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


Kapitel 2. The Substratum of Bodies

Locke develops his theory of the substratum of bodies as part of his account of our common ideas of substances. Every idea of substances, Locke asserts, contains the idea of substratum. The inclusion of this idea, which is also called the general idea of substance, is precisely what makes an idea of substances to an idea portraying (a class of) substances. In other words, to have or be a substratum is to be a substance. A substratum is thereby understood to be the “cause of the union“ of all the other properties unified in an idea of substances.<82> Moreover, Locke presupposes general agreement rather than demonstrates that our common ideas of bodies, finite spirits, e. g. human minds, and God include the idea of substratum.<83> But this idea of substratum, Locke argues, is confused. The stress of Locke‘s chief contention that our ordinary ideas of bodies and of spirits contain a confused idea of substratum, therefore does not lie on our possession of an idea of substratum, but on the fact that we just have a confused one. Now, as to bodies, Locke‘s explanations on substratum add up to this general claim.<84> This means in the light of Locke‘s notion of idea that his chief reasoning is an argument for the thesis that our conceptual comprehension of the substratum of bodies is in some sense confused.

This “knowledge thesis“ comprises two aspects, namely what a substratum is and what it means that the idea is confused. I will agree on the latter issue with what has been suggested by other commentators, e. g. Ayers, even though I am going to disagree with their interpretations of what a substratum consists in. One can hold on to the suggested notion of confusion, but replace the understanding of substratum. With respect to substratum, however, an alternative reading will be developed. There are two diverse, common lines of interpretation. One is Bennett‘s view that a substratum is a featureless, pure logical subject, the other one highlights the substratum of bodies as a specific set of


(micro-) physical properties which has been differently determined by different commentators. Generally speaking, I agree with those proponents of the latter line of interpretation who identify the substratum of bodies as matter. But there is largely disagreement with respect to the details of the argument. In my eyes, one commonly misses the point of Locke‘s formula that a substratum is the cause of the union of coexisting properties. What he means by this depiction is usually not elucidated, but it reveals both that his comments are more coherent and that he advances a more subtle account than current readings suggest.

To come to grips with Locke‘s reasoning for the knowledge thesis, his notion of the substratum of bodies will first be discussed and subsequently the assertion that the idea of substratum is confused. Then, Locke‘s justification will be analysed that our ordinary ideas of bodies contain a confused idea of their substratum. Finally, the conclusion will be drawn that Locke conceives the substratum of bodies in the perspective of an ideal scientific comprehension of bodies. This account will be confirmed in chapter five on the theory of qualities which re-establishes that Locke‘s analysis of the ideas of bodies assesses our conceptual grasp of bodies from an ideal scientific viewpoint.<85>

a. The Notion of the Substratum of Bodies

At the beginning of the chapter on the ideas of substances, Locke officially introduces the notion of substratum. The advanced characterization is repeated in the course of the explanations and also in other passages.<86> Moreover, since ’to inhere in‘ and ’to support‘ are aquivalent to ’to subsist in‘, as will be explained below, the same use of ’substratum‘ appears at further places as well.<87> Locke introduces ’substratum‘ as follows:

Ҥ1. THE Mind, being, as I have declared, furnished with a great number of the simple Ideas, conveyed in by the Senses, as they are found in exteriour things, or by Reflection on its own Operations, takes notice also, that a certain number of these simple Ideas go constantly together; which being presumed to belong to one thing,


and Words being suited to common apprehensions, and made use for quick dispatch, are called so united in one subject, by one name; which by inadvertency we are apt afterward to talk of and consider as one simple Idea, which indeed is a complication of many Ideas together; Because, as I have said, not imagining how these simple Ideas can subsist by themselves, we accustom our selves, to suppose some Substratum, wherein they do subsist, and from which they do result, which therefore we call Substance. “<88>

In the first part of §1 Locke describes the process of acquiring the idea of substratum as part of our forming ideas of substances. After having repeatedly experienced the same set of properties in several particulars or in one‘s own mind (“that a certain number of these simple Ideas go constantly together“), these properties are regarded to belong to the same subject: “which (i. e. properties) being presumed to belong to one thing, ..., are called so united in one subject“. That is, the regularly experienced bond of properties results in our assumption that these properties are features of the same entity. In the same light one has to read the last part of §1 and in particular the notion of ’to subsist‘, since Locke refers back to what he has said before (“as I have said“): the repeated experience of the same properties being together leads to our assumption of a substratum, i. e. we suppose that these entities have a substratum (“not imagining how these simple Ideas can subsist by themselves, we accustom ourselves, to suppose some Substratum, wherein they do subsist“). This means, Locke specifies our supposition that properties belong to the same entity as the assumption of a substratum, namely as the supposition of a substratum being understood as what explains the experienced union of properties. To say that the same properties are found together because they are properties of the same entity, is for Locke to say that the bond of the properties is due to a substratum. A substratum is therefore what explains the bond of properties by elucidating how these features are properties of the same entity.

