Esser, Frederick: John Locke‘s Investigation into our Knowledge of Bodies


Kapitel 3. Lockean Resemblance

Locke introduces his notion of resemblance as part of some further considerations concerning simple ideas in II.viii. Evidently, however, as his examples show, Locke is interested in our ideas of bodies and in simple ideas only in so far as they make up ideas of bodies. His resemblance theory concerns therefore the simple ideas of our everyday ideas of bodies. Distinguishing the properties of bodies in primary and secondary qualities, Locke argues that ideas of primary qualities are resemblances whereas ideas of secondary qualities are not. He contends more specifically that contrary to popular belief our ideas of sensible qualities, e. g. of colours, tastes, etc., are not resemblances.

As the discussion of recent years and the variety of suggested interpretations prove, Locke‘s doctrine of resemblance is not easy assessable. First of all, one has to distinguish between two notions of resemblance. As Yolton has persuasively shown, at Locke‘s time the relationship between ideas qua intentional objects representing entities has been explained by saying that ideas resemble the entities they represent, e. g. by Cartesians.<135> That is: an idea represents the entities which it depicts because the idea resembles the entities, i. e. because the idea portrays the entities by properties which they possess.<136> And since the content of ideas determines which entities are represented, ideas have to resemble the entities they represent. If an idea represents its cause, it does so precisely - otherwise it would not represent its cause.<137> In turn, in this context ’resemblance‘ is on par with ’representation‘: an idea resembles the depicted entities in the sense that it represents the entities by an accurate depiction of them. However, this conception of resemblance is evidently not present in Locke‘s context of qualities, since ideas of secondary qualities are said not to resemble the properties which they represent; which would be a contradiction in terms, if the former understanding of resemblance were


manifest here.

More importantly, Locke‘s comments on resemblance are on the face of it only to a degree illuminating. It is therefore not surprising that a variety of interpretations has been suggested.<138> E. M. Curley suggests a “causal reading“ of resemblance which I will support and substantially supplement. According to Curley, in the case of resemblance the same concept denoting the idea also signifies the represented property in a scientific account of this feature as the cause of its idea.<139> In this sense, which will be delineated in what follows, I will speak of “conceptual resemblance“. Curley does however concede that in these passages Locke speaks of secondary qualities in connection with the reality status of qualities when really meaning ideas of secondary qualities.<140> The problem now arises for Curley‘s line of interpretation that these comments are linked with the topic of resemblance. This, it seems to me, leads Alexander to the conclusion that there must be more to Lockean resemblance than Curley contends.<141> Probably for the same reason, Ayers differently reconstructs the resemblance doctrine as a whole, dropping Curley‘s approach of a causal reading.<142> I will argue in this and the next chapter that Curley is


basically right as to remblance, but, as most interpretators, he is wrong about the reality of qualities. Curley‘s reading neglects however substantial aspects and thus does not apprehend the genuine topic of Locke‘s resemblance theory. The issue of resemblance, I will contend, is to determine how far our ideas portray bodies from a perception-neutral viewpoint that is identified with the scientific, causal standpoint.

As just alluded to, this account will be further deepened in the next chapter, since contrary to prevailing interpretations Locke‘s view of the reality of qualities is an ontological claim that is linked to the same perception-neutral viewpoint specific for Lockean resemblance.<143> In fact, this nexus between the two topics makes first and foremost intelligible why there is a correspondence between Locke‘s major contentions on the resemblance of ideas and his contentions on the reality of qualities. It likewise explains why, curiously, after having gone on stage the notion of resemblance virtually disappears again. For, even though the resemblance theory sets the stage for the theory of qualities, in later parts of the Essay Locke‘s claims on resemblance reappear in the guise of assertions about qualities and their reality status.

In this chapter, first, the relata will be clarified and, subsequently, Lockean resemblance as a whole, including his chief contentions on resemblance. Due to the close ties between Locke‘s explanations on qualities and resemblance, the reconstruction is however completed in the following two chapters in three ways. First, further textual evidence will be presented that in this context ideas and secondary qualities are to be understood as suggested.<144> Second, as part and parcel of the theory of qualities, Locke‘s reasoning is highlighted for his principal claims on resemblance and qualities.<145> Third, the analysis of the argument establishes furthermore that pace Alexander Lockean resemblance and its perception-neutral viewpoint are not bound up with corpuscularian theory, but are rather


theory-neutral.<146> On the other hand, this chapter also pertains to the other two, since it does not only delineate the perception-neutral perspective, but also sets out a line of interpretation what (ideas of) secondary qualities are which is equally decisive for a correct grasp of Locke‘s comments on qualities.

