by Matthew Carey Jordan
1. Introduction
The goal of this paper is to answer two questions: what, if anything, did Locke have in mind when he spoke of `substance in general'? and did Locke affirm the existence of substance in general? Concerning the first of these, I will argue that what Locke had in mind were bare particulars (or something very closely akin thereto). In the first part of this paper, I will show why this interpretation of Locke is preferable to its two main rivals. Concerning the second question, Locke was agnostic about the existence of substance in general. He may not have wished to deny its existence outright, but he certainly did not affirm it. This claim runs counter to most readings of Locke, and I will defend it in the second half of the paper. My defense will appeal to a number of texts whose significance for this debate seems to have been overlooked by many commentators. By examining Locke's view of the relationship between conceivability and possibility, in particular, we can do much to elucidate his view of substance in general.
An appropriate way to begin is by presenting two key passages from Locke's Essay Concerning Human Understanding. The first is found in a chapter on space, and provides much of the impetus for skeptical readings of Locke on substance:
The other is from a chapter titled «Of Our Complex Ideas of Substances.» Locke writes,
This passage, of course, provides us with Locke's famous definition of substance as «something, I know not what.» His use of the word `something' here, along with some key passages from Locke's letters to Edward Stillingfleet, the bishop of Worcester, have led many commentators to believe that Locke does affirm the existence of substance in general.
Before moving on to the pertinent philosophical issues, a brief comment on Locke's terminology is in order. As with `idea', Locke's use of `substance' in the Essay is not univocal. My concern here is with what he refers to as `pure substance in general' in the passage from II.xxiii cited above, viz. whatever it is (if anything) that underlies and supports the qualities of a thing.Foot note 3 For the sake of convenience, I will often follow Locke in referring to this just as `substance' or `substratum'. There is a potential confusion here, however, as Locke also uses `substance' to mean «a particular thing,» in the sense of an Aristotelian primary substance. Context is usually sufficient to make clear which usage Locke intends, but (as we shall see later) there is at least one important passage in which his meaning is less than obvious.
Commentators have suggested at least three distinct ways of understanding what Locke meant when he spoke of substance. Some (e.g. Jonathan Bennett) have argued that Locke's conception of substance is merely relational; substance is that which supports qualities, and nothing more can be said. On this interpretation, Locke's substance in general consists of «bare particulars» which do not themselves have properties, hence no positive content is (nor could it be) included in our idea of it. Peter Alexander rejects this view. He maintains that Locke's ontology includes two ultimate, irreducible kinds of substance: material and immaterial. When Locke speaks of substance in general, what he has in mind is one of these two kinds of stuff. Alexander has little to say about the nature of the latter (beyond suggesting that Locke may have held perceptivity to be its defining characteristic), but argues that material substance, for Locke, is essentially solid stuff of which all material things are composed. This is the substance as general essences interpretation. It stands in contrast to the third main interpretation of Locke's view: substance as real essences. This has been defended by Nicholas Jolley, and hinted at by others (e.g. Michael Ayers, R. S. Woolhouse). The real essences interpretation claims that when Locke speaks of the real essence of a thing, he has in mind its substratum. On this view, `real essence' and `substratum' differ in intension but not extension. The two ideas are not equivalent, but in Locke's mind they pick out the same thing.
The bare particulars interpretation is the traditional way of reading Locke on substance. It is not difficult to see why. We have already noted two key passages (Essay, II.xiii.19 and II.xxiii.2) where Locke explicitly -- and exclusively -- characterizes substance in terms of that which supports qualities. There are many others, including the following:
On the basis of texts like these, adherents of the bare particulars interpretation maintain that for Locke, substance in general is understood purely in relational terms. Through sensation and reflection, we come to have ideas of various qualities, both primary and secondary. And because we cannot conceive of such qualities existing «on their own,» we suppose that there must be some thing in which they inhere -- a thing which itself neither is a quality, nor has any essential qualities. Its nature is exhausted by its function; substance is merely that in which qualities inhere. One contemporary philosopher describes bare particulars as things which do not have qualities in the usual sense, rather, they are things to which qualities are tied.Foot note 8 Others seem more comfortable speaking of bare particulars as having properties, but emphasize that there is no property F such that it is essential to a bare particular that it have F in order to exist. As Jonathan Bennett puts it, «Lockean substratum-substance cannot have a `nature' at all.»Foot note 9
Besides being prima facie plausible,Foot note 10 the bare particulars interpretation makes sense in light of Locke's explicit concern to give an «Account of the Ways, whereby our Understandings come to attain those Notions of Things we have.»Foot note 11 We do seem to have an idea of substance, albeit an obscure and confused one, and thus Locke must explain where that idea comes from. He need not provide us with an elaborate metaphysical scheme, and except insofar as his epistemology places certain strictures on what can be known to exist, he does not do so. If Locke has bare particulars in mind when he speaks of substance in general, then (since no positive claim at all is made about their essence) he has arguably given an account of the sort necessary for his project to be complete, without engaging in the sort of «Speculations, which, however curious and entertaining,»Foot note 12 he generally resists.
