a given set of propositions, we must be sure that these are instructive propositions and not merely verbal, or trifling, truths. Locke says the following of these two categories: We can know then the Truth of two sorts of Propositions, with perfect certainty ; the one is, of those trifling Propositions, which have a certainty in verbal Certainty but not instructive. And, secondly, we can
224 know the Truth, and so may be certain in Propositions, which affirm something of another, which is a necessary consequence of its precise complex Idea but not contained in it. As that the external Angle of all Triangles, is bigger than either of the opposite internal Angles ; which relation of the outward Ang le, to either of the opposite internal Angles, making no part of the complex Idea signified by the name Triangle, this is a real truth, and conveys with it instructive real Knowledge (4.8.8). So, if we move from a complex idea, like that of a triangle, which we understand completely, to with more than one adequately comprehended idea and their relations could also result in this type of instructive knowledge, while simply enumerating components of the complex idea(s) as simple combination would be a merely trifling, or verbal, prop osition, though it might be certainly true. 8 This is not to say that such verbal propositions do not have their place, as for instance in providing definitions, but they do not expand our the shewing the me aning of one Word by several other not synonymous Terms of a term in my mind clear to someone else (3.4.6). 9 This is a necessary component of all discourse, but spelling out the meanings of our t erms hardly counts as an addition to our knowledge. Mathematics is, at least partly, so valuable because we can work from a set of fundamental ideas to a set of necessarily true but previously unknown propositions. From the foregoing, it should be clear h ow it is the nature of the relevant ideas, which are ideas of modes and relations, which are all perfectly adequate, that makes it possible to have 8 Note that Locke is here ignoring a distinction between verbal and trifling truths about which I say more in Â§2.1 9 would yield different results. I have suggested that definitions (understood perhaps more generally than Locke is intending) also have a role in explaining our ideas to ourselves (in Chapter 3).
225 demonstrative, non trifling knowledge of mathematics. This kind of knowledge qualifies as 10 ) for Locke. The foregoing explains why Locke thought mathematics was the archetype of the sort of science that could give us new, certain truths. Of course, none of this is to say that there may not be problems with this account. Unsurprisin position is Berkeley, whose objection rather entirely follows from his rejection of Lockean abstraction. 11 However, I take the discussion in Chapter 5 to constitute a rather thorough covering that material. Rather, in the following sections, I will consider several objections to Locke from more recent commentators (though Leibniz does make an appearance). 7. 2 Trifling Trut hs? Much more recently than Berkeley or Leibniz, a particularly helpful discussion of 12 c 10 See 4 2 .8, for example. 11 For an excellent discussion of this, albeit from a pro Berkeley perspective, see Douglas M. Jesseph, Philosophy of Mathematics (Chicago: The University of Chic ago Press, 1993), particularly 19 28 and 241 246. 12 Journal of the History Philosophy Vol., No. 4 (Oct., 1990), 511 should not Instructive Distinction with the Analytic The Locke Newsletter Instructiv e Distinction The Locke Newsletter 11 (1980). While the questions involved in that debate are interesting, they are also beyond the scope of this dissertation. The closest one comes in that discussion to a concern that is within my purview is th mathematical propositions should be understood as analytic or synthetic, in a Kantian sense. I take that question to be a side issue that I will pass by, though I can see returning to the subject at a later date.
226 merely analytic), 13 Cicovacki indirectly considers the worry that these conclusions might be theory delivers non trifling truths in this area. Here it may be useful to clarify the distinction between trifling and merely verbal truths, according to the agreeme nt or disagreement of the Ideas they stand for, without regarding whether our Ideas (4.5.8). 14 We might, for instance come to a merely verbal truth about unicorns (say, that they eac h have a single horn), but without any supposition that there really are any unicorns. On the other hand, Lockean instructive, non another which is a necessary consequence of its precise complex Idea (4.8.8). So, a trifling truth would be one that affirms a necessary consequence that is trivially true because it is contained in the complex idea in question. I will return to this issue in greater detail in the next section, bu t for now it will suffice to observe that a trifling truth is trivially true and not illuminative and a merely verbal truth is true but makes no pretention to refer to anything that is actually true in the world, as one might say. concern, we might particularly worry that mathematical claims turn out to be trifling if we think that the simple combination s of our mathematical ideas are overly comprehensive. If too much is allowed into the simple combination s of our ideas it seems th 13 Sybil W Instructive Distinction with the Analytic The Locke Newsletter 9 (1978), 46. 14 real Truth when these signs are joined as ou r Ideas agree; and when our Ideas are such, as we know are capable of having an Existence in Nature: which in Substances we cannot
227 recall what the point of simple combination is, and that is to tell us both what is the essence of an ape case of triangles. If we think that the normal simple combination of triangle includes something external angles that are each bigger than either of the opposite we have misapplied the notion of simple combination and have packed much more into that for triangle than we should have, thereby sinning against the inherent, reasonable Now, Cicovacki observes that in 4.2.2 (quoted in Â§1, above), in which Locke discusses our coming to see that the interior angles of a triangle equal two right angles, Locke is saying that we cannot have immediate intuitive knowledge of this equality but that we must com e to it by way of an intermediate comparison. 15 We do not merely compare the definitions to arrive at this knowledge of equality but have to work out the implications of the relevant ideas ( interior angles triangle and right angles in this case) and the c omponents of their simple combinations While this might seem to be a kind of demonstration from definitions, it seems clearly to be informative rather than trifling. On the other hand, it seems possible that those who have specialized or expert understa nding of particular ideas might have lengthier, more detailed, simple combination s of those ideas. An astrophysicist, for instance, might, and almost certainly does, have a much more robust understanding of event horizon than I do. I think it makes sense t o say that more knowledge claims involving those ideas are trifling, for that person. What might be an informative demonstration to most of us would be just trifling for Stephen Hawking. This does 15 Cicovacki, 515.
228 as we can maintain the distinction between informative and trifling. 16 In this regard, one might wonder about communal knowledge not what I know, but what is known and how this is related to this distinction between, say, casual and expert knowledge. Locke does not seem to have noticed this distinction. Rather, he explicitly talks about the knowledge had (or not) by individuals, and then simply assumes that what he says will apply to knowledge had more generally. This suggests that he was not aware of the d istinction I have drawn between relatively casual and expert knowledge, as without that distinction it is not difficult to assume that what is certainly known by an individual is so known generally. Although it hardly counts as a robust objection to think ing of mathematics as understood by Locke to be a science, one might also worry about the notion that our mathematical that is, merely verbal. Strictly speaking, yes, these propositions are only about our ideas. To say that the sum of the interior angles of a triangle equal two right angles is only to say that what is meant by those words and ideas about triangle here i s about the applicability of mathematics to the world of more or less concrete particulars. This issue hinges upon the relation between general mathematical ideas, almost all mixed modes, and particulars. 16 It is necessary that our simple combinations be limited, though, due to our finite mental capacity. Perhaps there are complex ideas for which we can have complete simple combination s but the vast majority of our simple combinations must be quite limited. Someone with an unlimited mental capacity, t hough, could plausibly have perfect knowledge of his or her ideas, and it would seem to follow from the above considerations that this person would be stuck with merely trifling knowledge claims.