a. The Relata of the Resemblance Relationship

Locke expresses his notion of resemblance rather methaphorically and vaguely. Taken by themselves, they open the gates for a variety of different interpretations. For example, Locke introduces ’resemblance‘ as follows:

“§7. 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 maynot think (as perhaps usually is done) that they are exactly the Images and Resemblances of something inherent in the subject; 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.“<147>

In this context Locke distinguishes between two meanings or conceptions of ideas. In a first sense, ideas are mental states, namely perceptions or sensations. In a second sense, an idea is the cause of such mental states, i. e. of an idea in the first sense. The two conceptions of ideas are linked by a causal theory of perception. An idea of the second sense causes an idea of the first one: the property of a body causes one‘s perception of this feature. As we will soon see, Locke uses ’idea‘ primarily with the first meaning to denote perceptions and less frequently with the second one to signify the properties of bodies which cause these mental states.<148>

Locke draws the distinction as a means to prevent the common mistake of regarding perceptions as exact images or resemblances of the corresponding properties causing the perception. But if one differentiates between the two kinds of ideas, the possibility


becomes plain not to understand every idea of the first sense as a resemblance of an idea of the second one. This intention of Locke becomes clear in §7 by his comparison between ideas (in the first sense) and words: most of our sensations do not have any likeness of something existing outside of us, just as there is no likeness between a name and the idea being signified by the name. The resemblance relationship holds between an idea taken as a mental state and an idea taken as the property causing this sensation. An idea resembles its cause, if it is like its cause. However, ’likeness‘ is left unexplained. The depiction of ideas resembling their cause as “exact images“ certainly does not help either. And the comparison of the dissimilarity between an idea and its cause with the dissimilarity between the idea and its name does not tell us in which respect there is a lack of resemblance. The same goes for the deep metaphorical characterization of ideas being resemblances as mirrors in §16. My point here simply is, as the diversity of interpretations of Locke‘s theory of resemblance (and qualities) indicates, that Lockean resemblance remains unclear if these explanations are taken only by themselves. What are ideas qua mental states precisely and in which respect are they compared to the property causing them?

As it became common to read Locke, at the beginning of the comments on qualities and resemblance, Locke introduces the following two notions of ideas and qualities that each other correspond: an idea is a mental representation of the property which causes this representation in sensation; and a quality is the power, i. e. property, which causes a mental representation in sensation.<149> The point however is, I contend, that Locke sticks to this conception of ideas and qualities throughout the argument in the required sense. “In the required sense“ means that Locke‘s use of ’idea‘ to denote also qualities is not decisive for his doctrines of resemblance and the reality of qualities. By contrast, according to one line of interpretation, in the context of resemblance ideas of secondary qualities are mental states characterized by a specific phenomenological content.<150> And since in connection with the topic of reality secondary qualities are moreover conceived as mental


states being characterized by a specific phenomenological content, secondary qualities are thus equated with ideas of secondary qualities in the latter sense.<151> This implies in turn that ideas of secondary qualities are regarded as self-representing, namely as representing secondary quality which is a mental state characterized by phenomenological features. Two (ideas of) secondary qualities would then be distinguished in the way the sensation of yellow varies from the impression of blue, or a poundering headache differs from a dull one. Given the nexus between the issues of resemblance and the reality of qualities, the here developed reading will be further established in the following chapter which entails a thorough rejection of comprehending in this context (ideas of) secondary qualities as self-representing mental states.

In §8 Locke moves on to distinguish between ’idea‘ and ’quality‘. The explanations of §8 show that ideas are conceived here as the content of intentional states or respectively as the intentional states themselves:

“§8. Whatsoever the Mind perceives in it self, 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 wherein that power is. Thus a Snow-ball having the power to produce any Idea of White, Cold, and Round, the Powers to produce those Ideas in us, as they are in the Snow-ball, 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.“<152>

Ideas, Locke explains in the first sentence using traditional philosophical terminology,<153> are the “immediate object of Perception, Thought, or Understanding“.<154> This means, ideas are the content of intentional states. Since in §7 ideas are said to be perceptions or sensations, Locke apparently does not differentiate between an intentional state and its content, putting thus the stress solely on intentionality. This implies, in §8 he uses ’idea‘ with the first meaning of §7, declaring this signification as his official use of the concept. Furthermore, a quality is explained as the power or propensity of bodies to produce an


idea, i. e. to cause intentional content or an intentional state having intentional content respectively.<155> This peace of §8 is clear, independently of the other parts. Thus, on the face of it ’idea‘ denotes primarily the intentional content of mental states caused by qualities.