In other words, ’substratum‘ is an explanatory notion introduced to designate that feature of substances which accounts for the experienced union of the same properties. Correspondingly, by ’some x subsist in y‘ Locke means roughly ’the bond of some x is due to y‘. And likewise one has to understand similar expressions being used in subsequent paragraphs and elsewhere; i. e. ’to support‘ and ’to inhere in‘ denote the same relationship


between a set of properties and a substratum, namely, as Locke puts it, that the substratum is the cause of the union of coexisting properties.<89> Furthermore, Locke‘s reference to regularity means that the assumption of a substratum is caused by the recurring experience of the same set of properties. This does not mean, however, that Lockean substratum is defined by explaining this recurrence. One does not suppose a substratum to explain the fact that one has repeatedly experienced a set of properties. This can be taken from other passages where Locke uses similar phrases to depict substratum, but without relating this to regularity.<90> Rather, as Ayers has pointed out, the supposition of a substratum is a specific response of the mind to the (repeated) experience of the union of properties to explain the union of the features, but not the regularity of the experience.<91>

The point seems to be that the experience of a set of properties cries for an explanation of how these properties are interrelated, and our epistemic response is to assume that a real or ontological bond exists between the features.<92> The general character of this kind of experience, namely that it is not confined to any particular properties, explains why Locke‘s conception of substratum applies to all kinds of features, i. e. to simple ideas of sensation and of reflection. This is the reason why a substratum is purely functionally understood as what accounts for the union of properties and as applying to all kinds of substances. Locke expresses the functional conception of substratum by saying that “we have no Idea of what it is, but only a confused obscure one of what it does.“<93> It is important to note that ’substratum‘ is defined as a general conception which applies uniformly to all kinds of substances, namely to bodies, finite spirits like human minds, and God. For if one overlooks the uniformity of Locke‘s conception, one is easily driven into the direction Alexander goes, namely to understand ’substratum‘ as denoting a mixed bag of different stuffs which do not have a generic conception of a substratum is



As we have just seen, according to Locke the experience of the same set of properties leads to one‘s supposition of “some substratum“ to explain the union of the features found together. A substratum is thus functionally characterized as what explains the bond of properties which one encounters. In this sense, a substratum is characterized by a causal relationship: a substratum causes the union of features. To put it in slightly paradoxical terms, a substratum is the bond of properties qua being the cause of their bond. This causal understanding of Lockean substrata suggests, of course, that a substratum is to be identified with a set of other properties which substances have, e. g. microphysical ones. A substratum is the set of those properties of substances which explain the bond of the experienced features of an entity. In what follows the leading question will be to assess more precisely which properties of bodies comprise their substratum.

But before turning to the matter, a radically different line of interpretation should be discussed first. For, according to Bennett, the advanced approach is already on the wrong track and sets a completely false perspective in which to interpret Locke‘s further explanations. Bennett maintains what he calls the “Leibnizian interpretation“,<95> since he believes that Leibinz has already interpreted Locke correctly: for Locke a substratum is a “pure logical subject“. ’Substratum‘ is like the dummy concept ’thing‘ which can be used, even if one gives an exhaustive description of an entity‘s features: ’This thing has the


properties A, B, C, etc.‘<96> In this sense a substratum is a featureless subject to which one predicates properties only to make explicit that these properties belong to the same entity. And a substratum is pure logical in so far as it is stripped off any features. This means, if one equates a substratum with a set of explanatory basic properties, one is on the wrong track. According to this reading, Locke puts forward an argument which explains the substratum of a substance as a featureless something which serves as a subject in predications to depict a substance.

However, this view faces three difficulties at least. Firstly, given Locke‘s notion of obscure and confused ideas, as understood below, the idea of substratum would be neither obscure nor confused. Locke‘s knowledge thesis would become obscure itself. Secondly, to elucidate a substratum as a pure logical subject, would introduce incoherences in other parts of his explanations. Given Bennett‘s view, there would be no point for Locke to discuss whether the substance of finite spirits could exist or not. If the substratum of finite spirits were a pure logical subject, it would not make sense for Locke to argue that it exists.<97> Thirdly, in the very same context, Locke also suggests that there are two different kinds of substrata, namely one for bodies and one for spirits. This becomes evident in the Stillingfleet correspondence where Locke clarifies his explanations by asserting that the substance of finite spirits need not to be necessarily immaterial, but could also be material.<98> But a distinction between two kinds of substrata would be implausible, if substrata were pure logical subjects. The alternative line of interpretation faces certainly its own difficulties, in the way it has been developed by Alexander or Ayers.<99> Yet, despite


Bennett‘s rebuttal of criticism, the objections to the former views are not as big as the incoherences implied by his own one.<100>

In §2 Locke moves on to argue that we have zero or only limited knowledge of substrata. He claims subsequently that corpuscularians, children or the laymen do not really know what the substratum of bodies is. In particular, the knowledge thesis is expressed in idea-theoretic terms, namely that one does not have a clear and distinct, but only a confused and obscure idea of substratum.<101> We will come back to what this claim exactly means. More importantly, when exemplifying one‘s limited knowledge of substrata, Locke delivers a corpuscularian conception for the substratum of bodies. This account will help to examine what Locke means by substrata. The crucial question of interpretation is how to understand the corpuscularian model:

