Skip to content


How the problem of knowledge arises

Locke's common sense - and ours - derives from an adherence to the distinctions enshrined in the accepted usage of the language we speak. Thus, for example, Locke knows, as we know, that to answer the question - "How do you know that what you say is true?" - you must, if you are to follow the accepted practice in such matters, present your evidence in one of two ways: 

If you are putting forward a statement about the world, a synthetic statement (e.g., 'The grass is green.') you must present empirical evidence (show that the way things are in the world in fact corresponds to what you have said about it). If you are making an analytic statement, e.g., 'All bachelors are unmarried males.' you would present linguistic evidence (show that the rules governing the use of the terms you have used have in fact been followed in this case). 

There is then no 'problem of knowledge' for common-sense: the rules for determining whether someone knows that what they say is true, are perfectly straightforward and everyone who understands how to use the term 'knowledge' also knows how to apply these rules. It follows that if there actually is a problem concerning knowledge it must concern the fact that there is some barrier to the actual application of these common-sense rules, some difficulty, for example, in determining what the world is, in fact, like, or what the rules governing the use of terms in our language actually are. In the latter case there is usually not a problem. In effect, what the majority says, goes, and finding this out is simply a matter of listening to how people use the terms in question. Thus a child may be inclined to think that a whale is a fish and among his peers this is the accepted usage. As he grows up, he learns a new usage, viz., that a whale is a mammal, and consequently comes to regard this classification as the correct one, on the basis of accepted usage among the majority of adults. 

However, with regard to the former case (the difficulty of determining what the world is like, and thus being able to determine whether what we say about it is true) there does seem, at least, to be a genuine problem. In the analytic case the evidence about how other people use their language is easily obtained and unequivocal. If in doubt I simply ask around. But in the case of determining what the world is actually like, a systematic difficulty arises, once it is realized, following Descartes, that the evidence of our senses is only evidence for the way the world appears to be, not for the way it actually is. Furthermore, everyone is in his own 'dark room' and thus, in the nature of the case, this predicament cannot be circumvented by 'asking around'. This would be useless since there is no way of determining whether the 'resembling pictures' which appear in your dark room are at all like those that appear in mine. 

The objectivity of empirical knowledge is thus systematically put in doubt, a doubt which according to the theory which raises it, will be impossible to assuage. How the problem becomes a problem for common sense How is common sense to deal with this problem? Its premiss is that the evidence of our senses is valid evidence for the way things are in the world. This premise has been challenged by the Cartesian theory. The natural response of the advocate of common-sense is to challenge the theory, but the difficulty here is that the Cartesian theory is being presented as no more than a newly discovered part of common-sense. It has always been a part of common-sense that appearances can be deceptive and as Descartes pointed out, if we can, in theory, be deceived under all circumstances, how can we determine under what circumstances we are not being deceived?1 

We seem then to have a situation in which common-sense is displaying an inherent contradiction. The senses are, on the one hand, regarded as providing us with evidence of the way the world is, and on the other, of being capable, on any given occasion, of providing us with false evidence of the way the world is. But the whole virtue of common-sense (the whole virtue of the discriminations which common usage indicates that we are capable of making) is its overall consistency. The very fact that a given usage exists is all the evidence we need to prove that people can play the language-game2 in question. In effect, there cannot be inherent contradictions in common usage since this would amount to saying that a given discrimination both can and cannot be made. Now in fact, the language-games of veridical perception and deceptive perception exist side by side, and therefore there must exist adequate criteria for telling when a perception is veridical and when it is deceptive. It would seem to follow, quite straightforwardly, that any apparent inconsistency indicated by the existence of these two seemingly contradictory uses of our language, could easily be revealed as illusory by simply examining the criteria which govern the language-games which seem, at first glance, to be in conflict. This is, of course, what common sense would advise us to do. This advice is backed up by a powerful sanction: if you do not resolve the apparent conflict, if you do not come to realize that the rules governing the two language-games which are, apparently, in conflict, are, in fact, quite consistent with each other, then one or other of the language-games will have to be abandoned as 'unplayable' and the consequence of this is that many other related language- games will also become unplayable.

