LOCKE S EMPIRICAL THEORY OF COGNITION: THE THEORY OF IDEAS Let us then suppose the mind to be, as we say, white paper, void of all characters, without any ideas; how comes it to be furnished? Whence comes it by that vast store which the busy and boundless fancy of man has painted on it with an almost endless variety? Whence has it all the materials of reason and knowledge? To this I answer, in one word, from experience. In that all our knowledge is founded; and from that it ultimately derives itself. Our observation employed either, about external sensible objects, or about the internal operations of our minds perceived and reflected on by ourselves, is that which supplies our understandings with all the materials of thinking. These two are the fountains of knowledge, from whence all the ideas we have, or can naturally have, do spring. Locke, An Essay Concerning Human Understanding (1690) As we ve discussed, Locke wants to develop a theory of the human mind; he wants to help us understand its nature, powers and capacities. 1 And we know that he wants to do so in a way that satisfies the empirical constraint: namely, that such an explanation must ultimately be understood in terms of the deliverances of sensory experience and the various mental operations of which the mind is capable. And as we have discussed, he seeks to motivate this empiricist program in two stages: first, he attacks the truth of the alternate theory, nativism; second, he argues that an empiricist theory of human cognition is better equipped to explain the nature, function and contents of human cognition over any form of nativism. Now that we have looked at his attempt to execute the first stage, it s time to try to understand his attempt at executing the second stage. He does this by offering and defending his positive theory, his so-called way of ideas. But note that in order to do this, he must develop a theory regarding the nature of the human mind and its operations that satisfies two desiderata: (i) it must not only help us understand how the mind works in a way that is not itself plagued by theoretical problems; (ii) it must also be able to appropriately handle issues regarding our actual mental content. Regarding (i), he understands that he cannot say everything there is to say on the matter. No, he expects that it will take the work of other empiricists to help develop a complete and satisfactory theory. Even so, we will see that the version of the theory that he does offer has been attacked in substantive ways (both by antiempiricists, e.g., Leibniz, and empiricists, e.g., Berkeley, Hume, and Reid). Regarding (ii), it is important to keep in mind what is at stake: his account must either (i) explain the nature, content and means by which we come to have the concepts and general propositions we have and understand, or (ii) explain why we think we have such things despite the fact that we do not. Can he do that? What does that theory look like? Locke s way of ideas As we know, being an empiricist, Locke takes it that we can trace the origin of a human being h s thoughts to h having had sensory experiences of the relevant type. But, before he can work out the details of such a theory for us, he must first answer the question what are thoughts? Whatever they are, clearly they are mental states, i.e., states of the mind. But notice that that is not very helpful; while all thoughts are mental states, not all mental states are thoughts (e.g., pains and itches are not thoughts). So, what makes some mental states thoughts and not others? What is the relevant difference? To help answer this, let us consider the following English sentence: APPLE. The red apple I am eating is sweet and juicy. 1 Recall that his interest is the discerning powers of a man, not the physical considerations of the mind (Essay Introduction, 2).
What is it to think APPLE? The first thing we can say is this: thoughts such as APPLE are intentional states, that is, they are mental states that are about things, they have contents.2 Since thoughts have this unique feature, we must now ask: what are the things that our thought states are about? In other words, what are the objects of thoughts? Locke answers this with his theory of ideas; he tells us that Idea is the object of thinking, and that which [a person s] mind is employed about whilst thinking being ideas (Essay 2.1.1). So, according to the theory of ideas, ideas are the contents of a thought. In other words, to have a thought is to have a mental state whose object is an idea. But what does this mean? What exactly are ideas? This is what Locke seeks to explain in the Books II-III. Before we go forward, it will be helpful to make a distinction regarding two kinds of thoughts: object thoughts and propositional thoughts. Some of our thoughts seem to be of the former type, some the latter. To make this distinction, however, we need to distinguish two kinds of contents (again what Locke calls ideas): de re content and de dicto content.3 Notice that, in one sense, to think APPLE is have a mental state with de dicto contents: it is to entertain and take up an attitude toward the truth of the proposition. In other words, it is to consider and decide whether or not APPLE is true, whether or not it