229 As Cicovacki observes and as I said in Chapters 5 and 6 these complex ideas are, for Locke at least, mental categories, or archetypes, against which we compare the constituents of the world. The passage from the Essay already partly quoted from, is: I doubt not but it will be easily granted that the Know ledge we may have of Mathematical Truths is not only certain, but real Knowledge ; and not the bare empty Vision of vain insignificant Chimeras of the Brain: And yet, if we will consider, we shall find that it is only of our own Ideas The Mathematician co nsiders the Truth and Properties belonging to a Rectangle, or Circle, only as they are an Idea them existing mathematically, i.e. precisely true, in his Life. But yet the knowledge he has of any T ruths or Properties belonging to a Circle or any other mathematical Figure, are nevertheless true and certain even of real Things existing: because real Things are no farther concerned, nor intended to be meant by any such Propositions, than as Things real ly agree to those Archetypes in his Mind. Is it true of the Idea of a Triangle that its three Angles are equal to two right ones? It is true also of a Triangle where ever it really exists. Whatever other Figure exists, that is not exactly answerable to th at Idea of a Triangle in his Mind, is not at all concerned in that Proposition. And therefore he is certain all his Knowledge concerning such Ideas is real Knowledge: because intending Things no farther than they agree with those his Ideas he is sure wha t he knows concerning those Figures, when they have barely an Ideal Existence in his Mind, will hold true of them also, when they have a real existence in Matter; his consideration being barely of those Figures, which are the same where ever, or however th ey exist (4.4.6). 17 Mathematical ideas are formed by the mind, at least partly to serve as archetypes of things we might meet with in the real world. These archetypes are, because of their simplicity and distinctness, able to serve as the foundational elem ents in a demonstrative, instructive science that is first and foremost concerned with our ideas, and is only secondarily applicable to the non ideal world to the extent that the concrete particulars of that world agree to, or match up with, those archetyp es. To the degree that those particulars fail to match up with the corresponding archetypes, we must be less confident in our applied mathematical conclusions. Whether or not 17 I have included the entire passage here because of its ext reme clarity and relevance, both for the subject at hand and for its applic ation in the following sections.
230 the foregoing consideration of the Essay mind when he speaks of this sort of knowledge. 7. 3 ma thematics, or, more accurately, geometry. 18 The core of her most fundamental argument against Locke is: certainty of demonstrative knowledge (in particular if this is supposed to amount to apriority and to carry with it necessity) relies on instructiveness of demon strative knowledge [of geometry, presumably], however, appeals to the idea as a construction against a spatial background; the content of such knowledge is in effect smuggled in via the simple idea of space. What we can know with certainty a priori of Mr. Micawber is just what Dickens said of Mr. Micawber, hence is not instructive. The instructive knowledge we have of triangles depends on the simple idea of space and hence, without further explanation, is not certain. 19 have either certainty or instructiveness regarding our geometrical claims, not both. Although it is far from clear that Locke would acknowledge the sense of apriority or necessity Carson seems to want to ascribe to him, I will for now forego a direct disc 18 British Journal for the History of Philosophy Vol. 10, No. 3 (Aug., 2002), 359 Locke Studies Vol. 5 (2005 ), 19 Intuition and the Axiomatic Method ed. Emily Carson and Renate Huber (Dordrecht: Kluwer Academic Publishers, 200 6 ), 3 on Locke on Locke Studies Vol. 7 (2007), 21 46. As the most prominent recent critic of Locke on mathematics, I devote considerable space to her objections. I think this is justified by the fact that this discussion illustrates con nections with the discussions of previous chapters, and serves as a fairly detailed application of 19 Carson, 2002, 371. Carson makes the ver y same argument in Carson (2006 ) in the middle paragraph on page 9. The allusion to Dickens and Mr. Micawber is originally due to Cicovacki.
231 that any of our ideas are as premises in reverse order: instructiveness before certainty. I will then, briefly, return to the above general argument and show that Carson has failed to adequately support her position. However, it will be important to first get clearer on an issue at the heart of her critique: the status of geometrical ideas as simple or mixed modes. 7. 3.1 Simple vs. Mixed Modes As mentioned in the above argument, Carson takes the instructiveness of our geometrical background; the content of such knowledge is in effect smuggled in via the simple idea of s 20 This position requires her premise that our ideas of shapes are simple modes, but I will argue that they should be read as mixed modes. Moreover, as I will show, even if they are supposed to be simple modes, these ideas do not smuggle their spatial background but include it outright. Carson observes that ideas of geometric figures at least implicitly include that of space and, typically, that of Euclidean space. This point appears again and again in her critique of Locke. 21 Carson takes Locke to be committed to the position that the ideas of geometric figures are simple modes, which, because not mixed modes, cannot include anything but the simple idea of space If this were so, it would be very hard indeed to understand how our geometrical claims co uld be understood as instructive, as Carson points out. To the extent that geometry is 20 Ibid. 21 identity of real and nominal modal essences fails because he does not take into account the role of a background
232 instructive, it must include some notion of the space in which its claims apply (e.g. Euclidean space). Ergo, because she takes geometrical ideas to be simple modes, she thinks the idea of Euclidean space take place we would not have instructive knowledge. Although Locke and most of his commentators, s eems to think that mathematical or more accurately, geometrical ideas are mixed modes, Carson argues that they are really simple modes at least as far as the theory of the Essay goes Numbers clearly are simple modes, being nothing more than the result of repeating the simple idea unit ( 2 12 .5, 2 13 .1), but Carson points to the discussion in 2 12 and 13 as evidence that ideas of figures are also simple modes. variations of the same simple idea 22 Moreover, Locke seems to think that mixed modes are put together arbitrarily, while our ideas of geometrical shapes are constrained, by the simple idea of space. 23 In the making of an idea of a mixed mode, we are free to put together whatever ideas we like so long as they imply no contradiction, but in the making of a geometrical idea we are obliged to include the idea of Euclidean space 24 So, while we can arrive at certain knowledge involving mixed modes by an analysis of the ideas we combined to arrive a t those mixed modes, because we have maker knowledge of those modes, we cannot do the same for ideas of figure, because they are not simply combined by us, but have this external constraint. 25 22 Carson, 2006 point line or plane. 23 Ibid., 22 24. 24 Assuming that we are engaged in Euclidean geometry. 25 Because of these considerations, and contrary to the usual position taken by commentators, Carson thinks that we can have better, actually demonstrative, knowledge of the ideas associated with ethics than we can of those of mathematics. This is because, on her reading of Locke, the former ideas are mixed modes, while the latter are simple
233 Carson repeatedly emphasizes the importance of the way we come to have simple and primarily 26 However, it is hardly clear that Locke thought of t his a s the primary distinction. In 2 12 for instance, where Locke first introduces simple and mixed modes, the distinction is clearly a different combination s of the same simple Idea Idea of several Ideas of several kinds on of how we work with our simple ideas to come up with modes in 2.13 that Carson draws our attention to seems very much to be making a secondary here saying that how we come to form simple or mixed modes is the primary distinction between them, which he seems to have already g iven us in the earlier definitions of those terms. Rather, he is simply discussing a different distinction between these two types of ideas. There is no reason to think Locke considered this distinction even as important as the initial distinction given in 2.12, let alone the primary distinction. Of course, Locke recognizes both distinctions; the problem is that Carson seems to ignore, or at least deny the importance of, the distinction of what is included in these ideas. The emphasis she puts on the forma tion of modes seems to be doing some work in her reasoning. but cannot have th e same of simple ones, making the notion of a demonstrative science of mathematics problematic on a Lockean model. 26 Ibid., 20, bold emphasis added.