The account itself, while differing in important respects from the Aristotelian/Scholastic account of knowledge of substances, is rooted in that very tradition. Like many other early modern scholars, Locke's own philosophical training was Scholastic in character, and there can be no doubt that his mature thinking was shaped by it (though there is significant debate about the degree to which it was so shaped). Locke would have been quite familiar with the idea of prime matter, «pure potentiality» which is actualized by substantial forms. It is not at all difficult to see similarities between this doctrine and that of bare particulars which exist when tied to some property or other, but have no properties of their own. Locke's own words on the subject are reminiscent of Aristotle himself. Compare the following two passages (from the Essay Concerning Human Understanding and Aristotle's Metaphysics, respectively):
Obviously, these accounts are not identical -- for one thing, Locke takes knowledge of qualities to be prior to knowledge (if we have such knowledge at all) of substance; Aristotle seems to reverse this order -- but they are not terribly dissimilar. Given Locke's own exposure to Aristotle's philosophy and the Scholastic doctrine of prime matter to which it led, as well as the deep similarities between that doctrine and the bare particulars interpretation of Locke, it would not be surprising to find that Locke indeed had something of this sort in mind when he spoke of «substance in general.» Add to this the fact that it is natural to read Locke in this way, and the bare particulars interpretation is on solid footing. Indeed, it seems clear to me that this interpretation is correct.
This is not to say, however, that there are no problems with it. First and foremost, many philosophers have found the idea of bare particulars to be so absurd that it is simply inconceivable that a thinker of Locke's stature would countenance it. The principal philosophical challenge to bare particulars, of course, is that it seems nonsensical to speak of a thing which exists, but which exemplifies no properties. This is compounded by the fact that, on any construal of the doctrine, bare particulars seemingly must exemplify the property of being such that they exemplify no properties. Thus the very notion seems to be incoherent.
This objection need not trouble us, for the following reasons. First, the mere fact that a position appears to be philosophically indefensible does not entail that it is not Locke's view. The principle of charity dictates that we ascribe to Locke the most plausible view consistent with his writings, but insofar as those writings strongly lend themselves to a bare particulars interpretation, we have good reason to understand Locke in precisely that way -- metaphysical convictions of our own notwithstanding. Second, it has been argued that the doctrine of bare particulars can, in fact, be defended against this sort of objection and is not as untenable as many philosophers seem to think it is.Foot note 15 Third, we can modify the interpretation in a way that is faithful to the basic contours of the bare particulars doctrine, but which avoids the charge of incoherence. In all the passages we have seen, Locke discusses our idea of substance in general. The content of that idea, as has been noted, is merely relational: substance is that which supports qualities. This does not rule out the possibility, however, that there are other things true of substance itself, and that we just don't know what those truths are. It may be that Locke «intends an implicit contrast between our idea of substance and that of an omniscient being ... God's idea [of substance] would be whatever enables him to see what it is for properties to be coinstantiated, that is, to be properties of a single thing.»Foot note 16 It does not seem to me that such a re-interpretation is necessary, because I do not think that Locke himself took the notion of bare particulars to be incoherent -- nor do I think he had mere ideas in mind. Nonetheless, if the incoherence objection is taken to be a serious difficulty for the bare particulars interpretation, then it should be noted that this reading avoids the problem without revising the interpretation in any significant way.
A second problem for the bare particulars interpretation, as Edwin McCann has noted, is that such a position seems incompatible with Locke's corpuscularian view of matter.Foot note 17 If material objects are ultimately composed of infinitesimal solid corpuscles, then what metaphysical work is left for bare particulars to do? It is generally agreed that Locke subscribed to Boyle's philosophy of matter, and granted this, it is not at all clear that his ontology has room for bare particulars.
In section 2.3, I will argue that belief in corpuscles is not incompatible with belief in bare particulars. More importantly, however, we should note here that any alleged conflict between Locke's corpuscularianism and his bare particulars view of substance is only problematic if Locke in fact affirmed both. I will argue later (section 3) that he did not, thereby dissolving this problem. For now, it remains to be shown that the two main rivals to the bare particulars interpretation -- substance as general essences and substance as real essences -- face far more serious difficulties.
Peter Alexander sees Locke as asserting the existence of two ultimate kinds of substance. There is material stuff, which is essentially solid, and there is immaterial stuff, whose essence is less clearly identified.Foot note 18 Everything that exists is composed of one of these two kinds of substance. This interpretation is not obviously supported by the text, but Alexander contends that when all is said and done, it is the most reasonable way to understand Locke. Indeed, if Locke is to be understood as claiming that there is such a thing as substance in general, this view is more attractive than the bare particulars interpretation, if for no other reason than that it avoids the philosophical difficulties associated with that view.
In support of his interpretation, Alexander cites two important passages from Locke's correspondence with Stillingfleet. The first is a text which raises difficulties for the real essences interpretation, which Alexander sees as the only serious alternative to his position:
Here Locke can be understood as saying that there is some stuff which underlies the real essence of a thing; God modifies that stuff -- material or immaterial, depending upon the sort of being -- and substance thus modified is the real essence of the thing. For obvious reasons, this does seem problematic for anyone who wishes to maintain that Locke identifies substance with real essence. (Prima facie, it also appears to be problematic for the bare particulars interpretation -- though only if Locke should be read as affirming the actual existence of substance in general.) However, while this passage is certainly compatible with Alexander's claim, no mention is made here of there being exactly two ultimate kinds of substance. It is a bit of a stretch to say that the passage supports the substance as general essences interpretation.