“§2. So that if any one will examine himself concerning his Notion of pure Substance in general, he will find he has no other Idea of it at all, but only a Supposition of he knows not what support of such Qualities, which are capable of producing simple Ideas in us; which Qualities are commonly called Accidents. If any one should be asked, what is the subject wherein Colour or Weight inheres, he would have nothing to say, but the solid extended parts: And if he were demanded, what is it, that that Solidity and Extension inhere in, he would not be in a much better case, than the Indian before mentioned; who, saying that the World was supported by a great Elephant, was asked, what the Elephant rested on; to which his answer was, a great Tortoise: But being again pressed to know what gave support to the broad-back‘d Tortoise, replied, something, he knew not what. And thus here, as in all other cases, where we use Words without having clear and distinct Ideas, we talk like Children; who, being questioned, what such thing is, which they know not, readily give this satisfactory answer, That is it something; which in truth signifies no more, when so used, either by Children or Men, but that they know not what; and that the thing they pretend to know, and talk of, is what they have no distinct Idea of at all, and so are perfectly ignorant of it, and in the dark. The Idea then we have, to which we give the general name Substance, being nothing, but the supposed, but unknown support of those Qualities, we find existing, which we imagine cannot subsist, sine re substante, without something to support them, we


call that Support Substantia; which, according to the true import of the Word, is in plain English, standing under, or upholding.“<102>

Locke argues that even corpuscularians and their sophisticated understanding of bodies fall short in giving an adequate explanation of substratum. According to Locke, all of us, i. e. including the corpuscularians, talk like children when speaking of the substratum of bodies, since we do not know what it is. On the other hand, since Locke evidently believes corpuscularian theory to be true, the corpuscularian model of bodies certainly embodies the most intelligible conception of what one does not know, namely the most qualified account of what explains the experienced bond of properties. To point out the ignorance of corpuscularians, Locke draws the picture of three successive stages of explaining the union of properties. He does this in analogy to the way an Indian philosopher makes plain the “upholding of the world“. First there is the world being upholded, i. e. what has to be explained; then there is an elephant holding up the world; then a tortoise holding up the elephant; and then a one-knows-not-what holding up the tortoise. That is, the Indian does not know what truly explains the upholding of the world. Likewise, Locke expounds a corpuscularian-theory inspired model of three different levels of explanatory properties. First there is the union of properties, i. e. what has to be explained; then extension and solidity; and then a one-knows-not-what. Now, the decisive question is how to understand this model and its implications for Locke‘s conception of the substratum of bodies.

Mandelbaum has suggested to identify the substratum of a body with its real essence.<103> He maintains the common view, criticized in part two, that the real essence is a microphysical structure which is specific for the specimens of a sort. The basic argument for this interpretation is that corpuscularian theory conceives the union of experienced properties in terms of real essences. That is, the model Locke proposes in §2 is taken to imply that the substrata (of sorts) of bodies are their real essences. If the argument were true, it would be a strong one, since Locke‘s affiliation to corpuscularian theory indicates indeed that he comprehends the union of properties along corpuscularian lines. I will


contend, however, that a corpuscularian understanding of the bond of properties consists in fact in a different view. But before delineating the alternative account it is worth noting that Mandelbaum‘s reading has difficulties to square Locke‘s explanations.

First of all, Mandelbaum‘s interpretation and similar ones hinge decisively on their reading of Locke‘s corpuscularian model, since there is virtually no additional textual evidence supporting their reconstruction. Only §3 of the chapter on the ideas of substances suggests prima facie that for Locke the substratum of a body is its real essence, since he uses there an expression to denote the substratum of a sort which he otherwise applies to its real essence, namely ’the particular internal Constitution, or unknown Essence of that Substance‘. On the other hand, given strong evidence for an alternative interpretation of ’substratum‘, one can also read Locke as being only (verbally) imprecise in not distinguishing here properly between substratum and real essence. In any case, in §3 Locke does not clearly equate ’substratum‘ with ’real essence‘, so that this passage alone does not support sufficiently Mandelbaum‘s reading if he is mistaken about the corpuscularian understanding of bodies. To turn the tables, given the importance of both notions, one would expect Locke to make plain such a significant identification, e. g. when introducing ’real essence‘ officially.<104> But at no place Locke asserts explicitly that ’real essence‘ denotes substrata.<105>

More importantly, whether his view of corpuscularian theory is correct or not, critics have shown that Mandelbaum interprets effectively Locke‘s explanation in §2 as incoherent.<106> The objection runs as follows. If one identifies the substratum of a body with its real essence, the real essence appears as the one-knows-not-what on the third level. In addition, according to corpuscularian theory the real essence of a body is explained from a microphysical viewpoint in terms of the extension of cohering solid


elementary particles, i. e. corpuscles. This means, if the substratum were a real essence, microphysical solidity and extension would be assumed to be explanatory basic, i. e. to be on the third stage. Consequently, as Yolton and Ayers have indeed maintained,<107> when Locke says in §2 that a corpuscularian cannot explain solidity and extension, he must be read as referring to macrophysical solidity and extension of bodies. But if this were so, the critic now urges, Locke would not draw in §2 a picture of successive stages of explanation, as he apparently intends to do. For macrophysical solidity and extension does neither elucidate the experienced set of properties nor their bond. Macrophysical solidity and extension are features of the same level as the experienced properties are. Even worse, for Locke macrophysical extension and solidity are characteristics of bodies and therefore part of the set of features whose bond is to be specified. In short, if Locke intended to elucidate the substratum as the real essence by its explanatory role with respect to other properties, he would have failed bluntly.