234 However, Carson is right that Locke seems to suggest in 2 13 which is explicitly about simple modes of space, that ideas of various figures are simple modes. For instance, he says t distance is a different Modification of Space, and each Idea of any different distance, or Space, is a simple Mode of this Idea (2.13.4). H Power, by varying the Idea of Space; and thereby making still new Compositions, b y repeating its own Ideas and joining them as it pleases, is perfectly inexhaustible: And so it can multiply Figures in infinitum (2.13.5). T these infinitely various figures to be simple modes of space. This is reinforced when Locke says Figures both in their Shape, and Capacity, in infinitum all which are but so many different simple Modes of Space (2.13.6). It would be hard for Locke to get more explicit than that: our ideas of figures are all simple modes of space. There is at least, one important point to be made here. N ote that this discussion seems to be about our ideas of particular figures (this particular triangle or circle) not of general ideas of types of figures ( triangle or circle as archetypes of shapes various particular existent shapes might fall under) Carson does not seem to notice this distinction. Regarding, for instance, the general idea of triangle both Locke and a space between three lines, is the real as well as (3.3.18). While Locke thinks that lines are simple modes of space, and space is a simple idea, they are different ide as 27 and Locke seems to think that the idea of triangle essentially includes at least those ideas. It also seems to include the idea of figure (which I have argued should be read as being 27 I am aware that there is some oddness here. If triangle is made up of ideas, such as line angle and spa ce which are all either space or simple modes of space it might be thought that triangle itself is only a, rather complex, simple mode of space I confess that I am not sure what to say to this beyond that Locke does not seem to have thought about this i ssue, and, as suggested above, I do not find this reading of Locke intuitive.
235 figured or its equivalent 28 ) geometry throughout the Essay suggest that Locke considers these general ideas as mixed modes. This, in turn, suggests general ideas. However, I think the situ ation is a bit worse for Locke, because of the apparent conflict between what Locke says here, in 3.3, and what he says in 2.13. What Locke says about ideas of figures being simple modes of space in 2 13 .6 seems to be either just about ideas of areas of p articular shapes, or figures, or simply wrong. It is possible that Locke was unaware of the distinction between, say, the idea of triangle with the idea of a triangular space The latter seems to be plausibly a simple mode of space; that is a modification in the extended sense Locke has in mind of the simple idea of space arranged in such and such a way, extended in a particular way and to just such an extent. Such an idea would be a particular arrangement of instances of a single simple idea, space The id ea of triangle though, as just mentioned, seems to necessarily include more than the simple idea of space. The same is true of any idea of any particular figure (though plausibly not of any idea of a particular figured space). The idea of a particular fig ure, like triangle includes lines, the space enclosed, angles, 29 and the at least implicit idea of the kind of space in which they are conceived. The last is particularly important, as will become clear shortly. Possibly the more charitable thing to say is that Locke was unaware of the distinction at hand, but there seems to be no reasonable way to think that actual ideas of shapes of figures, or shapes, as such, should not be considered to be mixed 28 See Chapter 6. 29 Locke does not give us a story about the derivation of the idea of angle If it is a simple mode of space then the conc ern mentioned in footnote 27 applies he re as well. It might then be necessary to show that Euclidean space is also a simple mode of space
236 Moreover, the discussion in Chapter 6, in which it was argued that the simple idea of figure should be understood as being figured lends strong (if not conclusive) support to this reading. Any given shape, which might in itself be a simple mode, when considered as a shape, must i nclude the simple idea of being figured (or having shape ), in addition to whatever combinations of the simple idea of space or extension are involved, regardless of whether or not the idea of the shape usually also includes the idea plane figure as I thin k is the case. As a simple idea, being figured is not a mode of space (a distinct simple idea). Therefore, the idea of a shape (as a shape) must be a mixed mode. This reading also seems to make the best sense of everything else that Locke says about such ideas, mostly triangle So, it seems, Locke must have, at least, been unclear in 2 13 .6, ignoring an important distinction, or, at worst, made an outright mistake about the status of an important set of ideas. Regardless of which way to go, it seems to make sense to think that Lockean ideas of geometric figures are mixed modes, not simple. d modes. She seems to think that this is because Locke treated them as such and because treating them as such might enable us to have instructive knowledge about them. However, I think the more fundamental reason is that this is the only reasonable way to think of these ideas, as I have just argued. In Lockean terms, ideas like triangle square and circle must be mixed modes. All of this is important in understanding how it is that mathematical claims can be ys that because ideas of shapes are simple modes, they do not carry with them the content that makes instructive claims possible. That only comes when those simple modes are considered in conjunction with the idea of the kind of space in which the shapes a re constructed (usually Euclidean space). As her general argument shows, she
237 takes this to have several unwelcome consequences for Locke. However, if the ideas of shapes used in geometry are mixed modes, including things like being figured and plane figur e then we can have instructive knowledge of them by virtue of considering various implications of those elements of their simple combination s and comparisons with those of other ideas. The following discussion makes this a bit clearer, and I return to Car following section. Carson says that there is an important difference between simple and mixed modes, and this may be so, but if geometrical ideas really should be understood to be mixed modes then it is not clear one thing, we can have certain knowledge of mixed modes, though the question remains of how instructive our geometrical claims are remains. I will return to this questi on after considering three further issues suggested by the foregoing discussion. First, if we understand the ideas of geometric figures as mixed modes, there seems to be no reason to think that the idea of the kind of space in which such a figure is consi dered cannot be included (at least on some level) in the simple combination s of such ideas. In discussing the free activity of the mind in forming modes, Carson observes that it is not just up to us what kind of line we can use to join the ends of two alre two lines, it is now determined by something other than the free activity of the mind how a closed 30 Eucli dean (or whichever) space. However, the mind is also constrained by the kind of figure it wants to form. In fact, the implicit inclusion of such a constraint seems clearly necessary, even oblem if we take Lockean 30 Carson, 2006 23.
238 ideas of geometric ideas to be mixed modes after all. If geometric ideas are mixed modes, including more than just the simple idea of space, then there is no problem with thinking that one of the ideas combined to make the mixed mo de of, say triangle is that of Euclidean space Second, we might also wonder if the idea of Euclidean space, for instance, is not clearly implicit even if we take geometric ideas to be simple figure with three sides straight sides. If that is so, it would seem that he builds Euclidean space into this very notion, since three points define a 6. He does not, however, really define the term and acknowledges both straight and crooked lines, though he seems to default to straight lines. However, note that the way Locke talks ab out configuring simple ideas of space into simple modes of space in 2.13.5 6 to come up with the idea of a figure must imply that they are being configured in some particular way. We can join lines, repeat them, etc., as we please, but in doing so, if we a rrive at some particular, determinate figure, of which we then have an idea, the way that we have modified, in this sense, the idea of space builds in the notion of the kind of space we were working in, imagining, or presupposing when we got the idea of th at figure, either when it is spelled out that the idea of the kind of space involved is implied even by a supposedly simple mode, it is hard to see why this, as such, should be seen as a problem for Locke if the 31 31 Another point is suggested by footnotes 27 and 29 If it turns out that Euclidean space should be considered a simple mode (which Cars on seems to think is not the case), and we can make sense of a complex idea combining simple modes of the same simple idea (like space ) being a simple mode, then ideas of geometric shapes might be simple modes after all. However, then it seems clear that C
239 Third, we might wonder about the status of the idea of Euclidean space itself, in Lockean terms. Locke does not seem to have given this any explicit thought; it seems likely that he did not consider any alternative, which might have resulted in his not taking this idea as one he needed to deal with. It does seem to me, though, that this is an idea that is available to u s through abstraction. If we can arrive at our idea of plane (as it is normally used in Euclidean geometry) through abstraction, it seems that we can also arrive at the idea space organized on three planes, through abstraction and the formation of a mixed mode. I take it as fairly clear that we can form the idea of plane by way of abstraction. Alternatively, Euclidean space might be thought of as a particular simple mode of space As suggested by some of the earlier discussion, if plane is just a simple mod e of space then, perhaps my first derivation of Euclidean space might result in a simple mode, too. 32 In any case, it seems that Locke should not be seen to have a problem with Euclidean space as such. In summary, while ideas of particular shapes might be best thought of as simple modes, ideas of shapes as such (e.g., triangle square circle ) are mixed modes. Even if Locke did not see this, it is required by his theory as presented in the Essay as show n by the arguments in Chapter 6 and this sub section. However, even if one agrees with Carson that we should think of them as simple modes, we have good reason to think that the inclusion of Euclidean space as an implicit component of those modes is entail 32 On either of these latter two readings, Euclidean space would be a simple mode of space, which, according to Carson, is being illicitly included in other simple mode ideas of space. However, as the previous footnote suggested, this seems to result in the combination of two simple modes of the same idea, which seems to end up being another simple mod objection.