 A Digression concerning Descartes' Evil Genius argument Before we go any further we should note that the Cartesian argument that leads Locke into difficulties is not valid. This argument claims that we are sometimes deceived by our senses and since we might always be being deceived (by the evil genius), we can never know for certain what the world is really like. It concludes that we can only know about things in the outside world in terms of our ideas (=perceptions) which amounts to saying that we can have no certain knowledge of what things in the world are like. The new 'way of ideas' restricts our knowledge to our ideas. This argument is invalid because it overlooks the fact that in order to talk about being deceived by my senses I must understand what it would be like to have a sense experience which was not deceptive. Thus how did Descartes ever get the idea that his senses were deceiving him? Only, we must presume, by being undeceived on some occasion, by seeing, for example, that something that he thought was green was really blue. We all know that appearances can be deceptive and we know what to do about it. E.g. I have tried on a blue suit in a store with fluorescent lighting. It seems a pleasant shade of blue but - glancing up at the fluorescent lighting - I say to the salesman. "Lets see what this blue looks like outside." We both walk out into the sunshine and I look at the suit. I then say "It now appears to be a different colour blue, I wonder what colour it really is?" This sounds funny and is something we would never say except for a joke. This is because the real colour of the suit is the colour that it has under normal circumstances (i.e., in normal daylight). This is what we mean when we talk about the 'real colour' of something. Under flourescent lighting it appears to be a certain colour. I would be deceived if I thought that this was its real colour and everyone would understand what such a deception amounted to. However if, as in the joke, I wondered whether I was still being deceived as to the colour of the suit when I looked at it in daylight, nobody would be able to understand what I was talking about. This should be our response to Descartes' argument which uses the device of the evil genius. 

This device allows Descartes to suppose that even when perceptual conditions are normal (all who are involved agree that these are standard conditions) it is still possible that I am being deceived about the colour of my suit (and everything else as well). What we must resist, if we are to resist Descartes' argument, is the temptation to believe that we actually understand what 'being deceived' by the evil genius would amount to. If you think you do understand what being deceived in this case would be like, this comes to the same thing as not seeing the joke when the man says (looking at his suit under normal lighting conditions) "It now appears to be a different colour, I wonder what colour it really is?" If someone actually said this we would think that he did not understand how to use the term 'appear' and 'real' in relation to perceiving colours. 

Similarly someone who imagines that they understand what the 'deception' perpetrated by the evil genius would be like, does not understand what a deception is. To understand what it is to be deceived you must understand what it would be like to discover that you had been deceived. In the evil genius story, we are not told what it would be like to discover that the evil genius has been deceiving us. It follows that it is, therefore, not possible to understand what it is to be deceived by the evil genius. We simply do not understand what being deceived amounts to unless we understand what being undeceived would be like. And Descartes never tells us. This is an important point because it gives us some insight into the terrible mess that Locke's finds himself in as he tries to defend or justify his common sense beliefs within the context of the new way of ideas presented by Descartes. Because Descartes does accept the evil genius argument, he has to acknowledge that all our sense-perceptions - all our ideas about the world using Locke's term - must be regarded in a different light. We cannot see them any longer as pictures of the world outside. Now they are simply 'pictures' whose reference to reality has been systematically cut off. They now stand on their own, as it were, and the 'evidence' that they provide can no longer support any knowledge about the world. 

Integrating the veridical and deceptive language-games As we have already seen in Lecture Ten, if - following Descartes - you give up the idea that knowledge (of the sort claimed in synthetic statements) consists in an agreement between what we say about the world and the way it is3 then you will be forced to the conclusion that knowledge consists in the agreement of our ideas with each other. Your problem then is that you have to confront the unpleasant fact that there are no rules by which we can tell whether two ideas 'agree' with each other (See Lecture Ten, page 60). The new language-game is unplayable. Now it is important to realize that the alternative strategy, i.e., attempting to square the apparent contradiction which exists between two language games, will not simply restore the status quo once the contradiction is resolved. The integration of the two language-games deepens our understanding of the discriminations which the integrated language-game is capable of making. And this deeper understanding spreads to related language-games and thus eventually to our whole appreciation of what we mean by what we say. 

Thus for example, suppose I recognize, in the light of the previous digression, that there is really no inconsistency between the language-games of veridical and deceptive perception. Thus I come to understand that these two games are in fact dependent upon each other for their meaning. In other words, I simply would not know what it was to be deceived by a perception unless I understood how to discover whether the perception was deceptive. I find out, in practice, that there are standard conditions which, when met, are sufficient to ensure me that my perception is not deceptive. I learn, for example, that an apple changes colour under different lighting conditions, and that its true or real colour is defined as, say, its colour when seen in daylight. In recognizing this, my understanding of what a veridical perception amounts to changes and deepens. I perhaps used to think that a veridical perception was something like a correct or accurate reproduction of the original - the thing out there in the world that I was looking at - and that what made this perception veridical was the fact that my perception reproduced the same colour as the original. I thus was assuming that if, per impossible, I could 'see' the apple directly, i.e., without using my eyes, it would 'appear' to be the same colour as I saw it to be using the eyes in my head. 