The second passage occurs in the context of the thinking matter controversy. Stillingfleet was one of a number of Locke's contemporaries who were troubled by his claim in the Essay that «GOD can, if he pleases, superadd to Matter a Faculty of Thinking.»Foot note 20 Whether or not this is so is one of the major points of contention in their correspondence, and at one point Locke writes,
According to Alexander, Locke here denies that it would be possible to remove solidity from pure substance in general and then add thought to it; this would be «the substitution of one substance for another rather than the changing of one substance into another.»Foot note 22 But a careful reading does not bear this out. Locke's concern in this section of the correspondence is to show that thinking, understood as a power had by (Aristotelian primary) substances, is not incompatible with the quality of solidity. Stillingfleet explicitly affirms the existence of substance in general, and (as Locke points out) he believes that God could transform a material substance into a spiritual substance. Perhaps Stillingfleet is an advocate of bare particulars; the possible scenario for which he allows would then be something like the following. There exists some bare particular B which has the property of solidity. An omnipotent God could remove the solidity from B without destroying it. God could then give B the power of thinking, and the transformation from material to immaterial -- according to Stillingfleet -- would be complete. Locke, however, takes it one step further. Since Stillingfleet is committed to this state of affairs being possible, Locke uses it to demonstrate the compatibility of thinking with solidity:
The first thing to notice is that Locke's point in these passages is not to articulate his own positive doctrine of substance in general, but to show that Stillingfleet's own metaphysical commitments are inconsistent with his (Stillingfleet's) claims about the (im)possibility of thinking matter. Nonetheless, even if Locke were understood to be advocating a view of the nature of substance, the passage cited by Alexander simply does not mean what he claims it does. The key statement is «if the same substance remains not, body is not changed into an immaterial substance, but the solid substance, and all belonging to it, is annihilated.» Clearly, Locke is presenting Stillingfleet with a dilemma: either (A) God must destroy the material substance and replace it with an immaterial one, or (B) God may add both the quality of solidity and the power of thinking to one and the same substance. Stillingfleet rejects (A), so he must accept (B). Alexander's interpretation ignores the dialectic, and suffers accordingly.
While no passage in Locke can be cited in direct support of the substance as general essences interpretation, there are several texts which are problematic for it.Foot note 24
Locke's statements here are not strictly incompatible with Alexander's interpretation, but it seems very odd for him to say something like this if he understands substance in general in terms of general essences. Were that his view, one would expect an amendment to this passage, e.g. «I deny that there is any such thing as one general substance which underlies everything, but let us take note that there are exactly two such kinds of substance, and everything that exists is composed of one of them.» The fact that Locke refrained from making such a comment -- and in such a natural place to do so -- strongly suggests that he believed no such thing. Even more difficult for Alexander is Locke's statement that «the general idea of substance [is] the same every where.»Foot note 26 Here there is no hint of a distinction between material and immaterial substance, and «given Locke's general carelessness about observing the distinction between ideas and the things they are ideas of, he may be read as saying that substance or substratum is the same in [both].»Foot note 27 In the absence of any compelling textual evidence for Alexander's interpretation, this statement by Locke seems to count decisively against it.
Commentators who emphasize Locke's corpuscularianism as a key to understanding his philosophy -- and there are many such commentators -- often incline toward the substance as real essences interpretation of Locke. On this reading, Locke's distinction between substance in general and real essence is merely conceptual. In Michael Ayers' words, there is a «merely logical distinction ... between substance and real essence ... There are not two underlying levels, first the real essence, then, beneath it, the substance.»Foot note 28 Since the observable macroproperties of physical objects are, for Locke, determined by their microphysical structures,Foot note 29 the appeal of this interpretation is readily apparent. Furthermore, since substance in general is supposed to be whatever it is that supports qualities, and since Locke explicitly speaks of «real Essences ... on which all the properties of the Species depend, and from which alone they all flow,»Foot note 30 the case for this interpretation is quite strong. And as with the general essences interpretation, this way of understanding Locke is considerably more attractive, philosophically speaking, than the bare particulars interpretation. The «corpuscular hypothesis ... supplied [Locke] with a more significant content for the concept of body than philosophical tradition offered. There is a conceptual necessity about our concept of body which is more meaningfully filled by `insensible particles' than by `substratum'.»Foot note 31 The real essences interpretation also has the advantage of simplicity; it would be explanatorily superfluous to posit the existence of bare particulars and real essences, when one thing could do the work of both.
One difficulty for this interpretation, however, is that when Locke speaks of qualities depending on substrata and on real essences, he seems to have two different kinds of dependence in mind. Nicholas Jolley cites II.xxiii.3 of the Essay, where Locke writes,
and notes that it is «surprising,» on the real essences interpretation, «that he should have continued to describe the relationship between substance and the observable properties in terms of `inherence', when what he was actually talking about was causal dependence.»Foot note 32 When Locke speaks of our inclination to posit the existence of a substratum as a support for qualities, he consistently uses language of this sort (`inherence', `subsisting', etc.). We have already seen a number of examples of this in section 2.1. When Locke speaks of real essences, on the other hand, he consistently speaks of qualities which flow from those essences:
This does not entail that the two dependence relations could not possibly be the same, but if Locke had the substance as real essences interpretation in mind, he certainly could have expressed himself much more clearly. When Locke's words are taken at face value, it seems obvious that substance in general is that in which qualities inhere; qualities depend upon it insofar as they are not things which can subsist of themselves. The dependence here is ontological in character. Real essences, on the other hand, are what determine which observable qualities a thing has; qualities depend upon real essences insofar as a real essences cause things to have such-and-such particular qualities. This sort of dependence is determinate in character. Locke is quite plain about this:
Thus we are faced with the same textual difficulty that plagues the general essences interpretation: in all of Locke's work, there is no passage in which he actually affirms it, in spite of abundant opportunities to do so. In the words of one commentator, «on the whole, [the substance as real essences interpretation] leaves entirely unexplained large and central stretches of both the Essay and the Stillingfleet correspondence in which the notion of substance is discussed.»Foot note 37 One might also note that the notion of real essence, while not as hotly contested an issue as substance, is also discussed at length in the Essay and in the Stillingfleet correspondence -- yet time and time again, Locke fails to make explicit his alleged identification of real essence with substance in general.