240 7. 3.2 The Instructiveness of Geometrical Knowledge geometrical claims. Recall that this part of Car by the use of the idea of a spatial background that we can make instructive claims about geometrical ideas like triangle 33 The allusion to Dickens and Mr. Micawber indicates the problem Carson has with th e instructiveness of claims involving mixed modes, and is simply the position that such claims are trifling, which was discussed in Â§1. Carson thinks this worry is avoided by Locke through our geometrical ideas being simple modes of space. The contents of result in such truths being non instructive. Our learning something instructive about those ideas inst which we construct geometrical figures. This is how Carson thinks Locke can make the claim that we have instructive, non trifling, geometrical knowledge. 34 There are two responses to this line of thought that follow from the foregoing. First, the conc ern about the lack of instructiveness of our true geometrical claims is misplaced. Even if the constituent ideas of such a claim are mixed modes the claim need not be uninstructive. If we are careful to distinguish between what is actually part of the cont ent of an idea, or what is properly included in its simple combination and what is entailed by that content, we can see that instructive, even surprising, knowledge can be derived from our mixed mode ideas. The informativeness arises from considering our ideas and following out implications of their individual simple combination s or the comparison of two or more ideas and their simple 33 Carson, 2002, 371. 34 The problem with this, she thinks, is that where the knowledge is instructive it is not certain.
241 combination s. So, if we can properly understand Lockean geometrical ideas as mixed modes, rather than simple ones, we need not worry about a conflict between the instructiveness of our true claims in that area and their certainty. Second, I have argued that we can, and should understand Lockean geometrical ideas as mixed modes. Although Locke himself seems to be a bit confuse d about this issue, the mixed mode reading is the best way to understand most of what Locke says about these ideas and enables us to understand that constraints like that of spatial presupposition are included (e.g., plane figure is (or ought to be) includ ed in the simple combination of triangle ). Even if we should go with the simple mode reading, I have argued that such spatial presuppositions are entailed by the simple modes Carson and Locke seem to have in mind. The notion of Euclidean space for instanc e, is not smuggled in but comes with, by entailment, our geometrical ideas, even if they are simple. As a result of these considerations, it 7. 3.3 The Certainty of Geometrical Knowledge According to Carson, the certainty of our mathematical and geometrical claims is supposed to derive from the fact that the ideas involved are entirely of our own making. This is ( 1) We have certain kn owledge only when our ideas are of our own making. ( 2) Mixed modes are of our own making, but simple modes are not. ( 3) Our mathematical and geometrical ideas are simple modes. ( 4) Therefore, we cannot have certain mathematical or geometrical knowledge. 35 While this argument is valid, I have already given reasons to resist (3). In the following, I will consider (1) and (2) and argue that we should not accept them either. 35 This does not apply to moral knowledge because that does invo lve mixed modes; see footnote 25 above.
242 Carson takes the important ingredient in our having certain knowledge of geometrica l objects to be the fact that we put the elements of those ideas together somewhat arbitrarily. Emphasizing that Locke says the mind makes new combinations of our simpler ideas into modes ssembled ideas are consistent, 36 geometrical objects, those ideas must be the product of the mind, without any other constraint beyond consistency. Any other restrictio n present would compromise the ideal arbitrariness on which our knowledge depends. However, as we have seen, geometric ideas of figures all come with the idea of Euclidean space, or some other space, so they are not solely the creation of the mind. As Cars 37 Carson takes this to have several unwelcome result s for Locke. First, it is worth pointing out that Locke does not think we create any simple ideas in the way Carson seems to have in mind. She seems to be aware of this, in general, but the passage just quoted suggests that it might be possible to create the simple idea of space. If we were to thinking). If Carson is not thinki ng of the kind of extreme ideality I take her to be, then it is rather hard to see how the force of her point is to be understood (see below). 36 Carson, 2002, 363. 37 Ibid., 370.
243 The Essay can be seen as a way of explaining how we can be understood to develop all of our ideas and knowledge from experiential bases. For Locke, none of our ideas is reducible to Lo 38 She takes this to be a problem for Locke when it is really his most fundamental point. She seems to think at the latter are created, whole cloth, by the mind, but this is not the distinction Locke has in mind, since he does not think any of our ideas qualify as such. Rather, for a nominal essence it is the combination that Locke thinks is arbitrary to the degr ee that it does not rely upon an externally existent referent, while the real essence is what explains the properties the mode or substance has. Since there is no external essence that explains those properties in the case of mixed modes, the nominal essen ce just is the real essence for those ideas. Nevertheless, Carson repeatedly emphasizes what she thinks of as a necessary ideal arbitrariness as the only thing that will qualify an idea as being a true workmanship of the understanding. However, at least s ometimes and to some degree, Carson seems to be confusing being the workmanship of the understanding and being the pure creation of the mind. As an empiricist, Locke would think the latter is impossible. According to Locke, we do not simply make up, from n othing, any of our ideas. Rather we are obliged to work with the materials given by experience, of one sort or the other. What we are able to do in the case of mixed modes is to combine them in whatever way we please, just so long as there is no contradic tion involved. The ideas that we combine are, of course, constrained by two things: they must themselves be consistent if they are 38 Ibid., 369.
244 to be capable of fulfilling their role as archetypes against which we compare existents and they must ultimately be gotten f rom experience (that is, their components (what is combined) must be derived from experience) 39 So, of course, our geometrical ideas are constrained by the idea of the space in which they are intended to occupy, and this poses no problem at all for Locke. The only way a Lockean idea of triangle might avoid being so constrained is if the idea of the space it inhabited were abstracted away, but I can see no indication that this is what Locke intended. Moreover, even if that were the case, the other ideas invo lved in triangle would not only depend on the mind or be solely in the sense that Carson seems to have in mind; they would be, to some extent, the material furnished by experience. Locke, after all does not, and cou ld not, require any ideas to be the sort of pure creation of t he mind that Carson seem s to think we need. 40 As we have just seen, Carson takes what she thinks of as the non ideality of our geometrical ideas to be a significant problem for Locke. 41 Carson o bserves that Leibniz says 42 Carson points out that Leibniz often objects that Locke is confusing idea and object or image and acknowledges that this is often because Leib niz is trying to make Locke conform to a model of knowledge and science that Locke would reject. 43 39 Surely, Carson does not mean to deny the latter point, but it seems as if she takes the alternative as theoretically desirable. Also, she seems to not realize that if this is problematic for geometrical ideas it must be also problematic for other mixed modes. 40 Leibniz also seemed to have this misconception; see Carson, 2007. 41 In addition to the following see Carson, 2002, 364 371, particularly 369. 42 Carson, 2007, 34. 43 ke more generally.