A closer analysis of what a corpuscularian account of the bond of experienced properties consists in will reveal, however, that Locke does expound a corpuscularian model of the substratum of bodies in terms of three different explanatory levels. The first stage, i. e. the one of the experienced features, is of properties of the “macrophysical level“, e. g. colours and macrophysical figure. According to corpuscularian theory every property of the macrophysical level can be reduced to or explained by specific microphysical properties. For example, a colour is identified with a minute particles being in motion on the surface of coloured bodies and passing on motion by impulse to other tiny particles which finally affect our senses.<108> These microphysical properties are described in terms of microphysical solidity, extension, and mobility. Features of the macrophysical level are “the various modifications of the Extension, of cohering solid Parts, and their motion“.<109> Thus, from an explanatory viewpoint a macrophysical body is conceived, by and large, as a compound of cohering solid corpuscles which has a particular extension. (Motion will be discussed below.) In this sense, properties of the macrophysical level can be re-identified with microphysical features in terms of microphysical solidity and extension. Even


macrophysical solidity and extension are comprehended in this way. From this microphysical standpoint every body is understood as being solely solid and extended. Consequently, the second stage should be read as comprising microphysical extension and solidity.

To re-describe the experienced features of the macrophysical level as microphysical properties of an extended, solid substance is indeed to give an account of their union, namely that they are bound up by being interrelated features of a solid and extended entity from the microphysical perspective. Yet, this cannot be the whole story, since corpuscularian theory does not conceive a macrophysical body as an indivisible entity, but as a constitution of elementary bodies or atomic corpuscles. As indicated, a macrophysical body consists of cohering solid parts. Explaining the interrelationships between the properties of the macrophysical stage therefore leads to the question how the parts cohere which the body comprises of. As Locke puts it, there is a need to elucidate the extension of a macrophysical body, namely how its solid parts cohere.<110> Thus, if an explanation of the union of experienced properties is asked for, describing them in terms of microphysical solidity and extension simply passes on the bucket, namely to the interrelationship between microphysical extension and solidity. The identity relationship between the features of the first two stages illustrates this well. To explain the bond of properties of the macrophysical level is simply the same as to elucidate the union of corresponding microphysical features. Explaining the bond of properties of the macrophysical level in terms of microphysical extension and solidity, thus results in elucidating the union of microphysical extension and solidity. Corpuscularian theory requires a third stage of explanation.

To delineate the union of microphysical features, being conceived of as modifications of the extension of cohering solid parts, is to elucidate how the elementary corpuscles are united to one entity, i. e. to explain the coherence of corpuscles or the entity‘s extension respectively - in terms of other microphysical properties. However, as Locke painstakingly points out in the same chapter, neither corpuscularian theory nor any other hypothesis


can make intelligible how corpuscles cohere.<111> That is, for corpuscularians the bond of experienced properties, the substratum of bodies, is in fact a one-knows-not-what.

Thus in §2 Locke effectively employs a model of three stages of explanatory properties. The first one is about properties of the macrophysical level which serve as the defining features of sorts of bodies in our everyday ideas of them. The second one comprises properties of the microphysical level which correspond to properties of the first one. The third stage is of other microphysical properties, namely the substratum, which explains ultimately the union of all the other properties. In this sense, a substratum comprises the explanatory fundamental properties.

A comparison between real essence and substratum illustrates the differences. The real essence of a body comprises properties of the second level and represents an explanatory comprehension of features of the first one. That is to say, the real essence elucidates in microphysical terms properties being ordinarily known by our senses. By contrast, the substratum of a body consists of those microphysical properties which make plain the interrelationship of properties of the second stage. The substratum of a body accounts for the union of the features of which the body‘s real essence consists. A description of bodies as to the second level involves general features, e. g. being solid and extended, but it does not delineate how these properties are interrelated.

More precisely, on the second stage a body is conceived as an extended, solid entity which is being moved by impulse. For, according to Locke, features of the macrophysical level are explained as “the various modifications of the Extension, of cohering solid Parts, and their motion“.<112> Thus, mobility and its interrelationship to solidity and extension is likewise to be made intelligible on the most basic level as part of a theory of substratum. This means, an account of the union of a body‘s properties consists of an explanation of the interrelationships between microphysical solidity, extension and mobility. And, as with extension or cohesion respectively, we do not comprehend mobility. For in arguing that our understanding of finite spirits is equally in the dark as the one of bodies, Locke does not only point out that we, i. e. corpuscularians, have no clue of how corpuscles


cohere, but also none of how they pass on impulse.<113> Mobility therefore needs to be explained as well - in terms of other microphysical features. The substratum of bodies comprises the properties which elucidate the union of microphysical solidity, extension and mobility.<114>

As the discussion of Locke‘s view shows, he comprehends all bodies and all their features in the same, corpuscularian perspective. This suggests that he also conceives the union of properties for all bodies in the same terms. For example, he apparently assumes that corpuscles comprising bodies cohere in the same way. That is, he has a portrayal of the interrelationship of microphysical solidity and extension in mind which applies to every body. This understanding of the substratum of bodies is present when in the course of his explanations and at other places Locke identifies the substratum of bodies with a general stuff bodies are made of, namely with matter.<115> In other words, Locke conceives bodies as modifications of the same kind of material.<116> Thus the substratum of bodies comprises solely properties being common to all bodies, i. e. being specific for them qua bodies. A theory of the substratum of bodies consists in a general theory of what bodies are, namely in a depiction of the interrelationships of their properties in general terms.