In other words I had been understanding the process of perception on the analogy of someone observing the world through a pair of glasses whose lens do not distort the image which they convey. This person is convinced that if the glasses were removed, the world would look exactly the same as it does with them on. Now in the process of integrating the veridical and deceptive language games, I have discovered that a veridical perception is to be counted as veridical, when the prevailing conditions for perception are regarded as normal, the standard for normality being fixed by mutual agreement among observers. In other words, when I use the term 'red' correctly my actual 'interior' experience of 'redness' is neither here not there. It is the fact that I discriminate 'red' in line with the discriminations of the rest of the community, that constitutes the veracity of my perception of red. The shift in perspective brought about by this integration may be summarized thus: before, I thought that my experiences corresponded to the way the world is. When I saw something red, I assumed that the thing itself was red, i.e., the same colour as the one I perceive. Now when I see something red I appreciate that I cannot compare the colour I perceive, with the 'real' colour, I can only compare my use of perception terms with other people's use of them. The mark of truth (in "veridical perception") has shifted from a correspondence view (picture with thing pictured) to a coherence view of truth (i.e., I use the term 'red' truly, when my usage of this term coheres with other people's). 

There are, then, two possible responses to a common-sense contradiction: remove the contradiction by denying one of the alternatives, or remove it by seeing that the contradiction is only apparent, i.e., integrate the opposing language- games and revise one's understanding of the significance of these games. Locke's response to the 'integration' problem A third possibility is to engage in both sorts of responses in a more or less unconscious fashion, and this is what we find in Locke. The 'denial' response is illustrated by his revised definition of knowledge as the agreement and disagreement of our ideas (with each other). Because of the new 'way of ideas' he has to deny that knowledge can be about the world outside the circle of our ideas. But he recognizes that the new definition of knowledge that this denial forces upon him results in an unsatisfactory conception of knowledge. As he says "But of what use is all this fine knowledge of men's own imaginations to a man that inquires after the reality of things." (IV iv 1) 

Locke feels the force of this objection acutely and attempts to deals with it using his theory of Primary and Secondary qualities A variation on the 'integration' response is exemplified in his doctrine of primary and secondary qualities which he puts forward in Book II, Chapter viii, section 7: Locke uses his doctrine of primary and secondary qualities to try to solve the problem of how the evidence of our senses considered in the confines of the dark room which is our consciousness can be interpreted in such a way that it can be thought of as, on the one hand - o in the case of primary qualities - providing evidence for the way things are in reality, and on the other - in the case of secondary qualities - of not providing us with any evidence for how things are in reality. This doctrine will allow him to retain his Cartesian premiss - which carries the implication that our ideas are not copies of things in the outside world (and thus can give us no knowledge of it) - while at the same time bowing to common sense by showing that some of our ideas do reproduce characteristics of their originals in the external world (and so can provide us with knowledge of it). 

He introduces his discussion in the following manner: "To discover the nature of our ideas the better, and to discourse of them intelligibly, it will be convenient to distinguish them as they are ideas or perceptions in our minds; and as they are modifications of matter in the bodies that cause such perceptions in us: that so we may not think (as perhaps usually is done) that they are exactly the images and resemblances of something inherent in the subject [i.e. the body perceived]; most of those of sensation being in the mind no more the likeness of something existing without us, than the names that stand for them are the likeness of our ideas, which yet upon hearing they are apt to excite in us." (II viii 7) 

Locke is saying that though people generally have a naive view of perceptions (namely that they resemble the things perceived), in fact, most perceptions do not. However the Cartesian argument against the naive view of perception does not conclude that most of our perceptions do not resemble their counterparts in the world, it maintains that there is no way of knowing whether any of them do. Why then does Locke resist this conclusion? I believe it is because, as the champion of common sense, he wishes to find reasons for believing that we still have at least partial access to the the world (which we 'remember' in our nostalgic moments as we recall our direct knowledge of the world we once possessed as naive relaitis)He wants to find reasons to support a preCartesian view that our epistemological condition is such that the senses reveal - in some respects at least - the true nature of objects in the world. Primary and secondary qualities Before presenting his reasons for thinking that some of our perceptions are vertical, he introduces a technical term, viz:., 'quality'. "Whatsoever the mind perceives in itself, or is the immediate object of perception, thought, or understanding, that I call idea, and the power to produce any idea in our mind, I call quality of the subject [external object] wherein that power is. 

Thus a snowball having the power to produce in us the ideas of white, cold, and round, the powers to produce those ideas in us, as they are in the snowball, I call qualities; and as they are sensations or perceptions in our understandings, I call them ideas; which ideas if I speak of sometimes as in the things themselves, I would be understood to mean those qualities in the objects which produce them in us." (II, vii, 8). Now, on the causal principle, the fact that our ideas are all 'given' in experience implies that there is a giver, or something which is the cause of these ideas entering into our consciousness. A good Cartesian would admit that we can know nothing of the character of this cause of our experience and Locke, initially-in his definition of 'quality'-sticks to this view . It is harmless enough to attribute to the cause of our ideas, 'powers' to produce those ideas in us, and to label these powers 'qualities'. 