245 problem that is in 44 If geometrical ideas are imagistic (and to the extent that they are) they are inherently constrained by the spac e in which they are imagined, typically Euclidean. This, of course, is the same Carsonian objection discussed above but it is used here to point out a new difficulty for Locke. For Leibniz maxims 45 However, Locke has to rely upon experience to provide such constraints, and this seems to undermine the ideality of geometric ideas, which would undermine the distinction between i deas of modes and ideas of substances. This, again, is the same point, but here it is applied to the unwelcome result that triangle with interior angles totalin g up t o less (or more ) than 180 degrees. However, this is impossible, and Locke clearly grants this. As we have seen, the Lockean idea of triangle includes the fact that it is a plane figure, so it is restricted by the requirements of the idea of space or the a figure against a Euclidean background, the idea of that figure is not solely 46 She reinforces this emphasi do not depend only 44 Ibid., 35. 45 Ibid. 46 Ibid., 37, bold emphasis added.
246 space in which they are 47 This, Carson claims, undermines the Lockean position that the real and nominal essences of modes coincide. Although I have responded to much of this worry already and will return to the Leibnizian objection Carson develops in the next sect ion, the following further points should be made. I think there is a reason to not worry so much about the relative lack of arbitrariness involved in geometrical ideas. As we have seen, our simple combination s of various general ideas give us the criteria general idea of occupation gives us various requirements for some notion to count as such. Its simple combination must, for instance, include something like something an agent can d o or something for which one might be paid in some sense. Something like, say, a couch, does not meet these requirements, so it makes no sense to call it an oc cupation. Analogously, the simple combination s of any types of plane figures, like triangle mus t include that, plane figure to delimit the things that fall under those ideas. So, we h ave another reason to think our general ideas of shapes or figures are mixed modes; that is, they must, for instance, inc lude both a space bounded on three straight si des and plane figure which, as Carson observes, are not the same ideas. Now, if we are to assess whether a particular idea falls under a general idea, we have to see if the particular meets the criteria of the simple combination of the general. It is hard to see how this might be possible if the particular did not also include those elements, though in completely determinate ways. Carson points out that once we have two (straight) line segments of some lengths joined at some angle of our choosing, it is no t just an arbitrary decision of ours 47 Ibid., bold emphasis added.
247 lines, it is now determined by something other than the free activity of the mind how a closed figure can be formed from 48 Carson takes this other something to be the nature of Euclidean space, but the idea of a triangle includes this as well as the fact that a triangle is a plane figure. The particular idea of a triangle must be a plane figure a s well as a space bounded on three straight sides. 49 This is a constraint upon the idea of triangle, as Carson notes, but it is a constraint that applies just in case the idea under consideration is to fall under the general idea of plane figure which we normally take to be the case. Exactly this same sort of constraint applies to every other general idea, though, as we have seen in regard to occupation If our particular ideas are going to fall under general type ideas they must be constrained by the requirements, elements in the simple combination s of those general ideas. Citing Cicovacki, 50 hat 51 This suggests that we are still left with a worry about our being able to understand the ideas involved in geometrical objects adequately. Carson thinks we can have complete, adequate (in Lockean terms) understanding of our mixed mode ideas because they are entirely the workmanship of the understanding, but simple modes are not. In response to this worry, one might similarly worry about the adequacy of our mixed modes in so far as they are made up of ideas that are not just the workmanship of the 48 Carson, 2005, 23. 49 This seems to be a mixed mode, which, again, suggests that Locke was mistaken or confused in 2 13 50 Cicovacki (1990). 51 Carson, 2002, 370.
248 understanding. Surely mixed modes are often constituted, partly or wholly, by simple ideas and simple modes, which in turn are not at all arbitrary in the sense Carson and Locke use the term or onl y involve simple ideas. If our grasp of simple modes is problematic, then it seems that so must our grasp of mixed ones. 52 However, note that Locke takes the idea of triangle to be one that we are able to grasp most fully. 53 Rather than relying upon their b eing the workmanship of the understanding, Locke thinks we are able to fully understand our ideas when they are adequate and capable of being analyzed into component ideas which we are, in principle at least, able to understand completely (that is, they in clude only adequate ideas). Locke has given us a robust empiricist theory of how we come to grasp the ideas of space and number and Carson suggests no problem with that theory, though her concern over a lack of an account of the other something that determ ines the properties our geometrical ideas suggests that she does actually have such a problem. If she has a problem with what Locke says about how we come to have our idea of space in 2.4 or 13.2, she does not say what it is. Carson might worry about the f act that our idea of Euclidean space is not clearly explicated in Locke. Surely this is true, but it is unclear exactly how that worry might be developed, and I have (in Â§3.2) suggested how that particular idea should be understood to be available to Locke implicit, and seem s to flow from a Kantian viewpoint 54 of our ideas of space and number as satisfactory, there seems to be no reason to thi nk that geometrical ideas, whether simple or mixed modes, are problematic. 52 Note that Locke explicitly says that our ideas of both simple and mixed modes are adequate. See Chapter 6. 53 See, for instance, 3 11.22 and 23, as well as several other places in Book 4 54 Carson, 2006 footnote 13, 31 32.
249 So, to return to the argument presented at the beginning of this sub section, we now have reason to reject (1) and (2). Contrary to (1), Locke thinks we can have certain knowledge when the relevant ideas are capable of being completely and adequately understood, not when they are entirely of our own making. Contrary to the thought underlying both (1) and (2), none of our ideas are entirely of our own making in the sense that Carson seems to mean. As a result, Carson has not shown that there is a problem with the certainty associated with our mathematical and geometrical ideas. 7. esult that on latter, and the former has to go. However, I have shown that fro m clear that Locke would acknowledge the sense of apriority or necessity Carson (and Leibniz, actually) seem to want to ascribe to him. Also, the certainty we have from the adequacy of the simpler ideas from which we create our ideas of modes, simple as well as mixed. comes to. count of the instructiveness of pejorative. There seems to be nothing untoward about the inclusion of the constraints of
250 igure in our idea of triangle We can have instructive knowledge of the properties of a triangle because we might not have appreciated everything that follows from the things that we included in our idea. However, because it is the adequacy of the constitu tive ideas that underlies the certainty of our knowledge, we can consider those constituent ideas and their relations to each other and come to better, sometimes unexpected, instructive understanding. seems to fall apart, as one would knowledge of our ideas of mathematics though we can have it of our ideas of ethics. 55 This ence that geometrical ideas are simple modes that include the putatively problematic idea of Euclidean space and her position on the arbitrariness of New Essays 56 similarly, and systematically, runs afoul of the objections I have detailed above. It is to one particular Leibnizian objection Carson develops to which I turn in the next section. 7. 4 Another Carsonian Objection Working from a point made by Leibn iz, Carson objects that the distinction Locke attempts to make between mixed modes and substances is problematic. t hat that distinction i e their archetypes in the possibility of things, just as much as the idea of 57 Because we cannot be sure that we have a clear understanding of 55 Carson, 2 006 I return to this topic, and Carson, in the next section. 56 Carson, 2007. 57 Carson, 2007, 29. Carson goes on to ignore courage naturally, given her project, but, more interestingly, so does Leibniz. The remainder of that section of the New Essays uses strictly geometrical examples.