The distinction of §7 between two meanings of ’idea‘ corresponds to the one between ’idea‘ and ’quality‘ advanced in §8. What is called ’idea‘ with its second meaning in §7, is signified as a quality in §8. Consequently, since §8 is linked to §7, the resemblance relationship should hold between ideas as intentional objects and qualities as the cause of them! True, Locke‘s paradigm are sensations, since with respect to this type of intentional states the intentional object can be straight forwardly identified as its cause. But since Locke evidently regards his theory of qualities and resemblance to apply to all kinds of intentional states and not only to sensations, one should not misunderstand his explanations too narrowly to confine them only to sensations, as Alexander does.<156> In other words, ’idea of x‘ refers to x as the intentional object of any type of intentional state.

The second part of §8 is, no doubt, a linguistic muddle. The problem is to interpret the two occurrences of ’as they‘. I prefer to read them to refer to ’powers‘ because this would consider the semicolon to separate two different explications of powers in terms of ,as they‘. Given this interpretation, qualities are first explained as powers producing ideas (i. e. intentional objects) in us; ideas are then elucidated as powers in a sense in which powers can be said to be perceptions, namely as powers conceived as intentional objects. This implies, Locke uses ’power‘ with two significations corresponding both to the sense of ’idea‘ and of ’quality‘ in the first sentence of §8, and to the two meanings of ’idea‘ in §7. In the last part of §8 Locke, however, warns the reader that despite these conceptual clarifications he sometimes uses ’idea‘ to denote qualities as well. He thus reintroduces the ambiguity of ’idea‘ which has been explained in §7 and substituted by the distinction


between ’idea‘ and ’quality‘ in the first part of §8. There is in fact no need to read Locke‘s attempt to illuminate his key concepts as confused - but one might become thoughtful about its true impact on the innocent reader.

More importantly, it becomes evident what Locke is aiming at. He distinguishes between two ways to comprehend the same properties: to regard them as the objects of our intentional states and to conceive them as the causes of these intentional objects or states. This is expressed by his elucidation of ideas both as powers in our minds and as qualities of bodies. One even arrives at the same interpretation, even if, contrary to the above assumption, one reads the first occurrence of ’as they‘ to refer to ideas and the second one to qualities: qualities are then said to be ideas in the sense in which an idea can be comprehended as being in a body, and an idea is said to be a quality in the sense in which a quality can be conceived of being in the mind. Thus in either case Locke is struggling to distinguish between two different conceptions of the same property, namely between a feature as an intentional object and the feature as the cause of an intentional object. The topic of §§7-8 is to make plain the difference between properties of bodies as they are represented in our minds and as they are understood in a causal account of perception. Now, the crucial consequence is that, if Locke‘s account is coherent, the resemblance relationship holds between two conceptions of the same property. And since Locke introduces this understanding of ideas and qualities immediately before he differentiates between primary and secondary qualities, ideas of both primary and secondary qualities are prima facie to be comprehended in this sense, namely as properties of bodies conceived as the content of intentional states which are caused by this quality.

By contrast, to comprehend ideas as self-representing mental states characterized by phenomenological characteristics would introduce in Locke‘s argument unnecessary incoherences, if not confusion, since contenders of this line of interpretation understand only ideas of secondary qualities as representing mental states, but not likewise ideas of primary qualities. Only the latter ideas are conceived in the proposed way as intentional states or objects representing the property causing these ideas in sensation. In addition, this approach can easily lead to the view that, in the context of resemblance, secondary qualities are not properties of bodies, but of a perceiver, since a secondary quality is taken as the content of its idea, namely as the mental state being represented by the idea. But,


again, §§7-8 are placed immdiately before the distinction between primary and secondary qualities and thus treat them on par. One principal claim of the interpretation proposed in this and the next two chapters is that in connection with the resemblance theory and the reality doctrine there is no need to read Locke to deviate from his official understanding of ideas and qualities which he delineates in §§7-8. Quite the contrary, one can highlight Locke‘s explanations as conceptually coherent. For, as we will see, even those passages can be interpreted in the light of the here advanced view which on first thought seems to imply the rival reading that (ideas of) secondary qualities are self-representing mental states.<157> The thrust of Locke‘s distinction between ideas and their causes, i. e. between ideas as powers (or qualities) in the mind and powers (or qualities) as ideas in bodies, is not to differentiate between phenomenological features of a mental state and the property causing it, but to separate two conceptions of the same property. The epistemological significance which Locke attaches to this will be addressed below. First, however, Lockean resemblance will be reconstructed on the basis of our results.