However, he then goes on to say that these qualities are of two different sorts: "First, such [qualities] as are utterly inseparable from the body, in what state soever it be; and such as in all the alterations and changes it suffers, all the force can be used upon it, it constantly keeps; and such as sense constantly finds in every particle of matter which has bulk enough to be perceived; and the mind finds inseparable from every particle of matter, though less than to make itself singly [to] be perceived by our senses: [E.g.] take a grain of wheat, divide it into two parts, each part has still solidity, extension, figure, and mobility; divide it again, and it retains still the same qualities; and so divide it on, till the parts become insensible: they must retain still each of them all those qualities. For division (which is all that a mill or pestle or any other body does upon another in reducing it to insensible parts) can never take away either solidity, extension, figure, or mobility from any body ... 

These I call original or primary qualities of body, which I think we may observe to produce simple ideas in us, viz. solidity, extension, figure, motion or rest, and number." (II, viii, 9). Locke is saying that certain ideas among those which characterize those collections of ideas which we call 'bodies', are inseparable from them in two ways. In the first place, we never perceive a body which lacks any one of them, and, in the second place, we cannot imagine, or conceive of a body, however small, which lacked them. This first view has a solid empirical basis. It is simply a fact about, e.g., grains of wheat, that we always find these five sorts of ideas as constituents of our perceptions of them. 

Therefore, if we believed that the ideas of bodies which we experience were caused by something in the world, it would make sense to assume that something out there had the power to produce those five sorts of ideas in us. In the second place, Locke's argument maintains that a body divided beyond the range of perception, would nevertheless retain those primary qualities. This is trickier point since, by definition, the perceptions of bodies we experience cease to exist once those bodies are insensible (i.e., divided beyond the range of our perception). If I am not experiencing a perception I will not suppose that there is a body present to me. Now this does not mean that I could not have an idea of a body that was too small for me to see. But in this case my idea of a body would not be a perception of it, it would be a conception of it. If I have an idea (= a perception)-solidity, for example-I assume that there is an object 'out there' with the power to produce this perception in me.

 Following Locke I call this power in the object a quality of that object. However If I have an idea (= a conception) e.g., the idea of the 10th planet in the solar system, I do not assume that there is a planet out there which has the power to produce this idea (= conception) in me Thus if I have an idea (= conception) of a particle too small to be perceived, I cannot say with Locke, that this body has, or retains primary qualities. I cannot say this because, by definition, the insensible body of which I have a concept, lacks the capacity to produce perceptions - it is insensible - and therefore, by Locke's definition, it lacks primary qualities. There is, then, no warrant in Locke's argument for concluding that there are such things as 'insensible particles' characterized by the possession of primary qualities, since it amounts to saying that there are powers in objects to produce ideas in us which do not, in fact, produce any ideas in us, and he has, in the nature of the case, no evidence for this view. 

This is important, because in the next section Locke describes secondary qualities in terms of such insensible particles: "Secondly, such qualities which in truth are nothing in the objects themselves but powers to produce various sensations in us by their primary qualities, i.e., by the bulk, figure, texture, and motion of their insensible parts, as colours, sounds, tastes, etc. These I call secondary qualities." (II vii 10). Notice the phrase in the first sentence "which are nothing in truth in objects themselves but powers". There is here an implied contrast with primary qualities. The implication is that, unlike secondary qualities, primary qualites are perhaps something in the objects themselves. But, by definition, primary qualities (indeed, all 'qualities') are simply "the power to produce any idea in our mind." (II vii 8) So to now speak of secondary qualities as "but powers" indicates that primary qualities are somehow more than powers. What could this something more be? Where we speak of powers in the unknown causes of our perceptions we do nothing but assume a cause for our given perceptions. But when we assume that the powers which produce some ideas (primary ideas) in turn have the power to produce a second class of idea (tastes, smells, etc, secondary ideas), we need to provide some explanation, since the causal hypothesis only licences us to assume that for every idea there is a power in the unknown cause which is capable of producing this idea in us, a power for solidity, a power for yellow, etc. 

Why should we assume that the powers which produce in us the ideas of bulk, figure, texture, etc., should contain another power, namely, the power to produce in us ideas of a different sort (tastes, colours, etc.)? In particular, why should we assume that it is just when the powers to produce primary ideas are ineffective (i.e., when they are hypothetically present in 'insensible particles' and are, as such, lacking in any power to produce primary ideas in us) that they are deemed responsible for producing secondary ideas in us? This is a good question, because the causal hypothesis which Locke is employing with reference to the causes of our ideas, only licences him to postulate a single power in the unknown cause(s) of our perceptions for each single perception we experience. The speculation that several of these powers (the primary qualities) interacting in various ways, are capable of producing certain sorts of perceptions (colours, tastes, etc.) seems extraordinary. Why not stick to the 'for every perception a separate power' hypothesis? 