correctly describes the world. In such a case, to think APPLE is to have the proposition as the object of thought. Yet, in another sense, to think APPLE is to have a mental state with de re contents. For thinking APPLE seems to be more than a thought about APPLE s truth-value. Among other things, it also seems to involve thinking about the very apple itself that is the subject of APPLE. For it seems that in order to think APPLE is true or false, a person must also think about the referent of the subject term of APPLE, namely, the particular apple in question (call that apple Adam ). If so, then thinking APPLE is not just to think that it is red, that it is sweet, that it is juicy, that it s being eaten; it is also to think about the apple itself, to have it as an object of thought. Given this, we can now return to the distinction between object thoughts and propositional thoughts. The thought about the apple itself is an object thought (it has de re content); the thought about the truth of the proposition is a propositional thought (it has de dicto content). Given this, it should be clear that in order for Locke to develop a theory of the human mind on the basis of the claim that ideas are the objects of thought, he must develop a theory regarding both varieties (assuming that we can have both types). And this is exactly what Locke does. As we will see, Locke will develop of his theory of ideas as thoughts are about ideas by first developing a theory about object thoughts followed by a theory about propositional thoughts. Let us first focus our attention on his theory of object thoughts. As we will see, Locke s theory of ideas seems to be a form of representative realism.4 Before we state Two things. First, the concept of intentionality is not the same as the concept of intensionality. The former is a concept in the philosophy of mind; the latter is a concept in philosophical logic and semantics. Second, this is why pains and itches are not thoughts. Such mental states are mere sensations; they are not about something. For instance, Jones does not itch about something; she itches because of something. And Smith does not feel pain about something; he feels pain because of something. Of course, a tricky case is emotional pain. These pains (at least sometimes) seem to be about something. For instance, if someone asks Smith why are you sad (a kind of pain)?, it seems perfectly reasonable for Smith to reply with something like I m hurting over what Brown said to me. But is this pain really about something? Or is it pain that happens to be accompanied by Smith thinking to himself Brown said such-and-such to me and that hurts my feelings? If the latter, then Smith s pain is not strictly speaking about Brown s actions; it s caused by those actions and accompanied by thought states. 3 The term de re is Latin for concerning the thing; the term de dicto is Latin for concerning the proposition (or literally expression). 4 This has, and continues to be, the common interpretation of Locke s theory of cognition. Even so, some Locke scholars, e.g., EJ Lowe, argue that Locke actually rejects representative realism in favor of a form of direct realism. 2
what that means exactly, let us start with a few quotes. Whatsoever the mind perceives in itself or is the immediate object of perception, thought, or understanding, that I call idea (ECHU 2.8.8). But notice that that is not very precise. For it does not tell us about the relation between our thoughts about things and the ideas we are said to have. Says Locke, more exactly the mind knows not things immediately, but only by the intervention of ideas it has of them (ECHU 4.4.3, emphases mine). So, when you or I have a thought, Locke s passage seems to suggest, we are not thinking directly about that thing, we are thinking indirectly about it. Let us read a couple more quotes to get in mind what Locke seems to mean. According to the Scottish philosopher, Thomas Reid, Locke s view is thus: [Lockean] ideas are not the operations of any mind, but supposed objects of those operations. They are not perception, remembrance, or conception, but things that are said to be perceived, or remembered, or imagined (Essay on the Intellectual Powers (1785) 2.14, emphases mine). Moreover, says Reid, Locke s view commits him to the following position: [Although John Locke believes that] there are substantial and permanent beings called the sun and moon; they never appear to us in their own person, but by their representatives, the ideas in our own minds, and we know nothing of them but what we can gather from those ideas (ibid., emphasis mine). Given these passages, we can say the following. Given that to think APPLE is to think about Adam and think that Adam is sweet and juicy, it seems that Locke would say that when Jones thinks APPLE, Jones has an idea of Adam, an idea of sweet and juicy, and an idea that Adam (actually) has those properties. But the passages also suggest that Jones is not strictly thinking about Adam, those properties and its having them; he is thinking about his idea of Adam, idea of sweet, idea of juicy and idea of Adam having those properties. He does not think about Adam immediately, but only by way of thinking about ideas. Let s see if we can make this