b. Lockean Resemblance

The point of the preceding section is to demonstrate by a careful analysis of §§7-8 that, on the face of it, the resemblance relationship holds between ideas qua intentional object and qualities qua properties being represented by ideas and causing these ideas in sensation. The relationship thus compares two conceptions of the same properties. As we will shortly see, this naturally paves the way for Curley‘s interpretation of Lockean resemblance which will be significantly enhanced in the subsequent section. As indicated above, the problem of Curley‘s reading is that he maintains the orthodox view on the reality of qualities as well. According to him, Locke contends that secondary qualities are only real qua being properties of a perceiver, namely mental states characterized by phenomenological features, whereas primary qualities are real qua being physical properties of bodies. This implies, secondary qualities are taken, on the one hand, as properties of bodies in connection with resemblance and, on the other hand, as mental


states in the context of their reality status. In the light of this consequence, however, Curley‘s reading is hardly tenable. The reason is that Locke‘s comments on resemblance and on the reality of qualities are closely intertwined. Both claims are the two sides of the same coin. Alexander, who basically agrees with Curley on resemblance and reality, avoids this problem by interpreting terms like ’yellow‘ not as denoting secondary qualities, but ideas of secondary qualities.<158> Resemblance could thus be understood to relate to secondary qualities (in the sense of properties of bodies) and the reality status to ideas of secondary qualities (in the sense of sensations of secondary qualities, or properties of bodies respectively). In my eyes, this reading of Locke‘s terms like ’yellow‘ is straight forward contra-intuitive, given the nexus between them and Locke‘s comprehension of secondary qualities as properties of bodies in terms of microphysical primary qualities.<159> Any alternative reading which ascribes to Locke as advancing more or less coherently a different account should be preferred.

On the supposition of the orthodox view of the reality doctrine, I believe, the only convincing way out is Ayers‘s reading.<160> The price to pay is however that in the case of secondary qualities the notion of resemblance does not relate any more to two different conceptions of the same properties. According to Ayers, resemblance holds between ideas of primary qualities and primary qualities in Yolton‘s innocuous sense, namely that the former represent the latter. But ideas of secondary qualities are regarded to dissemble secondary qualities in the sense in which a mental state dissembles its cause. In which sense? Well, as Ayers himself alludes to, it is unintelligible what kind of dissemblance this should be. Moreover, in the context of resemblance and reality secondary qualities are conceived as mental states characterized by phenomenological features, whereas in other contexts secondary qualities are properties of bodies. And since both kinds of contexts are present in the comments on resemblance and qualities, this line of interpretation


understands Locke as employing ambiguously his central terms: ’idea of secondary quality‘ denotes either self-representing mental states characterized by phenomenological features or mental states representing properties of bodies; ’secondary quality‘ and ’yellow‘ (etc.) signify either self-representing mental states or properties of bodies.<161>

But there is no need to read Locke in this way, namely as talking of two kinds of resemblance relationships, of which one is unintelligible, and of using ambiguous terms. This will be argued in the next chapter on the reality doctrine. The point I wish here to make simply is that Curley‘s approach to read Lockean resemblance is natural in the light of §§7-8, but has to be secured and supplemented by an alternative interpretation of the reality status of qualities. Otherwise, Ayers‘s reconstruction is the better alternative. Curley‘s approach will now be unfolded and further deepened in the next section.

Curley elucidates the resemblance relationship in conceptual terms. According to him, in the case of resemblance the same concept denoting the idea also signifies the represented property in a scientific account of this feature as the cause of its idea.<162> As we have seen, in §§7-8 Locke is concerned with clarifying his vocabulary in order to distinguish between a property being the content of an idea and the property being the cause of the idea. On the one hand, there is the depiction of a property as it is represented by an idea. On the other hand, there is one‘s grasp of the same feature as the cause of the idea representing the feature. This suggests, to read the resemblance relationship as “conceptual resemblance“ holding between two different conceptions of the same property.

One can easily advance this line of interpretation. Thereby, it comes into play that a primary quality consists in a conceptual type comprising properties of both the macro- and the microphysical level. Like Locke, one can illustrate this in corpuscularian terms. For example, the primary quality form, i. e. figure, includes macro- as well as microphysical


figure. Both features are of the same conceptual type in the sense that ’form‘ denotes them both. This means, in an explanatory account of bodies a macrophysical primary quality is conceived in terms of its microphysical counter-part. For example, macrophysical form and solidity are understood in terms of the corresponding microphysical properties, e. g. microphysical form respectively solidity. By contrast, secondary qualities like colours do not reappear conceptually on the microphysical level. Conceived from an explanatory perspective, bodies do not have colours or any other secondary qualities, since secondary qualities are identified with microphysical primary qualities. For instance, yellow is equated with minute particles being in motion on the microphysical surface structure of bodies.<163> In this sense, a secondary quality is identified with a “combination“ of microphysical primary qualities, but not with microphysical secondary qualities. Primary and secondary qualities differ crucially in this aspect.