The reason is, I think, that for Locke, a causal hypothesis does not count for much unless it explains how the effect brings about the cause.4 This is regarded as important because if it were possible to somehow determine the mechanism whereby the cause (the unknown object outside us) brings about its effect (the resembling picture within the dark room) this might allow us to place some confidence in the hypothesis that the unknown object is like the picture of it that we see. That some story about the mechanics of causation could help us here is based (I suggest)) on the fact that the way in which the stone in a signet ring is incised, completely explains how the impression which it leaves in the wax comes to have the character which it has. When I look at this impression in the wax I know what the cause is like because I understand the mechanics of the process which brought the impression into being. 

If the causal history of some features of the 'resembling pictures' in the dark room could be given a mechanical interpretation (in a way analogous to the causal history which explains the nature o the impression signet ring) then these pictures could be regarded as supplying us with veridical information about what the world beyond the veil of perception is like. This attempt to find a justification for the view that our perceptions of the world are veridical (at least in some respects) is one of the main motives which fuel Locke's attempt to accommodate the 'new way of ideas' with the demands of common sense. After all, if at least some aspects of these ideas could be shown to resemble their causes 'outside', in the external world, common sense would be vindicated to ssome degree at least. Locke's causal explanation of how ideas might resemble their external causes in some respects 

Let us now look at Locke's attempt to establish "that the ideas of primary qualities of bodies are resemblances of them, and their patterns do really exist in the bodies themselves . . . ." (II viii 15) Locke hopes to reach this conclusion by setting out to explain how bodies produce ideas (= perceptions) in us through a series of transferences of motion. "The next thing to be considered, is how bodies produce ideas in us, and this is manifestly by impulse5, the only way which we can conceive bodies to operate in." (II viii 11) [In the first three editions he continues thus] "It being impossible to conceive that body should operate on what it does not touch . . . or when it does touch, operate any other way than by motion."6 He then goes on to explain the mechanics of how bodies produce ideas in us as follows: "If then external objects [the unknown causes of our perceptions] be not united to our minds when they produce ideas therein; [if they were so united, then according to the signet ring paradigm, the explanation of why our perceptions have the character they have would be obvious: they have been imprinted there and bear the stamp of their original cause,7 viz., external objects] and yet [i.e. but nevertheless] we perceive these original qualities in such of them as singly fall under our senses, it is evident that some motion must be thence continued by our nerves, or animal spirits, by some parts of our bodies, to the brains or the seat of sensation, there to produce in our minds the particular ideas we have of them" And since the extension, figure, number, and motion of bodies of an observable bigness, may be perceived at a distance by the sight it is evident8 some singly imperceptible bodies must come from them to the eyes, and thereby convey to the brain some motion, which produces these ideas which we have of them in us." (II viii 12) This 'and yet' clause needs to be explained. 

The reason that we nevertheless "perceive these original [i.e., primary] qualities" lies, in Locke's view, in the fact that according to his thought-experiment with the grain of wheat, even imperceptible particles must be thought of as having primary qualities. These tiny bodies come into our eyes and impinge upon the brain in a way that transfers motion to it and thus to the mind. It looks as if Locke is arguing that these imperceptible bodies - messengers from the external objects - do unite with the mind after all ( via a a character-preserving chain of impulses rather than via an 'intentional species (see note seven)) and this gives us the mechanical explanation of how ideas are produced. 

Moreover, these tiny bodies are characterized by primary qualities. This is what the grain of wheat thought-experiment set out to show. Using our paradigm, they are, in effect, thought of as tiny signet rings which transfer their inherent characteristics to the mind by impressing themselves upon the brain as we perceive external objects. In other words, Locke has been convinced by his argument that the perceptions caused by the primary qualities of external objects are effects which resemble or retain the character of their causes. He thinks this because the imperceptible bodies - the messengers from external objects - which explain how external objects 'make contact' with our minds are, themselves external bodies which somehow enter into us (or at least transfer their motion to our brains and thence to our minds). 

In this way they impress these characteristics directly upon us by uniting with the mind. Problems with Locke's explanation For the sake of the argument let us suppose that this 'signet ring' interpretation reflects the essence of Locke's explanation. It is perhaps the part of wisdom to do so, since otherwise his conclusion ("From whence I think it easy to draw this observation: that the ideas of primary qualities of bodies are resemblances of them, and their patterns do really exist in the bodies themselves. . . ." (II viii 15) seems like a non sequitur. Remember his conclusion regarding primer qualities leaps out quite surprisingly at the beginning of section l5 (In section 13 Locke explicitly indicates that he has no idea why the ideas of secondary qualities have the features they exhibit: God simply annexes them to particulars motions of insensible particles "with which they have nbo similitude" (II viii 13) and section 14 repeats the point that the causes of secondary aidea are but powers in the objects not resembling prototypes). 