Even worse for the real essences interpretation is that Locke does make explicit statements which seem to contradict it. We have already seen that Locke believes there is one idea of substratum, common to all things, but «if substance were identical with real essence, substance in gold would not be the same as substance in silver; what makes them particular substances is their different real essences or inner constitutions.»Foot note 38 If the real essences interpretation is correct, then Locke's words cannot be taken at face value when he claims that «the general idea of substance [is] the same every where.»Foot note 39
On this point, we should also note a criticism Locke makes of the Aristotelian doctrine of substantial forms:
There are two things to notice in this passage. First, Locke again makes it clear that real essences determine the sensible qualities of a thing, and therefore determine the species to which a thing belongs. So the substance as real essences interpretation would again contradict the earlier claim that substance is the same in all things. Second, it is interesting to see that Locke refers to something «they know not what.» This phrase is so closely tied to Locke's view of substance in general that it is hard to explain its occurrence here as mere coincidence -- especially since the fourth edition of the Essay (the one being cited here) was published after the controversy with Stillingfleet. Locke certainly knew what this phrase would remind his readers of. If `they know not what' is meant to conjure thoughts of a ubiquitous something in which qualities inhere, then Locke cannot have been identifying real essence with substance in general.
Textual issues aside, there are other problems for the real essences interpretation. The great appeal of this interpretation is that it makes sense of how Locke's corpuscularianism fits into his doctrine of substance, and with it, how his doctrine of substance fits into his overall project of offering an alternative to the Scholastic and Cartesian views of the world. However, the emphasis on corpuscularianism that drives this interpretation leads, itself, to two serious difficulties. First, it simply pushes the problem of substance in general back a step. We have already seen (in section 2.1) that the problem arises «because we cannot conceive how sensible qualities should subsist by themselves.»Foot note 41 Of course, this inconceivability is not unique to sensible qualities. The problem cannot be solved by appealing to the physical constitution of material substances, because that would leave us wondering what it is in which microproperties inhere. If the fact «that a certain number of ... simple Ideas go constantly together; which being presumed to belong to one thing ... are called so united in one subject, by one name»Foot note 42 is a philosophical puzzle when those ideas are the furriness, quadrupedality, and meowing of a cat, then it is also a puzzle when those ideas are the solidity, extension, and shape of a corpuscle. If there needs to be an explanation for the unity of the former set, there needs -- it would seem -- to be an explanation for the unity of the latter set.Foot note 43
Second, this interpretation seems to ignore that Locke is a dualist. Even if he affirms materialism about human persons, there can be no serious doubt that he acknowledges the existence of at least one immaterial substance: namely, God.Foot note 44 And while the substance as real essences interpretations is in line with Locke's philosophy of body, it is very difficult to see how it could be used to make sense out of the relationship between substance in general and particular immaterial substances. The real essence of a body is its microphysical structure. What, then, is the real essence of an immaterial thing? It is not at all clear how one would begin to answer this question, and it is perhaps telling that defenders of the real essences interpretation tend to focus exclusively on Locke's view of material substances. Perhaps there is an answer available; if so, it escapes me. For now, we may note that Locke was a metaphysical dualist, and any interpretation of his (somewhat opaque) views on substance in general must account for his (quite clear) views on substance in general. The substance as real essences interpretation fails to do this, and should be rejected.
Thus far, our goal has been to answer the question, when Locke speaks of `substance in general', what does he have in mind? I have argued that what he has in mind is precisely what interpreters have traditionally thought: bare particulars (or something closely akin to them). The other leading interpretations -- general and real essences -- are implausible, mainly for textual reasons. Neither position is explicitly affirmed by Locke, even though he has abundant opportunities to make his position on the matter clear. Furthermore, both of these positions seem to contradict what Locke actually does say on the topic of substance. Turnabout is fair play, however, and in the next part of this paper, I will argue that Locke was agnostic about the existence of substance in general -- in spite of the fact that this flatly contradicts a number of his own statements on the topic. My contention is that careful exegesis, along with an understanding of how Locke viewed the relationship between inconceivability and possibility, will be sufficient to show that he did not affirm the existence of substance in general.
