SOURCE: “The Substance of Berkeley's Philosophy,” in Berkeley's Metaphysics: Structural, Interpretive, and Critical Essays, edited by Robert G. Muehlmann, The Pennsylvania State University Press, 1995, pp. 89-105.

[In the following essay, Muehlmann thoroughly analyzes Berkeley's central metaphysical doctrines and some of the motivations behind them and concludes that many who have read his principles have been misled.]

In the Philosophical Commentaries, the “juvenile” Berkeley enthusiastically sketches out a bundle analysis of finite minds: a mind is constituted of episodes of volition and occurrences of ideas.¹ But it is clear that Berkeley endorses a substance analysis by the time the Principles appears. As early as PR 2 he says that a mind is “a thing entirely distinct” from its ideas; and numerous additional passages in both the Principles and Dialogues give voice to the position that finite minds are active substances distinct from, but both “causing” (some of) and “supporting” (all of), their ideas. That dramatic change in his position raises the question to which this essay is addressed. Why does Berkeley endorse the substance analysis of mind when he writes the Principles?

By the time I am able to provide an answer, I will have shown that this question is not equivalent to its negatively expressed counterpart, obtained by replacing ‘endorse the substance analysis’ with ‘abandon the bundle analysis’. Although it is obvious both that the notebooks' account of mind is inconsistent with the Principles' (since the former affirms what the latter denies and conversely) and that Berkeley does endorse a substance account of mind in the Principles, I show that, in one important sense, he does not abandon the bundle analysis. Rather, as I say, he conceals it. He conceals it in the sense that nearly all of the philosophical work provided in Berkeley's prepublication bundle, or congeries, account of finite mind is camouflaged as work now done by his published substance account; indeed it is difficult to find any ontological role, in Berkeley's two major works, that he explicitly and exclusively assigns to finite mental substances. Because he thus conceals the congeries account, I must address a second question. Why does Berkeley conceal the congeries account of mind in the writing of the Principles (and Dialogues)? The principal answer to this second question also serves as part of the answer to the first.

To quicken the sense of the puzzle raised by the first question, several things are noteworthy. One is that Berkeley is fully aware of the difficulties posed by his introduction of substances into the analysis of mind. He devotes considerable energy, for example, in the attempt to deflect one such difficulty, a difficulty exploited in the “parity argument.” This argument can be stated succinctly in ad hominem form: Berkeley's objections to material substances apply, with equal force, to mental substances; and parity of reasoning, therefore, should force him to reject the latter.² Nor is this the only difficulty Berkeley addresses. A related one arises within his epistemology.

As Berkeley reads his adversaries, their position is that substances, independently existing things, are not directly perceived; rather, they are known indirectly and inferentially: we directly perceive intervening “phantasms” or “sensible appearances,” and then our understanding, our reasoning based on these “appearances,” provides us with knowledge of substances. In stark contrast, Berkeley holds, first, that those so-called phantasms and mere appearances are actually real things (or ideas, as he prefers to call them when stressing their mind-dependence). He holds, second, that we know real things directly and noninferentially; and he holds, finally, that we also know our selves and our mental operations directly: we know the meanings of ‘my self’ and ‘this volition’ without the intervention of ideas and without inference. I have not the space to discuss these three points in detail, but for my purposes here the following will suffice.

While Berkeley devotes a great deal of attention to the first two, he seems to focus insufficient attention on the third. It is given a scant few lines in the first editions of the Principles and Dialogues and emerges only in a thinly stated doctrine of notions that Berkeley adds to the later editions. Underlying this doctrine, however, is a thesis he does address in some detail, the thesis that particular episodes of volition are active—that they are casually efficacious mental operations.³

The puzzle raised by the central question is thus magnified by the fact that Berkeley appreciates the difficulties entailed by his introduction of mental substances. But the puzzle becomes more intense when we realize that the notebooks' congeries analysis of mind is firmly grounded in Berkeley's antiabstractionism; that he stresses, particularly in the introduction but also in the main body of the Principles, the importance of antiabstractionism to his philosophy; and that the substance analysis of mind, emphasized at PR 2 and elsewhere, nonetheless runs completely counter to his attack on abstract ideas.

As if this were not enough, however, the puzzle is made acute by two additional facts. The first is that it is difficult to see what ontological role is played by the notion of mental substance in Berkeley's philosophy; mental substances do play an ontological role, but the role they play, while related to one of the traditional roles his adversaries assign to substances, constitutes a revolutionary departure from their analyses. The second fact sharpening the puzzle is that in many places where Berkeley gives voice to the substance account of mind, the notebooks' congeries analysis hovers, inconsistently, in the background. Numerous sections of the Principles make this second fact apparent, and we will do well to pay close attention to the more significant ones.