Thus His 'easy-to-draw observation' regarding primary qualities follows from the discussion in section 12, and, in my view, without the 'signet ring' gloss, it is hard to understand how he can draw it easily. Suppose then we grant the signet ring interpretation simply in order to make sense of this argument. Two things are still wrong with it. The first concerns a fault in the argument that involves the thought-experiment with the grain of wheat: when I imagine insensible bodies, I imagine them as having sensible characteristics, bulk, figure, etc., which I cannot actually perceive. I can imagine these tiny bodies because I am following an inductive rule. This rule is derived from my experience of dividing visible bodies. (Remember that after each division of the grain of wheat the ever smaller parts retained their primary characteristics. This allows me to formulate an inductive rule: namely that in any further division the parts will retain their primary characteristics.) The rule allows me to hypothesize that the eventual result of continuous divisions of ever smaller parts, namely, insensible particles, have 'sensible' characteristics which are the same as those which characterize sensible objects. From this I can conclude that these invisible characteristics of (tiny) external bodies resemble the characteristics of visible ones. The fallacy here is to thinks that an inductive rule derived from the examination of several instances of dividing visible bodies could allow me to form a rule for deducing what the inherent characteristics of invisible bodies are. 

Since this argument is faulty, we cannot conclude that the imperceptible bodies which impinge upon the brain ( in Locke's story of how we come to perceive external objects) are inherently primary-qualified. It follows that we cannot conclude that they impress these primary qualities on the brain after the fashion of a signet ring. The second mistake is to imagine that the transfer of motion at its final stage (underlined in the following quotation)) is intelligible: "it is evident that some motion must be thence continued by our nerves, or animal spirits, by some parts of our bodies, to the brains or the seat of sensation, there to produce in our minds the particular ideas we have of them." At this point the transference of motion becomes a mystery according to the Cartesian paradigm. Minds, after all, are 'untouchable' (lacking extension) and therefore incapable of receiving 'impulses'. However, the details of this transference of motion from the external world to the internal 'dark room' of the mind are not here in question. 

Locke knows, a priori that such a transference, which preserves the character of the cause, must be possible, because he has already concluded - via the chain of reasoning set off by the thought-experiment on the grain of wheat - that our ideas of bodies must resemble them with respect to their primary qualities. (His conception of the invisible portion of a grain of wheat is a conception of a body with primary qualities which are inherent in it.) Therefore, since only a transference of impulse - a 'character-preserving' kind of causal link - could accomplish this, something like this must happen at the final step where ideas are produced in our minds. To repeat his argument: "And since the extension, figure, number, and motion of bodies of an observable bigness, may be perceived at a distance by the sight it is evident9 some singly imperceptible bodies must come from them to the eyes, and thereby convey to the brain some motion, which produces these ideas which we have of them in us." (II viii 12) The singly imperceptible bodies have been conceived as having bulk, figure, number, etc. and therefore will transmit these characteristics faithfully to the observer within the dark room by 'impulse'. It is only on this assumption that I can make sense of the confidence Locke has in his argument. 

A review of the argument Let me set out once more the key steps in this interpretation: 1) First of all there is Locke's motivation: to reconcile the new Cartesian teachings about ideas with the common sense belief in veridical perception (that our ideas resemble the things outside our minds which cause them) and thus allow at least limited access to The Garden. 2) The causal hypothesis: causes 'outside' produce ideas 'within' via the motions of insensible particles. 3) These insensible particles are thought of as having primary qualities (Locke arrives at this view via the grain of wheat argument). But because our ideas of these insensible particles are conceptions of particles (we do not perceive them, by definition) we can no longer think of their primary qualities as powers since - again, by definition - as insensible, they manifestly have no power to produce any ideas in us. But if we do not think of these qualities as powers we must think of them as inherent characteristics of the insensible particles. 4) There would be no reason to think that these insensible particles eventually produce (through the transfer of motion) ideas in us which resemble them unless this transference of motion were somehow character-preserving (as in the signet ring paradigm). 5) This interpretation is encouraged by the fact that the conclusion he draws ("From whence I think it is easy to draw this observation - that the ideas of primary qualities of bodies are resemblances of them and their patterns do really exist in the bodies themselves" (II viii 15) would not make any sense unless the motions of the insensible particles which produce ideas in us were regarded as character-preserving motions. 6) This interpretation is reinforced by the fact that the motions of insensible particles which cause the ideas of secondary qualities are not character - preserving. This is because - as I see it - there is no argument parallel to the grain of wheat argument which could show that insensible particles must be thought of as having secondary qualities as inherent characteristics. Locke's explanation of the link between primary powers and secondary powers 