Now, to comprehend a property as the cause of a mental representation, is to grasp the feature in scientific terms. This holds, of course, whether properties are depicted by ideas of primary qualities or by ideas of secondary qualities. Our ordinary ideas of (macrophysical) corpuscularian properties therefore represent a (macrophysical) primary quality which is scientifically conceived as a (microphysical) primary quality. Ideas of secondary qualities, however, portray properties which are not highlighted again in terms of secondary qualities, but in terms of (microphysical) primary qualities. Thus, due to the conception of primary qualities as conceptual types, in the case of ideas of primary qualities there is conceptual resemblance holding between the term denoting the property as the content of the idea and the concept signifying the feature as the idea‘s cause: they both denote the same primary quality. The mental representation of solidity resembles the causal understanding of solidity since the former depicts solidity by the same term as the latter does, namely by ’solidity‘. By contrast, the idea of yellow dissembles the scientific comprehension of yellow because the former does not depict yellow by the same concept as the latter.

But is there not a stronger notion of resemblance present in Locke‘s explanations? Take, for example, hair, i. e. our sensation or ordinary conception of hair and our scientific


understanding of hair. Assume further that the view or comprehension of hair which we derive from looking through a microscope adds up to a scientific understanding of hair. True, the view through a microscope does not give an adequate explanatory account of the constitution of hair, but for the sake of the argument let the microscope serve here as a model. Given all this: does not our perception of hair through a microscope resemble our ordinary sensation of hair more specifically than just in the ascription of primary qualities? For instance, do both depictions not have more in common than that they both attribute any kind of figure to hair? Do they not both ascribe also a similar figure to hair? Is long and curly hair not likewise long and curly if perceived through a microscope?

However, Locke does not give prominence to this kind of resemblances. He rather emphasizes that one has very different ideas of bodies if one looks at them through a microscope.<164> For Locke there is no similaritiy which goes beyond conceptual resemblance. This point becomes more obvious with regard to the corpuscularian comprehension of bodies. For example, the surface structure of a dice might be smooth in macrophysical terms. But the surface is not only “very rough“ from a microphysical perspective. It is also described profoundly different, since in corpuscularian theory a body is regarded as a compound of corpuscles on whose surface minute particles are in motion.<165> In a macrophysical description, however, there are no tiny bodies buzzing on rough surfaces. Locke, criticizing corpuscularian theory with subtlety,<166> must have been well aware of this at first very astonishing insight about the “falsity“ of our everyday physical picture of bodies. One should therefore abandon the idea to read Locke in a more figurative way. According to him, in principle, the macrophysical description of a primary quality dissembles the corresponding microphysical portrayal in every aspect a part from having in common to depict the primary quality as the same primary quality, e. g. as form.

To conclude, Lockean resemblance is conceptual resemblance. The resemblance


relationship holds between the concept signifying a property conceived as the intentional object of a mental state and the concept denoting the same property comprehended as the cause of that mental state. Both concepts are of the same type; but there is no further similarity, since they each stand for an otherwise differently conceived feature. To pick up again the example, both concepts denote figure, but one stands for a smooth surface whereas the other one signifies a rough surface with minute particles being in motion. In this sense, it is intelligible to speak of conceptual resemblance between an idea and its cause. The same concept denotes both the idea and its cause, e. g. form. Consequently, on the backdrop of Locke‘s corpuscularian list of primary qualities, our ordinary ideas of macrophysical primary qualities resemble their cause only because they have counter-parts on the microphysical level. The concept signifying the idea of a primary quality of the macrophysical stage is of the same type as the concept denoting the idea‘s cause in terms of microphysical primary qualities. The same does not hold for ideas of secondary qualities because, as Locke argues, from the microphysical standpoint one does not conceive their cause in terms of secondary qualities.

This understanding of Lockean resemblance explains why according to his notions of ideas, qualities, and resemblance it is possible that only one of two ideas resembles its cause although both have the same cause, e. g. the idea of yellow and the idea of the respective microphysical features.<167> For, although the two ideas have the same intentional object, only the latter idea is signified by the type of concept by which the cause is denoted. The interpretation also fits with Locke‘s usage of ’resemblance‘ in II.viii, for example in §13:

“It being no more impossible, to conceive, that God should annex such Ideas [e. g. of colours] 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 deviding our Flesh, with which that Idea hath no resemblance.“<168>

After having advanced an explanatory, corpuscularian account of human sense perception and after having presupposed its truth as a hypothesis, Locke explains here


why ideas of colours are like the idea of pain which does not resemble its cause: in an explanatory account of bodies the idea of pain does not resemble its cause (i. e. the sharpness of the knife‘s shape) because one uses concepts of a different type to denote the idea of pain and its cause respectively.