251 substances, to establish the possibility of some given substance kind (say, zebra ) we have to observe that there are such things out there in the world. How else could we establish that such a combination of features is actually possible given the underlying, and unknowable, substances in which those features inhere? However, Leibniz says this h olds true of modes as well, because we cannot tell a priori 58 what is actually consistent ( which Locke requires for the adequacy of possible geometrical figure, bu t which, it turns out, is not actually possible, at least in Euclidean space. We must engage in a bit of geometrical reasoning, or an attempt to construct a regular decahedron, to see that such a figure is impossible. The critical point for Carson, referen cing 4 6 .11 of the Essay is that: In contrasting the idea of gold with the idea of a triangle in claiming that one is formed arbitrarily or vo Locke does not take account of the fact that just as the former is subject to phys ical or chemical constraints, the latter must satisfy geometrical constraints, i.e. the constraints of Euclidean space. Determining that an arbitrarily formed representation does tre atment of the consistency condition glosses over. 59 I think Carson is missing the point here that while we can combine ideas to come up with the notion of a regular decahedron 60 it is from the component ideas, including that of plane figure in Euclidean s pace that we are able to find out the contradiction. Because the set of component ideas is limited to simple ideas and modes all of them are adequate, so we can discover that their combination is inherently contradictory. However, there is no such set of adequate ideas to work with in the idea of a substance. The critical idea of an underlying 58 Neither Leibniz nor Carson use this term, at least not in this part of the discussion. 59 Ibid., 31 32. 60 Calling it a representation in this case seems to invite an implicit imagism and, moreover, this image would be, as it turns out, impossible to have
252 substratum upon which the physical or chemical constraints depend, is notoriously in adequate, so it is not just difficult to perceive an inherent inconsistency it i s impossible. What Locke ideas of substances, as to know what real constitutions produce those sensible qualities we find in them, and how those qualities flowed from thence, we could, by the specifick ideas of their real essences in our own minds, more certainly find out their properties, and discover what qualities they had, or had not, than we can now by our senses (4.6.11). If we had a sufficiently clear, adequate, idea of what real c onstitution, or substratum, produces the qualities gold has, we could make surer claims, along the lines of those we make in geometry, about gold and other substances. However, we do have a very clear and adequate idea of (Euclidean) space Locke is implic itly, but fairly clearly, relying upon the difference between the adequacy of our ideas of substance as substratum and space to support the difference between ideas of substance and geometrical ideas. 61 While Carson does admit that Locke implicitly acknowledged the background assumption of Euclidean space 62 regarding the transparency that Carson thinks is derived from the construction of geometrical shapes, it seems to me that this conceptual transparency is ultimately derived from the adequacy of ev ery idea that is included in a given geometrical idea. It is that adequacy that, effectively, makes the successful construction possible. Be cause those ideas are adequate we can create proper geometrical shapes, and see that a regular decahedron is impossi ble. In a sense, the consistency Carson mentions above is transparent because the ideas upon which it is (ultimately) based are all transparent However, according to Locke, we can do nothing like that with ideas of substanc es, because the critical 61 See Chapter 6 (Â§1.2) above, for a discussion of the adequacy of our ideas. 62 Carson, 2007, 32.
253 idea implicitly included is that of substance as substratum not space and that idea is far from adequately, let alone transparently, understood. 63 tances, we have no knowledge of the natural connections responsible for any particular conjunction of 64 She i s a bit confused though; i t is not the connections, as s uch, but the substratum upon which they depend of which we have no knowledge. This is brought to the fore by the considerations against the adequacy of ideas involving substance discussed in the previous chapter. According to Locke none of those ideas is a dequate, so we cannot plausibly be said to have knowledge (at between ideas of modes and substances is arbitrary fails, largely be cause Leibniz and Carson have failed to truly grasp the actual basis for that distinction: the adequacy of the ideas that go into our ideas of geometry, and other mixed modes, and the inadequacy of at least one of the ideas that goes into our ideas of subs tance kinds. A closely related issue, which was mentioned in the previous section: 65 C arson thinks this is based upon the previous point, that the distinction between modes and substances that Locke makes is arbitrary. According to Carson, clearly brings 63 Carson seems particularly blind to the actual source of the problem with substance See also, Carson (2005 ) 4 5. 64 Carson, 2002, 369. 65 Carson, 2007, 38.
254 out the distinction between what Carson takes to be the workmanship of the understanding, which, as we have see n for her is purely a product of the mind, and what is brought on board as a necessary contribution of non can formulate definitions as it pleases, but the possibility of the thing defined parallel lines [for instance] 66 Carson is clearly right that this point is supported by the reasoning in the thinking behind the earlier part of this section However, it also runs afoul of the response I gave above and that Locke surely does not intend any of our ideas to be purely the creation of the mind. Note that he says our ideas of modes are Ideas (2.31.3), which are not something that the mind creates unilaterally, but are constrained by the roles they are to play ( as types under which we organize ideas and existents ). My selection of that passage in 2 31 was deliberate Carson goes to that section to give evidence that Locke does have in mind utterly arbitrary collections of ideas, and that he thinks space is not included in triangle Idea of a Figure, with three sides meeting at three Angles, I have a complete Idea wherein I require nothing else to make it perfect (2.31.3). However, as Carson has pointed out time and again, the idea of triangle must include the idea of (Euclidean) space as well. So, she reemphasizes her 67 Here, again, in 2 31 .3, there is the potential problem th at Locke is contradicting the position I have been defending. I believe I can dispel that worry. 66 Ibid., 39. 67 Carson, 2007, 43.
255 In this paragraph of the Essay complex Ideas of Modes (2.31.3), and how they are archetypes formed by the mind a gainst which we compare reality, not the other way around, because they are not meant to be copies of things out there, in the world. They do not include a reference to something like an underlying substance as substratum So that, in the above passage, Lo Idea of a Figure, with three simple combination of triangle 68 wherein his point is not that there is nothing else included in this idea, but that there is nothing mysterious included in that idea. In the next sentences, Locke elaborates just this point, that the idea of triangle contains everything we need to come to clearly understand all the possible properties of Ideas of Substances it is otherw ise (2.31.3). In that case we want our ideas to capture how and in what way things really exist, but we cannot pretend that we have the same level of ideational perfection. It is true that Locke says that the mind does not conceive, that any Understandin g hath, or can have a more compleat or perfect Idea of that thing it signifies by the word Triangle supposing it to exist, than it self has in that complex Idea of three Sides, and three Angles: in which is contained all that is, or can be essential to it or necessary to complete it, w here ever or however it exists (2.31.3). This seems to fairly explicitly say that space is not included in the idea of triangle I think this is moving too fast, though. First, as is so often the case, one must remember the context in which a sentence occurs, and here that context is about the relative adequacy of our ideas of modes and of substances, not about the actual content of triangle Idea 68 stand ins for the complete simple combinations of those ideas.
256 we can see that Locke is trying to simply say that the idea of triangle includes all that is essential to it. His point being that this is not true of ideas of substances. Now, it is possible that Locke was simply careless, or very likely, never considered non Euclidean triangles or parallel lines as options. At most this would be a problem for the various claims he makes about triangles in the Essay As I have labored to show, this is not at all a problem for the overall theory of mathematics that is presented, for our general ideas of things like triangles and parallel lines can easily accommodate the inclusion of the idea of space nts. To return to Carson, she observes: substances because their real essences and nominal essences are the same (and therefore the real essence is transparent to us) is simply wrong. In so far as the properties of the triangle flow from the idea of the triangle which is its real essence, that idea is spatial, an d not ideal. It has no epistemic privilege over ideas of substance. 69 Her reasoning is that because our geometrical ideas are not pure creations of the mind, but include the extra mental aspects of space, they are on no different footing than our ideas of 70 However, as I have argued, Carson and Leibniz are failing to take into account the critical difference between space and substance as substratum ; that is that the former is clearly understood and adequate, while the latter is neither. Further, it is not my position that Locke did take a theory of s pa ce into account, at least explicitly, but that this is not 69 Carso n, 2007, 43. 70 Ibid.