We are now in a position to answer the question with which we began, viz., why Locke asserted that the powers which produce in us the ideas of bulk, figure, texture, etc. should have an additional or secondary power to produce in us ideas of a different sort (colour, taste, etc.). He says: "After the same manner that the ideas of these original qualities are produced in us, we may conceive that the ideas of secondary qualities are also produced, viz. by the operation of insensible particles on our senses. For it being manifest that there are bodies and good store of bodies, each whereof are so small that we cannot by any of our senses discover either their bulk, figure, or motion, as is evident in the particles of the air and water and others extremely smaller than those, perhaps as much smaller than the particles of air and water as the particles of air and water are smaller than peas or hail-stones: let us suppose at present that the different motions and figures, bulk and numbers, of such particles, affecting the several organs of our senses, produce in us those different sensations which we have from the colours and smells of bodies . . . It being no more impossible that God should annex such ideas to such motions, with which they have no similitude, than that he should annex the idea of pain to the motion of a piece of steel dividing our flesh, with which that idea hath no resemblance." (II viii 13) 

An assessment of this argument Now when Locke gave his explanation of how bodies produce ideas in us of the primary variety, he felt he needed to maintain that the causes of these ideas resembled them, for only so could he explain how our primary ideas come to have the character that they have and thus allow for a route back to The Garden via this veridical aspect of our experience. But now, with respect to secondary ideas, he feels no such need. Why did he not carry out a thought-experiment to show that insensible particles were inconceivable without our imagining them as coloured, having a certain taste, etc., on the model of grain-of-wheat thought- experiment? It is clear from what he says, that without such a chain of reasoning, his explanation - with its explicit appeal to God - falls flat. We can gain some hint as to why he did not follow a parallel chain of reasoning from the examples he gives which serve to illustrate his theory that only primary ideas have corresponding/resembling counterpart characteristics in the objects which cause them. "Pound an almond, and the clear white colour will be altered into a dirty one, and the sweet taste into an oily one. What real alternation can the beating of a pestle make in any body, but an alteration of the texture of it?" (II viii 20) 

The argument here runs something like this: However one may alter a body it remains a body, i.e., it still has primary qualities, but its other qualities change. Therefore they must depend for their character on the texture of the primary qualities. Therefore while primary qualities are, so to speak 'in' the bodies, secondary qualities are not because they change, while only the 'texture' of the primary qualities change. Therefore, secondary qualities are only 'in' the object as powers to produce ideas in us and these powers vary - i.e., produce different results - as the texture - the configuration of primary qualities of the body - changes. There is, consequently, no possibility of a thought-experiment along the lines of dividing the grain of wheat when we come to determine the causes of secondary qualities. They alter as we manipulate the objects with which they are associated, and they alter more or less unpredictably. Thus they provide us with no inductive rule for attributing colour, taste, smell, etc., to insensible particles. With no rule, we cannot find any rule to follow when we try to conceive, or imagine, the smell, taste, or colour of an insensible particle, and this lack of a rule tempts us to imagine that such particles are tasteless, colourless, etc. If we yield to this temptation, we will be quite willing to concede that aggregates of these particles are also tasteless, colourless, etc. in themselves, and are only said to have taste in so far as their 'texture' produces taste sensations in us. These sensations in no way resemble their causes - simply because we have no rule for 'explaining' - via character- preserving chains of impulses - the way in which bodies produce various ideas of secondary sensations in us. In effect, we have no explanation. 

God must bear the responsibility for this inexplicable correspondence between certain textures and certain tastes, and there the matter must rest. But, says Locke, this should not worry us, for in our experience of tertiary qualities - i.e., the power that one body has to bring about changes in another body, (i.e., the sun melting wax) "we cannot imagine that [i.e. the 'melting' effect] to be the reception or resemblance of anything in the sun, . . . ." (II viii 25) In other words we are certainly used to the notion of an effect being produced by a cause completely unlike it and therefore we should not balk at the idea of secondary ideas being the effects of causes (viz., the 'texture' of insensible particles) which in no way resemble them. 

Locke then offers two other examples, red and white colours in property disappearing in the absence of light, (II viii 19) and how the same water, at the same time may produce the idea of cold by one hand and of heat by the other (II viii 21). These instances, plus the pounded almond, illustrate the only three ways of telling empirically whether some characteristic of a thing exists independently or not.10 If under all three conditions, the thing retains one set of characteristics (its primary qualities) and if these same changes produce an alteration in another set of characteristics commonly attributed to the object in question (taste, colour, etc.,) then it would seem that the reality of the former set of characteristics is assured, since 'real' in this context is defined as 'maintains its character under all possible empirical conditions.' 