more precise. Representative realism teased out As noted, the above quotes suggest that Locke is committed to a representative realist theory of human cognition. What is that theory? It is a conjunction of two different, but important philosophical theses: metaphysical realism and representationalism. Let s begin with the former. Metaphysical realism is a theory about the nature of reality. Let us understand it as the theory that MP-REALISM. There is at least one entity e such that e exists and e has (at least some of) its properties mind-independently. As you can see, the metaphysical realist thinks that there is at least one probably many entity that exists and has (at least some of) its properties independent of every cognitive being and their thinking about that entity (for instance, that apples have a shape whether or not there are any cognitive being that exists). 5 What is representationalism? 5 Most people ascribe to some form of metaphysical realism. For instance, most of us ascribe to a view on
Representationalism is a theory of cognition, it tells us what it is for humans to have intentional states, i.e., mental states that are about things (e.g., thoughts). It purports to explain how the human mind accesses such things as the inhabitants and goings-on in the external world. As the Locke quotes say: no human has immediate direct cognitive access to objects; that all cognitive access is mediated by some mental entity. And following Descartes, Locke takes these to be ideas. Let us make representationalism more precise, however. We will understand it as the theory that states that RTC. A person S thinks about the F-ness of an object O only if: (i) S has an idea i, (ii) i represents the F-ness of O, and (iii) S is directly acquainted with i. So, according to RTC, unless a perceiver satisfies conditions (i) through (iii), she won t perceive any entity. For instance, RTC tells us that if Jones perceives the shape of Adam, then following must be true: Jones has an idea that represents the shape of Adam, and Jones is directly acquainted with the idea of the shape of Adam. In other words, unless Jones has that idea and the right relations occur, Jones cannot perceive (whatsoever) the shape of Adam or any other property for that matter. Locke s empirical version of RTC So far, RTC only tells us what it is for Jones or any human to have a thought about an object. It does not tell us about the source of the idea, e.g., the source of Jones idea of the shape of Adam. Thus far, Descartes a nativist would agree with RTC. So, given that Locke is an empiricist, we need to add to the analysis to make RTC unique to empiricism. To do so, we can say this. Empiricists brace what has come to be called the copy principle, namely, the claim that COPYPRINCIPLE. All ideas are copies of experiential impressions is true. 6 Given COPYPRINCIPLE, we can amend RTC and make it unique to empricism as follows: RTC*. A person S perceives the F-ness of an object O iff: (i) S has an idea i, (ii) i represents the F-ness of O, (iii) S is directly acquainted with i, (iv) O s being F caused S to have an experiential impression i*, and (v) i is a copy of i*. Notice that not only does RTC* include two new clauses i.e., (iv) and (v) but it also says that (i)-(v) are not merely necessary conditions. No, they are jointly sufficient; Jones satisfying all of them entails that Jones in fact perceives the properties of the objects in question. So, given RTC*, we can now amend the description of Jones having an idea of the shape of Adam to include the distinctly empiricist the nature of matter that not only that it really exists, but that it has various properties whether cognitive beings exist or not, e.g., it s composed of atomic parts in various lattice structures, etc. As we will see, Berkeley and possibly Hume have deep philosophical reservations about the coherence a commitment to both empiricism and metaphysical realism. 6 Alternatively, we could state COPYPRINCIPLE as COPYPRINCIPLE*: Unless S has an experiential impression of the relevant type, S cannot have an idea.
component of the explanation, namely, Jones perceives the shape of Adam iff the following is true: Jones has an idea that represents the shape of Adam, Jones is directly acquainted with the idea of the shape of Adam, the Adam being so shaped is the cause of Jones idea since Adam being so shaped caused Jones to have a sensory impression and Jones idea is a copy of that impression. Locke s distinction of kinds of ideas Let us now spell out exactly how ideas function so as to represent the world to our minds. According to Locke and the empiricists, in addition to COPYPRINCIPLE, the following principles are true: (a) There are only two types of ideas: simple ideas and complex ideas, and (b) All complex ideas are composed from simple ideas. 