In Locke's correspondence with Stillingfleet, the topic of substance (in particular, its knowability) takes center stage. As mentioned earlier, Stillingfleet was a professional clergyman. He had published a book called A Discourse in Vindication of the Doctrine of the Trinity in which he accused Locke (though he did not name him outright) of compromising orthodox Christian faith, Trinitarianism in particular. Since Stillingfleet understood divine triunity according to the traditional conception of three persons existing in one substance, he saw Locke's sarcastic comments about substance in the Essay as an implicit attack on the rationality of the Christian understanding of God. And there can be no doubt that Locke was quite sarcastic in places. We have already noted (in the introduction to this paper) his famous analogy of the Indian who claimed that the Earth was supported by an elephant, the elephant by a tortoise, and the tortoise by «something, I know not what,» just as philosophers are wont to claim that substance is what undergirds the sensible qualities of things. Locke went on to say of «those who lay so much stress on the sound of these two Syllables, Substance» that «It helps not our Ignorance, to feign a Knowledge, where we have none, by making a noise with Sounds, without clear and distinct Significations,»Foot note 45 and that
Later, in the chapter on substance, Locke writes,
On the basis of statements like these, Stillingfleet accused Locke of «almost discard[ing] Substance out of the reasonable part of the World.»Foot note 48 In his reply to Stillingfleet, Locke is quite anxious to rebut this charge:
Locke acknowledges that he does not think we can have a clear idea of substance in general, but insists that this is not the same as denying its existence. Indeed, he sees a striking similarity between his own description of substance as something we posit because we can't conceive of qualities subsisting alone, and a statement of Stillingfleet's about such subsistence being «repugnant» to us:
In light of such statements, it is easy to see why most commentators see Locke as affirming the existence of substance in general, and it is fair to say that there is a burden of proof on anyone who suggests otherwise. Michael Ayers is one who is dismissive of such suggestions:
Of course, in the correspondence with Stillingfleet, it cannot be denied that Locke is occasionally insincere and sometimes secretive. As Lex Newman has pointed out, «Given the charges of heresy that lurk, it is understandable that Locke would resort to being cagey.»Foot note 52 But we need not attribute to Locke either ambivalence or confusion. In spite of the «disclaimer» referred to by Ayers (and quoted above), Locke is remarkably consistent in his remarks about the existence of substance in general. When pressed by Stillingfleet, he repeatedly agrees that we are unable to conceive of qualities subsisting of themselves. He says (again, as quoted above) that he has the very same reasons for believing in substance as Stillingfleet does.Foot note 53 But never does he come out and say, «Yes, Bishop, there is a substratum.» Given the amount of space dedicated to this topic, and Stillingfleet's repeated badgering of him on this issue, it would be astonishing if Locke really believed in the existence of substance in general and yet completely failed to say so explicitly. Since he did not say so, it is reasonable to conclude that he did not, in fact, believe in it.Foot note 54 There are three main (kinds of) texts in which Locke appears to deny this. The first are passages like the one noted above, where Locke grounds the existence of substance in general on the inconceivability of qualities existing without a support, and draws parallels between this inconceivability and the repugnancy of which Stillingfleet speaks. Many commentators have construed this as an argument to the effect that the existence of a substratum is logically necessary. The second are passages where he distinguishes between agnosticism about the being of substance from agnosticism about our idea of substance. The third is one isolated passage where Locke comes very close to an explicit affirmation of the existence of a substratum. I will deal with these texts in reverse order.
Early in the correspondence with Stillingfleet, there occurs a passage which many have seen as conclusive proof that Locke does believe in substance. He writes, «having every where affirmed and built upon it, That a Man is a Substance, I cannot be supposed to question or doubt of the being of Substance, till I can question or doubt of my own being.»Foot note 55 It is surprising, however, that this statement has been taken to be so decisive, since it contains a rather blatant equivocation. The charge against Locke is that he has undercut the rationality of belief in substance in general, not substances, understood as particular beings. Clearly, the latter usage is being employed here. It is obvious from the Essay that Locke believes in substances in this sense, and he is correct to say that rejecting the existence of such beings would be tantamount to denying or questioning his own existence. That, however, is not what is at issue; this passage simply changes the subject. We must look elsewhere for evidence that Locke's sarcastic comments about substance in the Essay do not capture his real view. Before moving on, however, it is worth noting that Locke's equivocation here is further evidence of his unwillingness to affirm that there is such a thing as substance in general. If he wished to acknowledge its existence, why not say so plainly here, of all places?
Immediately prior to the statement just considered, Locke makes a claim which he repeats several times in letters to Stillingfleet. He says,
Now, the first thing to notice here is that Locke again refrains from explicitly affirming that there is such a thing as substance. As before, it is fair to wonder why he does so, since it is clear that a single, unambiguous affirmation of belief in substance in general would be sufficient to end the controversy. Second, he acknowledges that our idea of substance is obscure in virtue of its source; neither sensation nor reflection can provide us with a simple idea of substance, and the idea is not «suggested» to us by other ideas, like the ideas of existence and unity are.Foot note 57 Locke's point here is that he has not denied that there is such a thing as a general substratum, which is certainly correct. A non-denial, however, hardly amounts to an affirmation. And if our idea of substance is on such admittedly shaky ground, it is no wonder that Stillingfleet continued to press Locke on this subject. Finally, Locke's consistent use of words like `accustom' and `suppose' in this context suggest that he is less than totally confident in the rational justifiability of belief in substance in general.