This interpretation has advantages. It interprets satisfactorily Locke‘s corpuscularian model of the substratum of bodies in §2. Furthermore, a coherent or unambiguous conception of substratum is ascribed to him which squares his definition of substratum with its identification with matter. In addition, another problem of reading Locke as being coherent can be solved. As has been pointed out above, in §1 Locke clearly intended to introduce a notion of ’substratum‘ which applies to the substrata of all substances. Moreover, the model of §2 relates to all kinds of substrata, since it serves not only to illustrate one‘s ignorance of the substratum of bodies, but of substrata in general. Thus the substratum of human minds comprises of analogous properties explaining the union of the defining features of these substances. And, corresponding to the equation of the


substratum of bodies with matter, Locke speaks also of the substrata of human minds and of God as of general stuffs. The substratum of God is thereby portrayed as immaterial and the substratum of human minds as either material or immaterial.<117> The question arises, however, whether Locke truly has a uniform notion of substratum. It has to be a generic conception for different substrata which leaves open the ontological issue of human consciousness and which covers both matter and the immaterial stuff of God. If Locke‘s concept of substratum does not comply with this, one would effectively be thrown back on Alexander‘s two-substances-in-general view that ’substratum‘ is not a generic term for substrata, but denotes a ragbag which includes two distinct stuffs. To see that there is no such incoherence, a comparison with the concept of the melting point of a sort of bodies might help.

The point of Locke‘s uniform concept of substratum is that it serves as a generic place-holder for different, unknown substrata. ’Substratum‘ may denote various substrata, but different substrata have genuinely in common that they fulfill the same explanatory role. In short, Locke‘s concept of substratum signifies sets of properties which ultimately account for the union of properties of substances - whatever these sets of properties may (turn out to) be. His concept ’the substratum of a sort of substances‘ is thus of the same kind as ’the melting point of a sort of bodies‘. Melting points differ among classes of bodies, but the abstract concept ’the melting point of a sort of bodies‘ is the same in each case. We can refer to the unknown melting points of various classes by using in each case the same concept ’the melting point of a sort of bodies‘. The general concept of melting points denotes a particular (and perhaps unknown) melting point, if used for a specific class of bodies. Likewise, there is the generic notion of substratum. It is considered as part of the concept of a particular sort of substances, signifying a specific (and unknown) substratum. The dissimilarity is only in number, namely that there are many different melting points compared to two different substrata. Thus, it is indeed intelligible and


coherent for Locke to assert, on the one hand, that there is always one and the same idea of substratum in every idea of substances and, on the other hand, that there are two different substrata and that consciousness might be material or immaterial.

b. The Confused Idea of Substratum

The advanced account of ’substratum‘ discloses in fact what Locke means by our ignorance of the substratum of bodies. Given our everyday ideas or comprehension of bodies, one does not know what a substratum consists in. Yet, one does know the existence of substrata. For ’substratum‘ denotes the unknown properties which “cause“ or explain that the same set of properties go together. Crudely speaking, the experience of the union of properties is the experience of a substratum. Thus, the idea of substratum as being part of ideas of substances represents knowledge of the existence of a substratum.<118> One knows the existence of a set of explanatory properties, but not what these properties are.

Locke expresses this peculiar cognitive content of the idea of substratum in characterizing it as relative, obscure, and confused.<119> As other commentators have rightly said, Locke‘s portrayal of the idea of substratum as confused (or respectively as being not distinct) is decisive.<120> His other characterizations should be seen as an effort to convey the same.<121> An idea is confused or not distinct if it is “not sufficiently distinguishable from another“<122> idea. More precisely, two ideas are confused if two ideas have the same


content, even though there is a name for each idea which is taken to signify a different idea. This means, two names are intended to signify two different classes of entities, but they denote the same entities because they are related to the same idea.<123> Given Locke‘s technical notion of a confused idea, the question should be with respect to which other idea the idea of substratum is called confused.<124>

However, in this context Locke does not stick to his official notion of confused ideas. As it has convincingly been shown,<125> Locke‘s argument in II.xiii.18 entails an elucidation in which sense he understands the idea of substance, i. e. the idea of substratum, to be confused.<126> He criticizes those people who think that the term ’substance‘ stands for three different significations, namely three distinct (unconfused) ideas when used for God, finite spirits and bodies. And, given these intentions of those speakers, he recommends them to use three distinct names to indicate those three different ideas to avoid confusion in conversation, i. e. names aquivalent to ’substratum of bodies‘, ’substratum of finite spirits‘, and ’substratum of God‘. But this advice is not really meant to be a recommendation, since Locke argues subsequently that the term has only one non-distinct signification, namely that ’substance‘ stands for a confused idea. Thus Locke does not maintain what one would expect, given his technical term of confused idea. He does not claim that one uses three distinct terms with the intention to denote three different substrata, but fails to do so because one uses the three terms for one and the same idea of substratum. Instead, Locke speaks of one idea which is confused.