We have seen what it means that ideas of primary qualities resemble primary qualities, whereas ideas of secondary qualities dissemble secondary qualities. But one can dig deeper. There is a particular epistemological thrust attached to Locke‘s theory of resemblance. To see this, one has to spell out the implications of the already indicated explanatory role which primary and secondary qualities play as to ideas.

c. The Perception-neutral Viewpoint

As argued, the resemblance relationship contrasts our conceptual understanding of properties in so far as they are conceived as being mentally represented with our comprehension of the properties in so far as they are conceived as the cause of their mental representation. This means, to grasp a feature in causal terms is to conceive it independently of the mental representation of the property. More precisely, a feature is comprehended not in terms of the way it is represented, if it is understood in terms of microphysical primary qualities. And since this accounts for all properties and for all our mental representations of them, the grasp of a property in terms of primary qualities is to understand the feature independently of the way it is mentally represented. To say therefore that ideas of primary qualities are resemblances, is to claim that ideas of primary qualities depict primary qualities in the same terms in which one would conceive them independently of the way they are mentally represented by the ideas. The punchline of Locke‘s resemblance theory is to determine which ideas portray the properties of bodies from an idea- or perception-neutral conceptual viewpoint.

All this becomes manifest when one realizes that Locke does not discuss any explanatory or causal account of bodies and their properties, but more specifically a theory of qualities in conjunction with an account of our perception of them. As will be highlighted in the chapter on the theory of qualities, the assumption of a corpuscularian theory of sense


perception is the cardinal step in Locke‘s main argument for his contentions of resemblance and qualities in §§11-14. On the basis of a corpuscularian theory of sense perception, Locke demonstrates that ideas of primary qualities are resemblances whereas ideas of secondary qualities are not. Furthermore, as Thomas Heyd has forcefully argued,<169> in §16 and §§19-21 Locke presents additional empirical support by pointing out that the alternative, Aristotelian theory of sense perception and qualities cannot convincingly explain the results of certain experiments or everyday experience respectively. In §22 Locke even clearly asserts with reference to the preceding paragraphs that an account of qualities and our perception of them is at issue.<170> This likewise becomes plain at the end of II.viii in §25. And since in this context Lockean ideas are regarded as intentional objects, which will be underscored in the next chapter, and since sensations serve as paradigms for intentional states, Locke‘s theory of sense perception is in fact a theory of mental representation and intentional states. Primary qualities are those properties whose concepts are designed to be the vocabulary with which one comprehends properties without reference to a mental representation of these properties. If therefore a body is understood in terms of primary qualities, the body is grasped independently of the way it is represented in the human mind.

In connection with the resemblance theory, bodies are grasped from a specific epistemological perspective, namely from a perception-neutral, conceptual viewpoint: bodies and their properties are conceptually conceived independently of the particular way we represent them. The reason is that bodies are understood in terms of primary qualities if they are conceived as the cause of our mental representations of their properties: in a theory of mental representation, primary qualities serve as the kind of properties in terms of which bodies are comprehended as the objects of every mental representation. An idea resembling its cause is thus an idea representing its cause in those terms in which this property is grasped if conceived without any reference to the specific


way the property is mentally represented. A resemblance is a mental representation depicting its property in terms of the same concepts in which the property is understood from a causal or perception-neutral viewpoint. Thus, Locke‘s chief contention of his explanations that only ideas of primary qualities are resemblances and that ideas of sensible qualities are truly not resemblances, is an account of how one has to conceive bodies from a scientific, conceptual viewpoint. In the light of Locke‘s reasoning that the known corpuscularian properties are primary qualities, the point of his argument on resemblance apparently is to insist that, in principle, we do have a conceptually adequate comprehension of bodies even though we are ignorant of their microphysical features. In other terms, from the epistemological viewpoint of resemblance our everyday ideas depicting bodies in terms of macrophysical corpuscularian properties represent a conceptually adequate scientific grasp of bodies and their properties.

Given the presence of corpuscularian theory in the argument of resemblance and qualities, namely the assumption of a corpuscularian theory of sense perception and the use of the quality distinction, the question arises whether this epistemological perspective is meant to be a specific corpuscularian one or a general, theory-neutral viewpoint applying prima facie to all kinds of theories of perceptions. In other words: is the issue of resemblance confined to the truth of corpuscularian theory, as Alexander‘s reading implies?<171>

First of all, supposing Boyle‘s hypothesis as a premise to establish which ideas are resemblances does certainly not mean that Locke‘s notion of resemblance is part and parcel of corpuscularian theory. The assumption does not imply that the question which ideas are resemblances is specifically linked to the hypothesis, but only that the answer is premissed on corpuscularian theory. Moreover, the simple fact that Locke uses the quality distinction does likewise not imply that he does so with the Boylean meaning, since other natural philosophers like Galileo have used these terms before and there was no defined


or fixed signification for them.<172> Of course, what has just been said does neither indicate the opposite. But, on the face of it, the topic of resemblance seems not to be confined to Boyle‘s explanatory account. First, contrary to Boyle Locke develops a concept of resemblance which he introduces before, immediately before, the notion of quality and the distinction between primary and secondary qualities. This indicates that resemblance is not meant to be an “addition“ to corpuscularian theory. Second and more important, the resemblance relationship is conceived as one between an idea and its cause. This cause is what Locke then calls a quality. That is, the relationship does not make use of any characteristics which specifically apply to primary or secondary qualities.