Stillingfleet had picked up on this last point, asking Locke whether this «Custom [is] grounded upon true Reason or not?»Foot note 58 In reply, Locke repeats what he had already said about our inability to conceive of qualities subsisting alone, «Which I think is a true Reason, because it is the same your Lordship grounds the Supposition of a Substratum on, in this very Page; even on the repugnancy to our Conceptions, that Modes and Accidents should subsist by themselves.»Foot note 59 Clearly, what Stillingfleet wanted to know was whether we are justified in believing in a substratum. Locke avoids this question, instead acknowledging that we cannot help but form such a belief when we reflect on qualities themselves. His point is psychological, not epistemological. Whether this «true reason» is a good reason is not addressed. Locke's continued strategy of pussyfooting around the real issue does give us good reason, however, to suspect that he is not being entirely forthright about his own views. Stillingfleet is seeking one of two things: either an admission from Locke that the arguments of the Essay undercut the rationality of belief in substance in general, or an outright, unequivocal statement to the effect that the existence of substance in general can be established. Locke understands this, but never budges from his position that our idea of substance is grounded in custom and supposition:
Again, we see the same features of Locke's account emerge: (1) substance is (merely) supposed, (2) no admission of certain knowledge concerning substance, and (3) a noncommittal attitude toward the actual existence of substance in general. Newman gets it exactly right when he says that the consequence of these passages «is not to upgrade the certainty of Locke's account of substratum from supposition to demonstrative reason; the effect is to downgrade the certainty which Stillingfleet purports, to that of mere presumption.»Foot note 61
In sections 2.1 and 3.2 of this paper, we have seen that Locke explains our custom of supposing the existence of a substratum by noting that we cannot conceive of qualities existing «unsupported.» Because Locke speaks of inconceivability, many commentators have understood him to be making a substantive modal claim. As Peter Alexander puts it, «the statement that we cannot imagine qualities existing without support [is] making a logical, rather than a psychological point, that is, [it is] the assertion that it is inconceivable because logically impossible that qualities should exist without support.»Foot note 62 Prima facie, this claim is strengthened by the fact that, as we have seen, Locke claims parity between his notion of inconceivability and Stillingfleet's notion of repugnancy, and the latter clearly does ground belief in substance in reason What many commentators have overlooked, however, is that while Locke does not explicitly develop an account of modal epistemology, he does seem to have clear thoughts about what we may (and may not) infer about the world on the basis of inconceivability. Loosely, his view is as follows. There are at least two kinds of inconceivability; we can label them `positive' and `negative'. A state of affairs is positively inconceivable (PI) when that state of affairs is incompatible with knowledge we have of necessary truths (e.g. the essential properties of geometric shapes). States of affairs which are PI can be known to be impossible. On the other hand, a state of affairs is negatively inconceivable (NI) when we lack a clear and distinct idea of how that state of affairs could come to be. Thus positive inconceivability stems from our knowledge, but negative inconceivability stems from our ignorance. If a state of affairs is NI, it may, for all we know, be possible. We certainly cannot know that it is impossible, and thus we must be agnostic about its modal status.
Let S be the state of affairs described by the proposition `qualities exist of themselves, without any substratum in which to inhere'. For Locke, S is NI. We cannot know S to be impossible, and we should be agnostic about whether or not S is actual. Therefore Locke is guilty of precisely the charge of which Stillingfleet accuses him: there can be no certainty about the existence of substance in general, and hence no knowledge of it, either. The reason Stillingfleet is unable to make the charge stick, however, is that when he (Stillingfleet) speaks of S being repugnant or inconceivable, he means that S is PI. Locke avoids the charge of heresy by using the same words as Stillingfleet, but changing their meaning; what he means is that S is NI.Foot note 63 An examination of some key texts will bear this out.
There are two principal contexts in which Locke makes clear his views on positive and negative inconceivability. One is the thinking matter controversy, where he argues that, for all we know, God could create a purely material being with the attribute of thought. The other is his argument for the existence of God and his defense of ex nihilo creation. Here is what Locke says:
Notice that in each of the first two passages, Locke juxtaposes the modal claim in question with a necessary truth from geometry, our paradigm for knowledge of necessity, possibility, and impossibility. In the first one, he counsels agnosticism concerning the existence of thinking, merely material beings. In the second, he states that what we know about thought entitles us to the strong claim that thinking cannot come from non-thinking; the denial of this is PI. And in the third passage, he fends off a potential criticism by noting that creation ex nihilo is merely NI. It may well be inconceivable, but only because of our own cognitive limitations. In sum, Locke holds that inconceivability is sometimes an indicator of necessary truth, but only in the case of «certain Relations, Habitudes, and Connexions, so visibly included in the Nature of the Ideas themselves, that we cannot conceive them separable from them, by any power whatsoever.»Foot note 67 On the basis of everything Locke says about qualities, inherence, and substrata, it should be clear that the so-called «logical argument» for the existence of substance in general does not meet Locke's own criteria for logical necessity. We cannot be certain that there is such a thing, and Locke is entitled to nothing stronger than agnosticism on the subject. He should be read as affirming precisely that.