Yet, it becomes clear what he means. He criticizes those who think that their term ’substratum‘ stands for different substrata when referred to God, finite spirits and bodies. Thus Locke implies that the idea of substratum is confused in the sense that it does not distinguish the alleged differences between the substratum of God, finite spirits and bodies. One has only one idea which cannot be used to discriminate between various substrata. This usage of ’confused idea‘ is similar to his official notion. In both cases various classes of entities are not distinguished as intended. The dissimilarity is however


that according to the official notion various names do not discriminate between different entities and that in the case of the idea of substratum an idea does not distinguish between different entities. This account implies that the confusion of the idea of substratum expresses one‘s ignorance of substrata: one does not know what substrata consist in, but one does know that they exist.

c. Locke‘s Argument

So far only Locke‘s knowledge thesis has been analysed, but not his argument that our ordinary ideas of bodies contain a confused idea of substratum. Once he has established that ideas of substances include the idea of substratum, his argument takes off that this idea is only confused, namely that a portrayal of matter is not part of our everyday ideas of bodies. Crucially, however, for Locke there is no question that we have an idea of substratum, i. e. that we have knowledge of the existence of the substratum of bodies and spirits. This shows up in the case of bodies when he asserts without an explicit argument that even children acquire the general notion of substance and ascribe it to bodies.<127> Likewise, though he does discuss whether finite spirits have a substratum, Locke is prepared to regard denials of the existence of their substratum as confusion about one‘s ordinary conception of the human mind and not as really arguable.<128> For him everyday experience entails knowledge of the existence of the substratum of bodies and human minds. Similary, Locke argues for the idea of substratum being confused: everybody who makes plain to oneself the content of one‘s own idea, will agree that it is confused and obscure.<129> In other words, the “epistemic status“ or content of the idea of substratum is quite evident, once one becomes aware of the fact. Thus, when rejecting knowledge claims raised by the laymen or corpuscularians to have a clear idea or understanding of the substratum of substances, Locke sees himself as referring to the kind of understanding of matter which is contained in our concepts or ideas of bodies. To conceive that our ideas


of subtances include an idea of substratum and that this idea is confused, is simply to become aware of the content of the ideas.

Yet, there is another argument present in Locke‘s account. He maintains that the regular experience of the same set of properties leads to the assumption of a substratum. For Locke this account serves certainly not so much as a psychological explanation of a mental operation, but rather as a justification of our supposition of a substratum and the inclusion of a confused idea of it in our ideas of substances. One has to assume unknown features of bodies which explain how the experienced properties are interrelated, i. e. which causes the experienced bond of the properties.

Two objections illuminate the argument. True, one may say that properties making up a set have a bond in a logical sense, namely that they comprise the set. But why has one to ascribe a real or ontological bond to the properties in the sense that other features have to account for the fact that one experiences them as being together? For example, given a universe in which only elementary particles exist, a comprehensive understanding of “ordinary“ bodies would simply consist in a list of all their properties. In other words, given everyday experience of our universe, how does one know that the properties of bodies are complex interrelated and not explanatory basic? Our ordinary experience of a set of features alone does not convey the existence of unknown interrelationships between these properties. Experience of this kind does not entail the need to account for such interrelationships. There is a second problem. As we have seen, Locke drives at a notion of stuff. Imagine, however, a universe in which bodies have features in common only on higher explanatory levels, but not on the explanatory basic stage. For example, the mechanism or cause how their elementary corpuscles cohere might vary from species to species. Bodies were then not made of the same stuff, since there were no generic features of bodies. And if even specimens of the same sort, e. g. gold, do not share features on which the bond of their common properties depend, the notion of stuff evaporates completely. If the bond of properties differs from body to body, one could not even speak of a “stuff“ being specific for a sort. It shows up that the force of Locke‘s argument rests on the attractiveness and intelligibility of a corpuscularian-theory inspired understanding of bodies.

With respect to both problems, one might be tempted to reply that Locke has the recourse


to appeal to a wider range of experience. The defence runs as follows. Locke maintains that ordinary experience of bodies suggests the notion of matter.<130> In other words, everyday experience of bodies interacting, dissembling and unifying affords the assumption that bodies are of the same stuff, namely the supposition of a substratum which explains the bond of properties as to every body. This might solve the difficulty for bodies, but it is far from clear whether it does so likewise for substances like human minds. We do not have analogous experience of consciousnesses interacting, dissembling and unifying - at least not in the required sense.

d. Conclusion

The most pressing difficulties of Locke‘s theory of substratum do not relate to alleged incoherences or ambiguities of his conception of substratum, but to his justification of it, namely by his appeal to everyday experience. Yet, despite this predicament, one can agree with Locke‘s epistemological argument: to reject advanced accounts of the substratum of bodies as insufficient and to point out one‘s ordinary ignorance about their explanatory basic features, and to insist on the need of an account of matter as part of a comprehensive, scientific understanding of bodies.