7 Given clause (i) of EMPIRICISM, we see that COPYPRINCIPLE and principles (a) and (b) make explicit the empiricist s position on how it is we come to acquire our mental content, e.g., concepts, beliefs, knowledge, etc. Accordingly, anything that is in principle incapable of being acquired through some experience cannot be included in the mental content we have (even if we think otherwise!). As such, this tells us that unless we have sensory impressions at all or of the relevant type, we cannot have the simple ideas that make up our complex ideas; and if we cannot have simple ideas, then we cannot have complex ideas. For instance, since Jones idea of Adam is a complex idea, Jones had to have the relevant simple ideas (such as the idea of the colour red, the idea of Adam s shape, the idea of juicy, the idea of sweet, the idea of solid, the idea of being extended (in space), etc.). How do complex ideas come to be? They are invented by the imagination; not surprisingly, the process is called compounding. Locke s criterion of success: resemblance As we ve seen, Locke takes it that ideas are entities in the mind that represent qualities in objects. While we ll say shortly what he say s qualities are, whatever they are, our ideas represent those. Simple ideas represent individuated qualities; complex ideas represent things as unified bundles of qualities. They do not represent the bearers of the qualities. (We ll look at his theory of substance later.) But what is the criterion for success according to Locke s representative realist view? How does an idea successfully represent a quality in an object? Here s what he says: RESEMBLANCE. An idea i successfully represents a quality F of an object O iff: (i) O is F, (ii) O s being F caused i to resemble F, and (iii) i resembles F. So we can see that Locke thinks in terms of a resemblance relation between ideas and the qualities of the object that the idea represents. When and only when an idea is so related will it be a successful idea. As we ll see next, however, most of our ideas are never so related. This, argues Locke, is because they cannot be so related. This is drawn out easiest in his primary secondary quality distinction. The primary-secondary quality distinction Following Boyle and Descartes, Locke takes it that there are two kinds of qualities: primary qualities (PQs) and secondary qualities (SQs). 8 Let us begin by giving Locke s definition of quality. According to 7 This is often called the compositionality principle. 8 Locke also discusses tertiary qualities, the relational capacity of a body to affect (and be affected by) the
Locke, quality is defined as follows: QUALITY. Q is a quality = df. Q is the power of an object to produce ideas. So, according to QUALITY, to say that object o has a quality F is to say that o has the power to bring about F-like ideas. On primary qualities (PQs) Says Locke, PQs are the real or intrinsic qualities in material bodies. That is just to claim that they exist in the world mind-independently. Examples of PQs are size, shape, motion, rest and solidity. Given QUALITY, Locke is claiming that an object s having a size is to be understood as the object s power to bring about an idea of size in a cognitive being. What is more, says Locke, in some cases, our idea of an object s PQs conforms to the object s PQs. What does conformity between an idea and a quality turn out to be? Such conformity is understood as follows: CONFORMS. A person S s idea of a quality q conforms to the q-ness of an object o iff S s idea of q resembles the actual q-ness of o. For instance, Jones idea of Adam s size conforms to the Adam s size just in case Jones idea resembles Adam s size. On secondary qualities (SQs) According to Locke, SQs are not like PQs: they are not real or intrinsic qualities in material bodies. That is, they do not exist in the world mind-independently. Rather, SQs exist only if there are minds. Examples of SQs are colors, tastes, odors, sounds, how a body feels to the touch, etc. Locke tells us that there are two important features of SQs. First, SQs are the causal powers of material bodies to bring about certain sensory ideas in perceivers. Second, SQs are the result of the PQs of a material body. On this view, material objects don t really have SQs (e.g., colours, temperatures, etc.). Given these two things, our ideas of SQs never conform to the actual qualities of the objects. Why? Simply given that our ideas of SQs resemble neither the causal powers nor the PQs of material bodies. Given that the distinction is now clear, we need to ask why Locke and others think the distinction is a genuine one. Consider the array of arguments that Locke offers in favor of the distinction. Locke s arguments for the PQ-SQ distinction Locke s argument from the scientific image (i.e., the world according to our best scientific theories). Locke argues that PQs are necessary for scientific theorizing, that is, there is no way to give an accurate and complete description of the spatial-temporal universe without PQs. This is not true with respect to the SQs, however. Scientific theories can provide an accurate and complete description of the spatialtemporal universe without any mention SQs. For instance, a physicist can give an accurate and complete description of the spatial-temporal universe in terms of physical entities, processes and mechanisms without ever having to use colour terms (such as red, green, blue, etc.), temperature terms (such as hot, cold, tepid, etc.), sound terms (such as loud, soft, etc.), etc. Of course, whether he s correct is a matter of considerable debate, i.e., whether it s accurate and complete. properties of another body.