On the other hand, a defender of Alexander‘s line of interpretation can reply that the talk of ideas resembling its cause already implies a more specific characterization of causes than just simply being causes since otherwise there would be no distinction between ideas resembling their cause and ideas which do not. And these entailed characteristics, one can add, are spelled out by Locke in terms of the quality distinction. This means, Locke‘s notion of resemblance presupposes the quality distinction so that the former would be confined to corpuscularian theory if the latter is. In my eyes, however, the topic of resemblance with its reference to properties qua causes is clearly too general to be bounded up with any particular theory of bodies and sense perception. Thus, even if Locke‘s quality distinction were meant to be specifically corpuscularian, his notion of resemblance should not be taken to be so as well.

Fortunately, we need not to content ourselves with these considerations. There is conclusive evidence that the quality distinction is not understood as part and parcel of the corpuscularian viewpoint. But before arguing this point in chapter five, the next section on the reality of qualities will first highlight that the advanced interpretation of resemblance can be squared with other passages on resemblance that are connected with Locke‘s reality doctrine. The reality doctrine and the resemblance theory turn out to be the two sides of the same coin.



Yolton (1984), 61.


Yolton (1984), 38: “Ideas, then, in some ways (resemblance is frequently suggested) represent objects.“


Logically speaking, there is no room between the content of an idea and the constitution of the represented entity whether the idea is a resemblance or dissemblance. As Jackson puts it: “There would be no sense in saying that an idea does not resemble a power to produce it.“ Cp. Jackson (1929), 68.


I will not go into details with a line of interpretation which understands Locke‘s resemblance doctrine as the claim that ideas of primary qualities necessarily depict bodies as to properties which they possess, whereas ideas of secondary qualities might fail to do so. Cp. Aaron (1937), 126 and O‘Connor (1952), 67. As other commentators have rightly and thoroughly argued, Lockean resemblance has nothing to do with mistakes in perceptual judgements. Cp. Curley (1972), 152-55; Mackie (1975), 13ff. - Generally speaking, many interpretations will be refuted simply by re-interpreting their textual evidence. For instance, Jackson‘s interpretation of resemblance is based on the assumption that primary qualities add up to indeterminate macrophysical properties: “The ideas of the secondary qualities are, therefore, probably held to be produced by microscopic qualities, while the ideas of primary qualities are probably held to be produced ny macroscopic qualities. Locke‘s doctrine then will be that macroscopic qualities produce resembling ideas, while microscopic qualities produce non-resembling ideas.“ Cp. Jackson (1929), 68. Since I will argue that primary qualities are neither determinable nor, crucially, macrophysical, there is no basis for Jackson‘s thesis that Locke does only identify secondary qualities, but not primary qualities with microphysical primary qualities when they are conceived as powers to produce ideas. Cp. 5a.


Curley (1972), 150f.


Curley (1972), 141.


Alexander (1985), 198.


According to Ayers, Locke conflates Yolton‘s kind of resemblance with another one. Yolton‘s notion of resemblance is present when ideas of primary qualities are said to be resemblances. But Locke is conceived as making use of ’resemblance‘ in a second sense in connection with ideas of secondary qualities. Here, resemblance is taken to hold between ideas of secondary qualities qua blank effects and secondary qualities qua being the cause of the blank effects, i. e. between mental states characterized by phenomenological features and physical properties. Cp. Ayers (1998), II, 1064 , 1091, and 1092. Cp. also Ayers (1991), I, 63ff.


McCann likewise highlights resemblance in the light of the corpuscularian criticism of Aristotelian “real qualities“. However, McCann‘s approach does not lead to an elucidation of Lockean resemblance. He explains an idea, that is a resemblance, as being “qualitatively similar“ to the property it represents, without clarifying what he means by this expression. Cp. McCann (1994), 64.


Cp. 4c.


Cp. 5b.


Cp. 5b.


134, II.viii.7. Cp. 137, II.viii.15f for further vague depictions of resemblance.


Cp. 376, II.xxxi.2. Cp. my interpretation of this passage in 4c.


Cp. Alexander (1985), 114ff; Ayers (1991), I, 58; Curley (1972), 141-45; Mackie (1975), 15.