As we have seen, for Locke ’substratum‘ is an explanatory concept. In this sense, ’substratum‘ is certainly not a pre-theoretic notion. On the other hand, according to Locke the assumption of a substratum is a simple or basic epistemic response of the human mind, since even children obtain the concept. This means, ’substratum‘ is directly linked to experience and has thus an extension independent of the truth of any theory which allegedly delineates the details. Whether Aristotelians, corpuscularians, Indian philosophers, laymen or children, we all have the same idea of substratum and apply it to the substrata of substances, whatever beliefs or opinions each of us has as to what a substratum consists in. In this sense, the conception of a substratum is primitive enough to serve as a placeholder for what an ideal theory identifies as the substratum of bodies. Thus, despite his corpuscularian model of the substratum of bodies, Locke‘s notion goes


beyond it. Its application is not confined to the truth of corspuscularian theory. This conception of substratum is more manifest with respect to finite spirits, since Locke does not even present a convincing or well justified hypothesis on their substratum.<131> For him one‘s everyday experience of the multiplicity of properties of bodies and of one‘s own mind entails a primitive conception of matter and consciousness. A specification of substratum is therefore part of an ideal, or true and comprehenisive, scientific understanding of species, since ideal, i. e. adequate or perfect, ideas of substances include a clear and distinct idea of their substratum.<132> Summing up, in connection with its functional conception to explain elementary experience, the substratum of bodies comprises features which are ascribed to bodies in an ideal scientific account of them.

This account will be confirmed by the theory of qualities. As will be argued in chapter five,<133> the notion of ’primary quality‘ serves to comprehend bodies from an ideal scientific viewpoint and in a specific epistemological perspective. And since the substratum of bodies is understood in the corpuscularian model as what explains the interrelationship between the three corpuscularian primary qualities solidity, extension and mobility, the substratum is understood in the same perspective as well.

This has a substantial implication as to Locke‘s philosophical programme. His discussion of substratum shows that there he is not concerned with pointing out the merits, but rather the limits of corpuscularian theory. Locke emphazises the ignorance of corpuscularians about the substratum of bodies, and equates it with the ignorance of children, laymen and Aristotelians. Likewise, when delineating that our idea of the substratum of finite spirits is not more obscure or confused than our idea of the substratum of bodies, he heightens our limited, corpuscularian understanding of bodies, namely as to the cohesion of atoms and as to the mechanism by which impulse is transmitted from body to body. In both reasonings, Locke does not justify, but only assumes and criticizes a corpuscularian account of bodies. Pace Alexander‘s view that the


Essay is intended to establish corpuscularian theory,<134> the thrust of Locke‘s argument is not to justify a corpuscularian account of matter and bodies. Rather, given his notion of substratum and its connection to ’primary quality‘, Locke assesses our everyday understanding of matter and bodies from an ideal scientific viewpoint. In this perspective one has to read the knowledge thesis that we possess only a confused idea of substratum, i. e. that we do know the existence of substratum, but not what it consists in. The epistemological aspect of this standpoint will be highlighted in the following section on resemblance.



298, II.xxiii.6; 305, II.xxiii.14.


Cp. 174f, II.xiii.18ff; 297f, II.xxiii..5. Below, the first passage will be discussed more in detail.


After arguing that every idea of substances contains an idea of substratum, Locke first establishes generally that our idea of substratum is confused, and then more specifically that the idea of a substratum of spirits is as confused as the idea of a substratum of bodies.


Cp. 5b.


298, II.xxiii.6; 305, II.xxiii.14; 316, II.xxiii.37.


175, II.xiii.19f; 295ff, II.xxiiii.2-5.


295, II.xxiii.1.


298, II.xxiii.6; 305, II.xxiii.14.


298, II.xxiii.6; 305, II.xxiii.14; 316, II.xxiii.37.


Ayers (1991), I, 161f.


This claim leaves open of course whether we can also acquire an idea of substances in cases where we have experienced only a few times a particular set of properties, once we have acquired the conception of a substratum.


175, II.xiii.19.


Alexander (1985), 204-235, esp. 221-235. He maintains that Locke identifies the substratum of bodies with matter and the substratum of spirits with an analogous spiritual substance called spirit (224). According to Alexander, a Lockean substratum is a stuff, a “substance-in-general“, which is specific for a particular kind of substances, namely for bodies or spirits. As will be argued below, Alexander is right in identifying a substratum with a general stuff, but wrong in understanding ’substratum‘ as designating a mixed bag of two kinds of stuffs. The crucial defect of his view is to disregard Locke‘s generic notion of substrata and, consequently, to lack an explanation of how Locke can effectively denote two different stuffs with his uniform conception. Thus, Alexander‘s interpretation cannot account for those passages in which Locke introduces and uses his general notion of substance. - To avoid misunderstandings, pace Alexander, Locke does not truly assert that finite spirits are of an immaterial stuff. He instead holds an agnostic position leaving open whether human minds are material or immaterial; cp. Ayers (1991), II, 42-47. God, on the other hand, an infinite spirit, is immaterial. But Locke‘s possibility of “thinking matter“, i. e. of materialism, adds weight on the prima-facie incoherences of Alexander‘s two-stuff interpretation.