Locke s arguments from the manifest image (i.e., the world as it appears to us in everyday sense perception). In addition to his argument from the scientific image, Locke gives a number of arguments from the manifest image. These purport to show that there is an important difference between the appearances and reality. In other words, his arguments purport to show that the mere fact that things appear to human percipients as being F (e.g., being coloured, being hot or cold, being flavored, etc.) does not imply that they are F. As a matter of fact, there are some properties that we experience that we would never take to be in the object, e.g., consider being in pain. Arg1. The pain/nausea case. Although being stabbed by a knife causes pain and eating too much manna (bread) causes nausea, there is no good reason to say that pain is in a knife or nausea is in the manna. No, these are merely effects produced by the things and their primary qualities when in contact with our bodies. And given that the SQs are relevantly analogous to pain and nausea, there s no reason to believe that SQs are in objects. 9 Arg2. The fire case. Although getting too close to a fire causes pain, there is no good reason to say that pain is in the fire. Given that temperature (an SQ) and pain are relevantly analogous, there s no reason to believe that temperature is in the fire. Arg3. The buckets of water case. Imagine Jones places his left hand in a bucket of water that registers 50 F on a thermometer for thirty seconds (call this bucket A ) and his right hand in a bucket of water that registers 110 F on a thermometer for thirty seconds (call this bucket B ). Jones left hand will feel cold in A and his right hand will feel hot in B. After thirty seconds, Jones places both hands in a bucket of water that registers 75 F on a thermometer (call this bucket C ). C will feel hot to Jones left hand and cold to Jones right hand. So, the water in C will feel both hot and cold. But C is not hot and cold. It has one thermometer reading. We can explain this, says Locke, by the fact that all that has occurred is that Jones nerve endings were affected by the water in C differently given that his nerve endings in his left and right hands had been affected in a certain way by the water in A and B, respectively. Arg4. The color in darkness case. If Smith looks at an object that is coloured (a ripe lemon, say), and takes the object into a completely dark room or subject it to green light, the object does not change. All that changes is what Smith can perceive of the object. Arg5. The almond case. Take an almond. It appears brown on the outside and white on the inside. If you begin to crush it, eventually the pieces will become such that it is neither brown nor white, but some other color: grey. Of course, every piece still has a shape, a degree of solidity, a size, etc. What explains this, asks Locke? Simply this: while the PQs are retained through division, the almond acquires new SQs. What are those new SQs? The particular effects on the perceiver: namely, having a certain taste, a different colour, etc. The implications of the PQ-SQ distinction Accordingly, we have two problems we need to explain. First, since most of our declarative sentences predicate SQs of objects e.g., that roses are red, that violets are blue most of what we say is false. Locke recognizes this; he is committed to an error theory about SQs, namely, the view that 9 By relevantly analogous I mean this: two things, T1 and T2, are relevantly analogous iff T1 and T2 are importantly similar or alike in various ways that are relevant to the issue in question.
ERROR THEORY: All statements that include SQ terms are false. The second problem is this: although we think our sensory ideas correctly represent the world as it is, they do not. It certainly appears as if tulips are yellow and honey is sweet, but according to Locke, our sensory appearances do not conform to reality. This is especially worrisome given that most of the properties that we take objects to have are SQs. So, how do we explain the fact that a perceiver s sensory ideas of SQs purport to resemble objects, but in fact do not resemble the actual properties in the material body? Says Locke, we can given the dispositional theory of colour. According to the dispositional theory of colour (hereafter, DTC), grass does not have the property of being green. Grass simply has various PQs that bring about SQs. In other words, although it is false that grass is green, what is true of grass (as well as other objects that appear to us as being green) is this: GREEN: x is green iff x possesses the power/disposition, by virtue of the primary qualities of its microstructural parts and surface spectral reflective properties, to produce in normal human percipients in standard lighting conditions a type of sensory experience that we refer to using the English expression green. The same follows for any other colour, as well as flavors, temperatures, odors, sounds, etc. So, according to DTC, there can be no green unless three things are true. First, there is an object that has the relevant PQs. Second, there must be a normal human percipient in standard lighting conditions. Third, (while this is not explicitly stated in GREEN), there must be some mechanism by which the human percipient is comes into visual contact with the relevant object. If any one (or more) of these three things are not actual, then, according to DTC, there will be no green (at least not for human beings). One final thought about DTC Notice, DTC does not say that the English word green is defined by GREEN. It is an open question as to whether green is so defined. No, at most, DTC tells us what is true of an object that appears green to a normal human perceiver in standard lighting conditions, namely, it has the described disposition. Whether or not green is defined by GREEN is another question altogether. For it may turn out that GREEN is true, but green is not so defined. Notice that if green were defined by GREEN, then most English speaking people in the world would not know what the word green means. For most English speakers know nothing of the physical theory of colour. Of course, most English speakers perfectly understand what we mean when use the word (even little children!). So, given that it seems that most English speaking people in the world do seem to know what the word green means, it seems as though GREEN is not the definition of green. Of course, that is not a knock down, drag out argument against making the definition of colour terms a dispositional property of objects. Maybe colour terms are so defined. But it is certainly something that anyone who thinks that colour terms are so defined must be able to account for or explain away. 10 10 2014, Richard G. Graziano. All rights reserved. This material may not be used, or duplicated in part or whole without express written permission by the author.