At PR 98 Berkeley says, “[W]hoever should go about to divide in his thoughts or abstract the existence of a spirit from its cogitation, will, I believe, find it no easy task” (cf. PC 650-52). It will be no easy task because, as Berkeley argues so forcefully in his introduction to the Principles, such abstract ideas cannot exist.⁴ Yet since cogitation consists in the having, the occurrence in a thinker, of ideas, it is hard to see how the distinctness of a mind from its ideas can be squared with Berkeley's assertion at PR 98.

In another significant section, PR 27, Berkeley observes that some philosophers hold that they “can frame the idea of … power or active being” and that they have “ideas of two principal powers, marked by the names will and understanding, distinct from each other as well as from a third idea of substance or being in general, with a relative notion of its supporting or being the subject of the aforesaid powers, which is signified by the name soul or spirit.” But Berkeley then continues by echoing his notebooks' disdain for such philosophers: “This is what some hold; but so far as I can see, the words will, soul, spirit, do not stand for different ideas, or in truth, for any idea at all, but for something which is very different from ideas, and which being an agent cannot be like unto, or represented by, any idea whatsoever” (PR 27). The main point of this section is to bring out the fact that our awareness of spirit is not ideational—that being an agent, a spirit cannot be represented by something that is passive, such as an idea. But for my purposes here, we need only note that implicit in this section is Berkeley's rejection of the traditional substance analysis of spirit—his rejection of the view that a spirit is a “substance or being in general” and that it stands in a relation of inherence, of support, to the “principal powers” of will and understanding.⁵

At PR 11 Berkeley dismisses that “much ridiculed notion of materia prima, to be met with in Aristotle and his followers”; and, significantly, he says (with emphasis added), “[H]ere I cannot but remark, how nearly the vague and indeterminate description of matter or corporeal substance which the modern philosophers are run into by their own principles, resembles that antiquated” notion. The context makes it clear that the notion of a “corporeal substance” depends, as Berkeley puts it, “on that strange doctrine of abstract ideas,” specifically, on Locke's claim that “the principium individuationis … is Existence itself, which determines a Being of any sort to a particular time and place incommunicable to two Beings of the same kind.”⁶ While Berkeley is almost certainly aware of the fact that Locke advances this notion in opposition to Aristotle's materia prima, he nevertheless objects to it. Berkeley does so because, while he sees that Locke eliminates substance as individuator, Locke nevertheless retains substance as substratum or support: at PR 17 Berkeley says (again, with emphasis added) that this “idea of being in general, together with the relative notion of supporting accidents … appeareth to me the most abstract and incomprehensible of all other”; and he then argues, in the same section, that Locke's notion of “support” is incoherent because it leads to an infinite regress.

The significance of ‘resembles’ in PR 11, then, is that it suggests Berkeley recognizes that substances play two roles in his adversaries' philosophies: the notion of a “corporeal substance” is not exhausted by its role as individuator but has the additional role of substratum. Indeed, the significance of ‘resembles’ goes beyond even this, for Berkeley recognizes that such substances have a threefold role: in addition to the two just mentioned, his adversaries introduce substances as causal agents, a notion that Berkeley dismisses, at least with regard to “corporeal substances,” with an independent series of arguments, in PR 18-19 and 25. Not the least significant part of the puzzle addressed in this essay is the contrast between (a) the fact that Berkeley often suggests that minds support their ideas and says that minds are agents and (b) the fact that mental substances play neither of these roles in his philosophy.

It is obvious that mental substances cannot be individuators. In Berkeley's nominalistic ontology, only particular items exist.⁷ Consequently, Berkeley can safely dismiss as ridiculous any attempt to identify a mental substance with a mens prima, just as he does at PR 11 with materia prima. (And, if pressed, he could simply appeal to Ockham's razor.)⁸ Of course, though Berkeley has no need of mental substances as atemporal individuators, this does not show that he has no need for them as temporal continuants. Berkeley may introduce mental substances, in other words, in order to account for personal identity. There are good reasons to think, however, that this requirement is viewed by him as a largely theological (rather than as an ontological) requirement, and I return to it at the end.