If any doubt remains about Locke's real attitude toward the existence of substance in general, there is one more passage worth noting. It has not received any significant attention in the secondary literature, which is surprising, since it speaks directly to the issue at hand. We have seen that interpreters of Locke who wish to argue that he affirmed the existence of substance in general typically cite his repeated statements of the following kind: «Your lordship ... concludes that there is substance, «because it is a repugnancy to our conceptions of things ... that modes or accidents should subsist by themselves;» and I conclude the same thing, because we cannot conceive how sensible qualities should subsist by themselves.»Foot note 68 Following Lex Newman, I have argued that the comparisons Locke makes between his own views and Stillingfleet's are intended to weaken confidence in the latter, not to strengthen it in the former. There is another comparison made by Locke, however, which seems to settle the matter once and for all. After insisting to Stillingfleet that «the being of Substance is not shaken by what I have said,»Foot note 69 he goes on to say, «It cannot be doubted but there are distinct Species of separate Spirits, of which yet we have no distinct Ideas at all: It cannot be questioned but Spirits have ways of Communicating their Thoughts, and yet we have no Idea of it at all.»Foot note 70 Prima facie, Locke appears to be agreeing with Stillingfleet here; there are, he suggests, indubitable truths about which «we have no distinct Ideas at all.» But I do not think it is anachronistic to point out that it can be doubted that «there are distinct Species of separate Spirits,» and it can be questioned whether «Spirits have ways of communicating their thoughts.» Stillingfleet himself may have been unwilling to subject these claims to criticism, so in that respect Locke's examples are well chosen. But insofar as Locke's beliefs about substance are what is at issue, these examples are quite odd -- if what he wishes to communicate is a sincere, confident affirmation that there is such a thing as substance in general. Indeed, Locke had already addressed the issue of our knowledge of spirits in the Essay:
If the very existence of angels and other spirits is a matter about which we cannot acquire certainty (apart from divine revelation), then surely the same is true of the distinctions between such species and their capacity for interspiritual communication.
This is a very important point, and even at the risk of repetitiveness, the dialectic needs to be made clear. Stillingfleet has accused Locke of undercutting the rationality of belief in substance. This is tantamount to heresy, since if substance is called into question, so -- in Stillingfleet's eyes -- is the doctrine of the Trinity. Understandably, Locke wants to avoid being labeled a heretic, but he also wants to promulgate his philosophical views, including his agnosticism about the existence of substance in general. So in the correspondence with Stillingfleet, he softens the critical claims he made in the Essay and dances around Stillingfleet's accusation, while never actually affirming (or denying) that it is rational to believe in substance. Throughout, he hints at his real view, including in the passage being considered here. The existence of substance in general, he states, is just as certain as the existence of «distinct Species of separate Spirits.» That is to say, it is not certain at all.
Matthew Carey Jordan
The Ohio State University
jordan.384 [at] osu.edu
[Foot Note 1]
John Locke, An Essay Concerning Human Understanding, ed. Peter H. Nidditch (New York: Oxford University Press, 1975), II.xiii.19.
[Foot Note 2]
Essay, II.xxiii.2.
[Foot Note 3]
To speak of substance as «underlying» and «supporting» qualities is, of course, to be rather vague. These notions will be clarified below; at this stage, however, nothing more precise needs to (or should) be said.
[Foot Note 4]
Essay, I.iv.18.
[Foot Note 5]
Essay, II.xxiii.15.
[Foot Note 6]
John Locke, Letter to the Bishop of Worcester (London: A. and J. Churchill, 1697), text-fiche, pp. 10-11.
[Foot Note 7]
John Locke, The Works of John Locke, vol. 4 (London: Thomas Davison, 1823), 445.
[Foot Note 8]
J. P. Moreland, Universals (Chesham, UK: Acumen, 2001), 153. Moreland takes the «tied to» relation to be unanalyzable. Whatever it is, it differs from the exemplification relation; for Moreland, this is what accounts for bare particulars being genuinely bare.
[Foot Note 9]
Jonathan Bennett, Locke, Berkeley, Hume: Central Themes (Oxford: Oxford University Press, 1971), 62. Note the similarity between this statement and Gustav Bergmann's assertion that «Bare particulars neither are nor have natures» (cited in Moreland, 148).
[Foot Note 10]
I am speaking here of textual/interpretative plausibility, not philosophical plausibility.
[Foot Note 11]
Essay, I.i.2.
[Foot Note 12]
Ibid.
[Foot Note 13]
Essay, II.xxiii.1-2.
[Foot Note 14]
Aristotle, Metaphyics 1028a20, my italics.
[Foot Note 15]
Offering such a defense lies outside the scope of this paper; one good example is chapter 7 of Moreland's Universals.
[Foot Note 16]
Nicholas Jolley, Leibniz and Locke (Oxford: Clarendon Press, 1984), 77.
[Foot Note 17]
Edwin McCann, «Locke's Philosophy of Body,» in The Cambridge Companion to Locke, ed. Vere Chappell (Cambridge: Cambridge University Press, 1994), 80-1.
[Foot Note 18]
As noted earlier, Alexander sees perceptivity as the most likely candidate for the essence of Lockean immaterial substance; see his Ideas, Qualities and Corpuscles: Locke and Boyle on the External World (Cambridge: Cambridge University Press, 1985), 233-4.
[Foot Note 19]
Locke 1823, 82.
[Foot Note 20]
Essay, IV.iii.6.
[Foot Note 21]
Locke 1823, 470.
[Foot Note 22]
Alexander 1985, 231.
[Foot Note 23]
Locke 1823, 471.
[Foot Note 24]
Edwin McCann has argued that the substance as general essences interpretation is ruled out by Locke's commitment to the possibility of thinking matter, since God's superaddition of thought to a material object would necessarily entail that the «object would have two distinct natures, and would belong to each of the two general kinds of substance» (McCann, 80). This objection is mistaken, however, since Locke is quite clear that thought is not an essential attribute of immaterial substance; for Locke, the key contrast is not between material and spiritual substances, but between material and immaterial substances. For Locke, thinking is essential to spirits (qua spirits), but immateriality is not. On Alexander's interpretation, the two ultimate kinds of substance are material and immaterial, so there is no inconsistency here.
[Foot Note 25]
Locke 1697, 52.