Ayers (1991), I, 63ff. According to Ayers, in this context resemblance is taken to hold between the idea of a secondary quality qua blank effect and the secondary quality qua cause of the blank effect, i. e. between a mental state characterized by phenomenological features and a physical property. Cp. Ayers (1998), II, 1092.


Ayers (1991), I, 63.


134, II.viii.8.


Cp. Yolton (1984), 88-104.


134, II.viii.8. Cp. 47, I.i.8. Cp. likewise 104, II.i.1 where Locke expands his phrase of ideas being the immediate object of the understanding.


Jackson contends that one should strictly distinguish between powers and qualities, since only primary qualities are genuinely qualities and that secondary qualities are in fact powers. Cp. Jackson (1929), 55 and 59. Jackson is right in reading some passages in this light, cp. 135, II.viii.9 and 375f, II.xxxi.2. Here Locke indicates that according to his “taste“ one should call only primary qualities qualities, but not likewise secondary qualities. Calling secondary qualities ’qualities‘ is for him only to comply with an established way of speaking (135, II.viii.9). However, after all, he does name secondary qualities ’secondary qualities‘. And in this sense he apparently highlights qualities in terms of powers in §8 what subsequently serves him as the basis for distinguishing between two types of qualities, calling the latter ones secondary qualities.


Alexander (1985), 199.


Cp. 5c.


Cp. Alexander (1985). Alexander maintains, on the one hand, that secondary qualities are properties of bodies (118). But since he claims that colours (etc.) are truly ideas of secondary qualities (118) - and not secondary qualities, as I contend -, he takes the issue of colours (etc.) being real as a question of locating them correctly. Colours (etc.) are genuine properties of a perceiver, i. e. ideas, and not properties of objects causing ideas (125f). As to resemblance, cp. 198.


For instance, in Locke‘s definition of ’secondary quality‘. Cp. 135, II.viii.10.


Cp. Ayers (1998), II, 1064 , 1091, and 1092. Cp. also Ayers (1991), I, 63ff.


This is not exactly what Ayers says, but it is implied by his position. See above.


Curley (1972), 150f. In principle, Alexander maintains the same approach. Cp. Alexander (1974), 458f. However, he decisively obscures his account because he believes that there has to be a stronger conception of resemblance. For him, one has to explain in which sense one can intelligibly say that the idea of the property X has the property X, e. g. that the idea of extension is extended. One result of his effort is that the notion of resemblance is narrowed down to sensations so that other types of intentional states are excluded. I am affraid, I do not see Alexander‘s problem in the first place, namely why Locke is allegedly confined to the claim that an idea is extended in any stronger sense than that the represented entities are extended. Cp. Alexander (1985), 198-203.


545, IV.iii.13; 589,


301f, II.xxiii.11. In this context Locke focuses on our ideas of sensible qualities and compares them to ideas of the microphysical structures with which one would identify a sensible quality in an explanatory account of bodies. Yet, his comments make plain that ideas of macrophysical primary qualities vary also greatly from ideas of microphysical primary qualities.


Cp. 545, IV.iii.13; 589,


295f, II.xxiii.2; 308-12, II.xxiii.23-28.


545, IV.iii.13; 589,


136f, II.viii.13.


Heyd (1994), 19-27.


140, II.viii.22: “§22. I have in what just goes before, been engaged in Physical Enquiries a little further than, perhaps, I intended. But it being necessary, to make the Nature of Sensation a little understood, and to make the difference between the Qualities in Bodies, and the Ideas produced by them in the Mind, to be distinctly conceived, without which it were impossible to discourse intelligibly of them; [...].“


Alexander holds, first, that Locke‘s quality distinction is Boyle‘s one and, second, that, the Essay is aimed at establishing that corpuscularian theory can best explain our everyday experience and at spelling out its implications for philosophy. And since in the comments on resemblance and qualities, Locke discusses everyday experience and even experiments, it follows for Alexander‘s line of interpretation that the notion of resemblance is a corpuscularian term. Cp. Alexander (1985), 7 and 118.


Galileo introduced the distinction in modern philosophy in Il Saggiatore. Cp. Galileo (1968), VI, 213-372.

[Titelseite] [Widmung] [Danksagung] [Einleitung] [1-1] [1-2] [1-3] [1-4] [1-5] [1-6] [2-1] [2-2] [2-3] [2-4] [3-1] [3-2] [3-3] [Zusammenfassung] [Bibliographie] [Selbständigkeitserklärung]

© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.

DiML DTD Version 2.0
Zertifizierter Dokumentenserver
der Humboldt-Universität zu Berlin
HTML - Version erstellt am:
Fri Sep 27 15:57:21 2002