Of course, by the same token, our attitude towards the reality of the secondary qualities must be revised. Experience teaches us that these ideas can vary, and indeed disappear or contradict each other, without the object to which they are attributed, altering, disappearing or losing its identity. Therefore, it teaches us that these ideas cannot be in the object in any independent fashion (where this means, 'they do not retain their character under all empirical conditions') and so cannot be regarded as real. As Locke puts it: "The particular bulk, number, figure, and motion of the parts of fire and snow are really in them, whether any one's senses perceive them or no; and therefore they may be called real qualities, because they really exist in those bodies. But light, heat, whiteness, or coldness are no more really in them than sickness and pain are in manna. Take away the sensation of them; let not the eyes see light or colours, nor the ears hear sounds; let the palate not taste, nor the nose smell; and all colours, tastes, odours, and sounds, as they are such particular ideas, vanish and cease, and are reduced to their causes, i.e., bulk, figure, and motion of parts." (II viii 17) 

Concluding assessment 

To conclude this discussion of Locke's doctrine concerning primary and secondary qualities, two things need to be said: in so far as the doctrine is supported by examples drawn from observation of the different sets of ideas which characterize those collections of ideas which make up the 'objects' which appear to us in the 'dark room', it is unexceptionable. Experience teaches us that these 'objects' are characterized by two different types of ideas: one type varies without changing their essential character (primary ideas); the other varies unpredictably (the secondary ideas). The second type varies unpredictably because there is nothing in our experience of manipulating 'sweet' objects which would permit the prediction that pounding them in a mortar would change that 'sweet' object to an 'oily' tasting object. It might well work for almonds but what about pears or candy? However, insofar as the doctrine maintains that insensible particles instantiate primary qualities as inherent characteristics, it involves a speculation for which there could not, by definition, be any empirical evidence. We cannot experience insensible particles. Locke's attempt to break out of the circle of ideas in order to show that some aspects of our ideas are 'resembling pictures' of external objects is a failure. Common sense, which supposes this to be the case, cannot be integrated with the Cartesian epistemology which Locke has adopted. The kind of knowledge for which our senses provide evidence needs to be put on a new footing and in Lecture 16 we will see how Locke seeks to accomplish this. 

In the following Lecture we shall consider Locke's conception of substance. 

1 Note that in the Introduction this part of Descartes' argument is given special attention. 

2 'Language-game' is a term coined by Ludwig Wittgenstein in his Philosophical Investigations, (translated by G.E.M. Anscombe), Basil Blackwell, Oxford l953, § 7 ). It refers to the fact that a given cluster of terms mean what they do because they are used in a certain context - what he called a 'form of life' (see §19, Ibid). These terms can only be properly understood in that context and if they are used in some other context there are likely to be breed confusion without our being able to understand what has gone wrong. 

3 In Locke's terms an agreement between our ideas (thought of as 'resembling pictures') and the way the world is (the 'originals' from which the 'pictures' were copied). 

4 Edwin mccann notes t"It is surprising, in view of the importance of these considerations in Locke's overall argument from primary and secondary qualities that these sections (11-13 of essay IIviii) have been so little commented on " p 62 the Cambridge Companion to Locke. Cambridge University Press Cambridge l994. I could find nothing in the literature which dealt with the mechanics of causation' which Locke hypothesises as an explanation of how bodies produce ideas in us. 

5 It is this mention of impulse - and in the quotation which immediately follows this one, the mention of touch - that makes me think that something like the signet ring model of causality is before Locke's mind as he attempts to explain why the ideas we experience have the character which they have - a character which is preserved through the various transferences of motion which link them to their external causes. 

6 See note l in A.C. Fraser's Unabridged Edition of the Essay, Dover Publications, New York, 1959, p. 171. In the note Fraser explains that Locke dropped this last passage from the Fourth Edition since Newton's discussion of gravity had by then convinced him that bodies could operate in ways other than by impulse (i.e., the transference of motion by touching). 

7 Locke has given up this possibility though as we shall see , as a story of how perception works it is ideal since it allows for the vertical transmission of features which characterizes external objects to the features which characterize ideas. mccann very nicely summarizes the aristotelean scholastic theory which allowed 'external objects to be united to our minds when they produced ideas therein' " The Aristotelian Scholastic doctrine of qualities held that most, at least , of the sensible qualities of objects are real qualities,, that is, they are real entities existing of inheriting in the objects, and that perception of them involves the mind taking on the form of these qualities as they exist in the object. This is facilitated by the transmission though a medium - light for example-, in the case of qualities perceive by means of vision - of an intentional species that becomes the form of the relevant perception or act of mind; this intentional species is the from that exists in the object, except that this form exists not in matter as it does in the object but in the mind. The idea in the mind is thus q qualitatively identical with the quality in the body that initiated the whole causal process, since these two are the same i from or species; and so it can properly be said to resemble the quality as it is in the body imcann 1994 p64 

8 Fraser adds here: "'evident' - because bodies, [Locke] assumes, cannot otherwise than by continuity of motion occasion the motions in the organism on which our perceptions somehow depend. ." Ibid, note 4, p.171 

9 See note 3. 

10 Thus I try to change the character of the thing itself by physically altering it (the almond), I try to change it by changing the conditions of its perception (changing the lighting on porphry), I try to change it by changing the perceiver (put a cold hand and a warm hand in the same water).