[Foot Note 26]
Locke 1823, 33.
[Foot Note 27]
McCann, 80-1.
[Foot Note 28]
Michael R. Ayers, «The Ideas of Power and Substance in Locke's Philosophy,» Philosophical Quarterly 25 (1975): 16-7. It should be noted that Ayers has distanced himself from this interpretation in the years since the article was published. In Locke: Epistemology and Ontology (New York: Routledge, 1993), he writes, «Now the unknown cause of the union of the observable properties of iron, according to corpuscularian theory, is a particular or determinate constitution or modification of matter: precisely what Locke called the `real essence' of the species. Yet in general Locke does not seem to have thought of the unknown substance and the unknown real essence of anything as identical» (Vol. II, p. 40). Nonetheless, his earlier work is often cited in support of the real essences interpretation, and with good reason.
[Foot Note 29]
See, e.g. Essay, II.xxxii.24: «all that the most expert Man knows, are but few, in comparison of what are really in that Body, and depend upon its internal or essential Constitution...»
[Foot Note 30]
Essay, III.v.14.
[Foot Note 31]
John W. Yolton, Locke and the Compass of Human Understanding (Cambridge: Cambridge University Press, 1970), 44.
[Foot Note 32]
Jolley, 86.
[Foot Note 33]
Essay, II.xxxi.11.
[Foot Note 34]
Essay, II.xxxii.24.
[Foot Note 35]
Essay, III.iii.18.
[Foot Note 36]
Essay, III.iii.15.
[Foot Note 37]
McCann, 83.
[Foot Note 38]
Alexander 1985, 216.
[Foot Note 39]
Locke 1823, 33. The real essences interpretation cannot be salvaged here by claiming that the «idea of substance» is just the abstract idea of real essences. We have already seen that Lockean real essences are «inner constitutions» or «modifications,» and the passage cited here continues, «the general idea of substance being the same every where, the modification of thinking, or the power of thinking joined to it, makes it a spirit, without considering what other modifications it has ... on the other side, substance, that has the modification or [sic] solidity, is matter» (my emphases). Substance in general is what gets modified; real essences are specific modifications thereof. Also, as will be discussed later, whatever «idea of substance is the same every where» needs to apply to immaterial as well as material things. It is not at all clear that the substance as real essences interpretation can do so.
[Foot Note 40]
Essay, III.iii.17.
[Foot Note 41]
Locke 1823, 445.
[Foot Note 42]
Essay, II.xxiii.1.
[Foot Note 43]
It should be pointed out that this line of thinking does not commit Locke to the view that there is a substratum; as I have mentioned, I take him to be agnostic on that issue -- perhaps the problem described here is simply insoluble (at least for us). It does, however, seem to rule out corpuscularianism as a solution to the problem as Locke himself describes it.
[Foot Note 44]
See Essay, IV.10.
[Foot Note 45]
Essay, II.xiii.18.
[Foot Note 46]
Essay, II.xiii.20.
[Foot Note 47]
Essay, II.xxiii.2.
[Foot Note 48]
Edward Stillingfleet, «The Objections against the Trinity in Point of Reason Answer'd» in Three Criticisms of Locke (New York: Georg Olms Verlag, 1987), 234.
[Foot Note 49]
Locke 1697, 6.
[Foot Note 50]
Locke 1697, 22. The «two passages» mentioned here are Stillingfleet's statement that «... it is a Repugnancy to our first Conceptions of things, that Modes or Accidents should subsist by themselves, and therefore the Rational Idea of Substance is one of the first, and most natural Ideas in our minds» (Stillingfleet, 236), and Locke's claim that «`because we cannot conceive how simple ideas of sensible qualities should subsist alone, or one in another, we suppose them existing in, and supported by, some common subject.' Which I, with your lordship, call also substratum» (Locke 1823, 13).
[Foot Note 51]
Ayers 1975, 3-4.
[Foot Note 52]
Lex Newman, «Locke on the Idea of Substratum,» Pacific Philosophical Quarterly 81 (Sept. 2000): 317.
[Foot Note 53]
We shall see later (section 3.3) that this fact is meant to undermine Stillingfleet's position.
[Foot Note 54]
Note, however, that we may conclude merely that Locke did not affirm the existence of substance in general. His silence here does not give us reason to think he denied its existence; a non-affirmation is not the same thing as an outright rejection. Again, Locke should be read as an agnostic about whether there is such a thing as substance in general.
[Foot Note 55]
Locke 1697, 32.
[Foot Note 56]
Ibid.
[Foot Note 57]
Essay, II.vii.7.
[Foot Note 58]
Stillingfleet, 236.
[Foot Note 59]
Locke 1697, 34.
[Foot Note 60]
Locke 1823, 444-5.
[Foot Note 61]
Newman, 29.
[Foot Note 62]
Peter Alexander, «Locke on Substance-In-General,» Ratio 22 (1980): 97.
[Foot Note 63]
This interpretation is similar, though not identical, to that defended by Lex Newman in the paper cited earlier.
[Foot Note 64]
Essay, IV.iii.6.
[Foot Note 65]
Essay, IV.x.5.
[Foot Note 66]
Essay, IV.x.19.
[Foot Note 67]
Essay, IV.iii.29.
[Foot Note 68]
Locke 1823, 445.
[Foot Note 69]
Locke 1697, 33.
[Foot Note 70]
Ibid.
[Foot Note 71]
Essay, IV.iii.27, my emphasis.