257 e. The understanding of ideas that I have been developing in this dissertation allows for the inclusion of plane figure in the content, or simple combination of ideas like triangle 7. 5 A Lockean Success In the foregoing, I have argued that Locke gives us a theory of mathematics (and geometry) that secures certain, not merely trifling demonstrative knowl edge, and that this follows from the adequacy of the ideas involved. I have also argued, in Â§3, that the most recent although Carson has managed to show th at there is a problem with whether Locke thought of, or should have thought of, ideas of, at least, geometrical figures as simple or mixed modes. Locke seems to have, at worst, made a mistake, to have, at best, been a bit confused at one point in the Essay or, perhaps most likely, was simply unclear. Regardless, I have argued that not only must geometrical ideas like triangle and circle be mixed modes, but that even were those ideas simple modes there is no apparent problem for the theory in question. I take the foregoing to constitute my prima facie mathematics to be successful. At the very least, it has given us good reason to think that Locke is successful in doing what he thought was necessary to explain the way our mathematical knowledge, as we normally understand it, to be grounded in the general theory of the Essay particularly regarding its treatment of, and reliance upon, the adequacy of the relevant ideas ics is the best we can do, or even merely a good way to think of mathematics, is clearly an entirely different question, one which I leave for future consideration.
258 CHAPTER 8 CONCLUSIONS The conclusions of the previous chapters can usefully be organized into three categories: those involving the positive account of the reading of the Essay I have been developing, those responding to critics of Locke, and those suggesting problems for him. In the following brief sections I will review each of these catego ries before saying a final word about questions that remain. 8. 1 My Reading of the Essay thoroughly imagistic reading of Locke on ideas. In spite of the argument Ayers offers, it makes better sense to think that, at least for purposes of making sense of the Essay Locke did not think that all of our ideas are mental images. Rather, we should think that Lockean ideas include both imagistic and non imagistic mental objects. In Chapter 3 I developed what I have called the simple combination reading of Locke on ideas. That is, each non simple, or complex, idea is just a combination of its consti tuent simpler ideas. 1 Each of the constituent ideas involved in the simple combination will be either simple or complex, and the latter will have its own simple combination. In this way, all complex ideas are reducible to simple ideas, whether those are im ages or not. I argued that this reading both makes good sense of what the Essay says and respects the other commitments Locke has regarding ideas. In Chapter 4 I argued that Locke should not be understood to have a resemblance based theory of representati on, but has instead a causal theory of representation for simple ideas, and a basic use theory for other ideas. That is, a simple idea represents (at least) its cause because of 1 I also suggested that even simple ideas can be thought of as having a basic simple combination though in a rather attenuated sense.
259 the automatic mechanisms involved in our getting the idea in the first place, but a complex idea represents what it does because we use the idea to so represent. Some of our ideas do resemble what they represent, but that resemblance is not what makes them represent. That relationship is something that is caused by the nature of our experience or is established by the understanding in making sense of that experience. In Chapter 5 I supported the position that Locke had a single understanding of abstraction in the Essay method of abstraction does not, contra Walmsley, require a problematic strict separation of ideas, but is perfectly compatible with a mere partial consideration of some complex thing or idea. Finally, in Chapter 7, I argued that the informativeness of ou r mathematical and geometrical claims comes from the nature of the ideas involved in those claims. Because those ideas are not simple modes, but mixed, and have simple combinations the entailments of which we might not have fully appreciated we can have b oth certain and informative knowledge of In general, I have endeavored to show that the Essay has a largely consistent and unproblematic understanding of ideas. It may not be the best way to think of our mental objects, which is notably not to reach a clear understanding of what is meant by idea He is instead far more interested, for instance, in the limits of hu man understanding and establishing guidelines to avoid the excesses of religious or skeptical enthusiasm. For example, Locke is concerned to emphasize that we can only have knowledge to the extent that we have clear and distinct ideas. Where we lack such i deas, as in the case of substance as substratum which Locke thinks is the archetypical confused idea, we do not have knowledge. As a result, we should be wary of
260 committing ourselves fully to claims involving such ideas. Similarly, the nature of our ideas and the fact that we receive the simple ones passively, suggest that a reasonable limit on skepticism is warranted. There must be something beyond our minds that serves as a source for these ideas. On the reading I have developed, and which I have argued is at the very least compatible with what Locke says in the Essay he seems to have what most of what he needs for these higher level projects, at least as far as his understanding of ideas is concerned. 8. 2 Responses to Critics In developing the foregoi ng reading, I have offered responses to a number of prominent General objections that follow from strict imagism are avoided by the distinction between imagistic and non s to Loc ke on abstraction, including those that hinge upon separatism have been shown to miss the mark by failing to theory provides the material needed to avoid problematic claims about ideas of sensation that Locke himself makes and there is no good reason to think that the abstraction presented in the Essay cannot provide us with ideas like male timepiece hero and philosopher although such ideas might themselves be confused (like the idea of substance as substratum ). On the other te ideas are unfounded, particularly because the important foundational idea of thing or existent is a simple idea which we are not only able to derive from experience but is forced upon us by that experience
261 n disarmed by an emphasis on the extreme simplicity of our simple ideas, as Locke understands them, and the distinction between imagistic and non imagistic ideas. certain o r informative, not both, rests on a misunderstanding of the nature of the ideas involved, which should be understood to be mixed modes, not simple, as well as on an implicit belief that some of our ideas can be entirely of our own making, which is antithet distinction between modes and substances is unfounded fails because it mistakenly emphasizes how we derive those ideas and fails to appreciate the import antly different content of those ideas. objections to externalist understandings of Lockean representation fail to gain sufficient traction. Of course, if the controversy between internalism and externalism is resolved to some degree, this point will need to be revisited. 8. 3 Two Remaining Problems However, the development of my position has emphasized one long standing problem for Locke and unearthed another. qualities seems unfounded, as commentators as far back as Berkeley and Leibniz have insisted. While it might make sense to think that ideas of primary qualities resemble som ewhat more fully those qualities in bodies than the ideas of secondary qualities do, it can hardly make sense to say that the first set resembles (full stop) their respective qualities while the latter does not. Neither set of ideas really resemble in the sense Locke seems to have in mind. It is not clear, though, how significant a problem this really is for the Essay
262 I have offered in Chapter 4 is sufficient, it seems that the distinction at hand does not really carry much weight after all. However, if it is necessary to adopt a resemblance theory of representation (perhaps, say, because the aforementioned debate between externalists and internalists goes against externalist theories), then it seems that th e failure of this distinction might well pose a depends upon resemblance after all, and it is not at all clear how we are to understand that resemblance, particular ly if we cannot make good sense of the distinction Locke himself makes between ideas that resemble and those that do not, it seems very hard to see how we can properly The fact that understanding Locke as having a resemb lance theory is so problematic ( and I have arg ued that it is very problematic) might well count as a factor in favor of reading Locke as an externalist in this regard. Fortunately, as I have argued, the reading I have advanced does seem to do the wo from this issue. our complex imagistic ideas seems to require that they be made up of simple idea s that are like the minimal components of pictures: small bits of color and extension, or pixels even. However, the explicit account of simple ideas seems incompatible with such components, which amount to very tiny images of their own. Rather than an imag e of a perceptual minima (which would not be account actually gives us things like is extended and has color To say the least, it is unclear how these latter simple ideas can serve as the basic building blocks of, at least, our imagistic ideas. At simplified
263 taxonomy of ideas, and that a suitably Lockean account might be develop ed that overcomes this issue. In the absence of this revised account, though, it seems that the Essay has the significant problem of failing to give an adequate explanation of how it is that an important class of our ideas (complex imagistic ones) actually further is one of the most interesting of the topics that I am obliged to reserve for future work. 8. 4 Further Questions In addition to the two problems discussed above, there have been a number of questions I have had to pass by but are worthy of further consideration. One of the most fundamental questions in considering the Essay is whether Locke successfully defended his anti innatism. I have deliberately presupposed a (hypothetical) affirmati reading developed in innatism as well as his reasons for rejecting the innatist position. 2 The discussion in Chapter 3 suggested that general simple ideas could be thought of as types (no t tokens). Combining this thought with the other issues raised by my consideration of simple ideas (see abo ve), presents an opportunity to productively re examine the nature and role rs. A set of important questions suggested by the discussion of Chapter 4, on representation, involve a more thorough understanding of causal theories of representation. In particular, how 2 See Begging Status of the Anti Nativist Philosophy and Phenomenological Research Vol 69, No. 1 (July, 2004), 37 64, for a worthwhile starting point on this discussion.