Bennett (1987), 197. Ayers however argues against Bennett‘s interpretation of Leibniz. Cp. Ayers (1994), n. 59, 65.


Bennett (1965), 1f; Bennett (1971), 59-63. Bennett has recently modified his view, but in essence he re-asserted it. Cp. Bennett (1987), 197-201.


298, II.xxiii.5.


Works IV, 33-7.


Alexander, see n. above. Ayers effectively combines both Mandelbaum‘s - see below - and Alexander‘s line of interpretation. Cp. Ayer (1991), II, 37f.

For example, Ayers reads Locke‘s corpuscularian model in the same way Mandelbaum does. More precisely, in those passages (295f, II.xxiii.1f), where Locke clearly speaks of one idea of substratum, Ayers takes Locke to endorse a general notion of substratum which applies to all kinds of substances. There ’substratum‘ is an explanatory notion (1991, I, 49) signifying real essences of substances as what explains the union of properties (1991, II, 40). And at other places (e. g. 297f, II.xxiii.5), Locke is understood to use the notion of a general substratum to denote a general stuff which can be different in various substances (1991, II, 38). For example, bodies are of matter and God is of immaterial stuff. Ayers explains this ambivalence by a slip of Locke from the first into the second usage of ’substratum‘ in the course of his explanations (1991, II, 38). By contrast, the reading proposed below interprets Locke‘s model differently, reconciles both types of passages and ascribes thus to Locke an ambiguous notion.


Cp. Alexander (1985), 204-215, Ayers (1991), II, 51-64; Ayers (1994), 62-71. On the other hand, Bennett has recently rejected Ayers‘s criticism. Cp. Bennett (1987), 202-214.


295, II.xxiii.2. This is also indicated by the summary of §2 at the beginning of §3, where the idea of substratum is said to be confused and relative.


295f, II.xxiii.2.


Mandelbaum (1964), 39. Cp. Bolton (1976), 501.


Cp. Alexander (1985), 217-221 and 232; Alexander (1980), 104f; Alexander (1981), 1-6.


In some passages, Locke even clearly distinguishes the substratum and the real essence of a body. Cp. 383, II.xxxi.13. Locke‘s comments to Stillingfleet moreover suggest the here proposed reading, namely that the substratum, or substance, of bodies is matter and that their real essences are microphysical structure: the internal constitutions of bodies is depicted as the “modification of [...] substance“. Cp. Works, IV, 82. But a more elaborated reconstruction, e. g. Ayers‘s one, can deal with these passages, since one can take Locke to use ’substratum‘ here with a second meaning.


Mackie (1976), 81.


Yolton (1970), 45; Ayers in Tipton (1977), 89.


545, IV.iii.13; 589,


313, II.xxiii.30.


308-11, II.xxiii.23-27.


308-11, II.xxiii.23-27.


313, II.xxiii.30.


311f, II.xxiii.28.


Something similar must in fact account for Newtonian gravitation as well, since it is likewise regarded as inconceivable. Works IV, 467f. Cp. Rogers (1978), 225f.


Cp. 174f, II.xiii.18ff and 297f, II.xxiii.5.


Cp. 330f, II.xxvii.3f.


Works IV, 33-37; 623f, IV.x.10. In the Essay, Locke speaks of finite spirits as of immaterial spirits, suggesting that their substratum is an immaterial stuff. Cp. 305f, II.xxiii.15. There are also other passages from which some commentators concluded that Locke maintains dualism. However, in the Stillingfleet correspondence Locke clarifies his thoughts and insists on an agnostic position, namely that consciousness might be material or immaterial. Cp.Works IV, 33-37. As Ayers convincingly argues, passages which seem to imply dualism present in fact an agnostic account. Cp. Ayers (1991), II, 42-47.


Since Locke himself speaks of knowledge in this context, I do so as well. Given Locke‘s technical term of knowledge, one has, however, only sensitive or perceptual knowledge of substrata, namely when one experiences the bond of properties in sensation. If considered an idea of bodies only by itself, it is, at best, true. This means, it entails solely knowledge of the former existence of bodies and their substratum.


296f, II.xxiii.3.


Cp. Ayers (1991), II, 47f.


Cp. 296, II.xxiii.3. An idea of something is obscure if one has not properly grasped its content due to bad perceptual conditions, e. g. twilight. And having not fully grasped an idea implies not having full knowledge of what the idea is about. Cp. 363, II.xxix.2. Apparently and in a rather broad sense, this technical sense of ’obscure‘ means in the context of the idea of substratum that there are not proper conditions to grasp the substratum of bodies in detail, but only its existence. - By contrast, ’relative‘ is not a technical term of Locke in the first place and can be interpreted along the lines suggested here with respect to ’confused‘.


364, II.xxix.4.


364f, II.xxix.6.


367, II.xxix.11.


Cp. Ayers (1991), II, 47f.


174, II.xiii.18.


296, II.xxiii.2.


297f, II.xxiii.5.


296, II.xxiii.3.


330f, II.xxvii.3f.


God is a special case, since we acquire knowledge of his existence by reflecting on the idea of God, i. e. not from experience. Cp. 619-630, II.x. This is why God is virtually not discussed in the context of substratum.


383, II.xxxi.13.


Cp. 5b.


Alexander (1985), 7.

[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