264 Lockean representation works with his blank effect reading of Lockean ideas; and how should we understand more traditional causal theories of representation when it comes to a general application to Lockean ideas? While I have offered my own account for t he latter, I have left spelling out how other causal theories might be understood to do this for future consideration. mathematics and geometry in Chapter 7. For instance, s mathematical propositions as analytic or synthetic, in a Kantian sense? More fundamentally, is the subject ? This question seems to invite a more detailed look at the early modern empiricist understanding of mathematics and how we should understand that general project in light of contemporary positions on the subject. morality? Loc too brief comments in support of such a science have historically inspired skepticism among commentators as well as outright objections. Given the reading of Locke on ideas that I have developed, and, in particular, the understanding of demonstrat ive scientific knowledge discussed in Chapter 7, though, it seems that a careful consideration of the viability 3 3 (Oxford: Oxford Univ ersi ty Press, 2012), as well as Think Vol 10, No.28 (June, 2011), 77 87, and The Review of Politics Vol. 73, No. 4 (Fall, 2011), 581 608.
265 LIST OF REFERENCES Aaron, Richard I. Aaron John Locke Oxford: Clarendon Press, 1971 Aydede, Murat Language of Thought Hypothesis. The Stanford Encyclopedia of Philosophy https://plato.stanford.edu/ entries/language thought/, 2010 Ayers, Michael R. Locke, Vol. I: Epistemology London: Routledge, 1991 Baxter, Donald L. Philosophy and Phenomenological Research Vol. 57, No. 2 (June 19 97): 307 330. Berkeley, George Principles of Human Knowledge ed. Howard Robinson Oxford : Oxford University Press, 1996 Bolton, Martha Brand t e of Simple Ideas of Pacific Philosophical Quarterly 85 (2004): 301 321. Carson, Emily . British Journal for the History of Philosophy Vol. 10 No. 3 (Aug. 2002): 359 378. Locke Studies Vol. 5 (2005) : 19 38. Intuition and the Axiomatic Method ed s Emily Carson and Renate Huber Dordrec ht: Kluwer Academic Publishers (2006): 3 21. Locke Studies Vol. 7 (2007): 21 46. Chappell, Vere. The Cambridge Companion to John Locke ed by Vere Chappell Cambridge: Ca mbridge University Press, 1994: 26 55. Pacific Philosophical Quarterly 85, 3 (2004): 338 335. Cicovacki, Predrag Journal of the History Philosophy Vol., No. 4 (Oct. 1990): 511 524. De Rosa, Raffaella Begging Status of the Anti Nativist Arguments Philosophy and Phenomenological Research Vol. 69, No. 1 (July 2004): 37 64. Forde, Steven. The Review of Politics Vo l. 73, No. 4 (Fall, 2011): 581 608. Gibson, James Ethics Mind Vol. 5, No. 17 (Jan. 1896): 38 59.
266 . Cambridge: Cambridg e University Press, 1960 (1917) Goldberg, Sanford C. ed. Internalism and Externalism in Semantics and Epistemology Oxford : Oxford University Press, 2007 Hacking, I an. Why Does Language Matter to Philosophy? Cambridge: Cambridge University Pres s, 1975 Hanna, Robert Theory The Review of Metaphysics Vol. 44, No. 4 (June 1991): 775 805. Hume, David A Treatise of Human Nature ed s. David Fate Norton and Mary J. Norton Oxford : Oxford University Press, 2000 Jacovides, Michael. The Philosophical Review Vol. 108, No. 4 (Oct. 1999): 461 496. Jesseph, Douglas M. Philosophy of Mathematics Chicago: The Un iversity of Chicago Press, 1993 Kroll, Journal of the History of Ideas V ol. 45, No. 3 (July Sep. 1984): 339 359. Leibniz, G.W. New Essays on Human Unde rstanding trans. and eds. Peter Remnant and Jonathan Bennett Cambridge: C ambridge University Press, 1996 Lennon, Pacific Philosophical Quarterly 85, 3 (2004): 322 360. Locke, John An Essay concerning Human Understanding ed. Peter H. Nidditch Oxford: Clarendon Press, 1975 . The Locke Newsletter 2 (1971). LoLordo, Antonia LoLordo. Philosophy and Phenomenological Research Vol. 77, No. 3 (November 2008): 705 724. . Oxford : Oxford University Press, 2012 Lowe, E.J. Locke on Human Understanding New York: Routledge, 1995 Ludlow, Peter Ludlow and Norah Martin, eds. Externalism a nd Self Knowledge Stanford: CSLI Publications, 199 8 Mackie, J.L. Problems from Locke Oxford: Clarendon Press, 1976 Moore, Terence Think Vol. 10, No.28 (June 2011): 77 87.
267 Ott, British Journal for the History of Philosophy Vol. 20, No. 6 (2012) : 1077 1095. Pappas, George S. Ithaca, NY: Cornell University Press, 2000 Sextus Empiricus. Outlines of Skepticism eds. and trans. Julia Annas and Jonathan Barnes Cambridge: Cambridge University Press, 20 00 Shapiro, Lionel Shapiro. Pacific Philosophical Quarterly 91 (2010): 554 586. Stuart, Matthew . British Journal for the History of Philosophy Vol. 1 6, No. 3 (2008): 511 533. Tipton, The Empiricists: Critical Essays on Locke, Berkeley, and Hume Margaret Atherton ed. La nham, MD: Rowman & Littlefield, 1999: 1 18. Walmsley, Jonathan Abstraction: a Response to M.R. Ayers British Journal for the History of Philosophy 7: 1 (1999): 123 134. Warnock, G.J. Berkeley Notre Dame: Unive rsity of Notre Dame Press, 1982 Wolfram, Sybil Instructive Distinction with the Analytic Synthetic Distinction. The Locke Newsletter 9 (1978). Instructive Distinction A Reply. The Locke Newsletter 11 (1980). Wright, Crispin, Barry C. Smith, and Cynthia Macdonald, ed s Knowing Our Own Minds Oxford: Clarendon Pre ss, 1998 Yolton, John. Locke and the Compass of Human Understanding: A Selective Commentary on the Cambridge: C ambridge University Press, 1970 . Perceptu al Acquaintance from Descartes to Reid Minneapolis: University of Minnesota Pre ss, 1984 . A Locke Dictionary Oxford: Blackwell, 1993
268 BIOGRAPHICAL SKETCH Ron Claypool was first introduced to modern philosophy as an undergraduate at the to understand the material and present his thinking in a clear, rigorous man ner. By the end of the course, Ron was hooked on philosophy and fascinated by the British empiricists. In December 2006, Ron graduated from USF with a BA, magna cum laude, and a double major in history and philosophy, and set his sights on pursuing his PhD in philosophy. At this time, Ron focused his independent study on the work of John Locke and David Hume, becoming interested in their understanding of mathematical ideas, or concepts. When he was accepted into the graduate program at the University of Fl orida in 2008 it was with the intention of working on the philosophy of mathematics of the early modern British empiricists. t became increasingly apparent understanding of ideas was an important first step in developing a broader understand of the dissertation Locke on Ideas and his PhD in philosophy, granted in the summer of 2017.