It is not as obvious that mental substances cannot be causal agents. For one thing, Berkeley consistently describes them as such. The following passage, for example, is not atypical. Berkeley has Philonous say that “I know or am conscious of my own being; and that I my self am not my ideas, but somewhat else, a thinking active principle that perceives, knows, wills, and operates about ideas” (D 233). But while Berkeley in the Principles and Philonous in the Dialogues frequently make such assertions as these, little effort is devoted to explaining how it is possible for us to have this sort of knowledge, and next to nothing is said about the ontological role this “one individual principle” is supposed to play. The continuation of the above passage is interesting, however, since Philonous advances an argument that one's self must be “one individual principle,” although, note bene, his argument does not address the question of this item's causal agency: “I know that I, one and the same self, perceive both colours and sounds; that a colour cannot perceive a sound, nor a sound a colour; that I am therefore one individual principle, distinct from colour and sound; and, for the same reason, from all other sensible things and inert ideas” (D 233-34). The argument seems to be that since sensible qualities (of one “sensory modality”) cannot perceive sensible qualities (of another), there must be some item, not itself a sensible quality (or, presumably, a passion or volition), that does the perceiving, and this can only be a mental substance. Philonous's argument, however, is one for which the notebooks' Berkeley would (and the “mature” Berkeley should) have had nothing but scorn. In the case of immediate perception, the occurrence of a sensible quality (an idea of sense) simply is a perceiving; or, to put it differently, an immediate perceiving is not distinct from its object; it is its object. Furthermore, since ideas of imagination are sometimes representative of ideas of sense, the former are, broadly speaking, perceptions of the latter. Moreover, mediate perception requires the possibility that, again in an important sense, a sound does sometimes “perceive a color”: involved in the perception of a sensible object (including that object's color), and in some sense “mediating” it, is the perception of an auditory idea.⁹ Finally, even conceding that no idea can perceive another idea, there is no difficulty in accounting for the sense in which I perceive a color or sound without supposing that ‘I’ refers to “one individual principle,”¹⁰ a mental substance: one can hold, as the notebooks' Berkeley does, that ‘I’ refers to a congeries of ideas (and volitions) and that ‘I see this color’ constitutes a (partial) “explication” of the meaning of ‘I’.¹¹ Indeed, Berkeley suggests as much in PR 49 when he advances his antisubstantialist account of the meaning of ‘a die is hard, extended and square’.

Shortly before advancing the above argument Philonous says, “My own mind and my own ideas I have an immediate knowledge of”; and shortly after this, he tells Hylas “you neither perceive matter objectively, as you do an inactive being or idea, nor know it, as you do your self by a reflex act” (D 232). From these two assertions it is safe to conclude that knowledge of the self and its volitions is not inferential and that nothing intervenes—there is no tertium quid—between the “reflex act” and the mind and its operations that act makes known. But what is this reflex act and what sort of entity is its target? Berkeley provides no answer.

The end of the later passage (cited above from D 233-34) is designed to reinforce Berkeley's claim at PR 2 that a mind is entirely distinct from its ideas. And it is noteworthy that he does not say here that a mind is distinct from its volitions. Indeed, when he explains, at PR 27, why we cannot have ideas of our minds, the explanation rests entirely on the fact that our episodes of volition are active: ideas, being passive, cannot represent them. Now, Berkeley holds that causation is activity, but it is particular episodes of volition, not mental substances, that play this causal role. If a mind is constituted, in part, of episodes of volition, then the mind can safely be described as an “agent” or “active principle”; but it is nonetheless clear that causal agency resides, fundamentally, in individual episodes of volition. Moreover, since only such episodes are causes, a mental substance could not, in any case, cause one of its own episodes of volition without first willing it.

Finally, Berkeley also dismisses the support role of substance, explicitly with regard to material substance and also with regard to mental substance. Seeing this requires an examination of several additional passages, and in the course of this examination, while working up to an explanation of how minds are supposed to support ideas, we discover additional evidence both that the notebooks' analysis hovers in the background and that Berkeley takes some pains to conceal this.

The two most significant sections for my purposes are PR 49 and PR 91. The former, like PR 27, is about minds, but it is much more explicit in its import. Berkeley insists at PR 49 that sensible qualities are not in the mind “by way of mode or attribute”; rather, “they are in the mind only as they are perceived by it, that is, … only by way of idea” (the italics on ‘only’ are added). This is Berkeley's answer to an objection he anticipates, an objection premised on the very substance account of mind he seems to have endorsed earlier in the Principles. The objection is that on the Scholastic-Cartesian account,¹² a mind is a subject of prediction, and thus Berkeley's introduction of a substance into his analysis of mind entails the absurdity that sensible qualities are predicable of minds. Berkeley puts that objection this way: “[I]f extension and figure exist only in the mind, it follows that the mind is extended and figured; since extension is a mode or attribute, which (to speak with the Schools) is predicated of the subject in which it exists” (PR 49) Again, if Berkeley has introduced a Scholastic-Cartesian mental substance into his analysis of finite minds, it is hard to see how PR 49 can be squared with such an introduction.¹³

In this connection, it is also worth noting that there are many passages in which a substance analysis of mind seems to have been, not reintroduced—for until the end of the notebooks, Berkeley seems never to have taken this as a serious philosophical possibility—but rather, injected. The very first mention of mental substance at PR 7, for example, begins with the sentence “From what has been said, it follows, there is not any other substance than spirit, or that which perceives.” But the continuation in this section alludes to the argument of PR 49 when Berkeley says that “for an idea to exist in an unperceiving thing, is a manifest contradiction; for to have an idea is all one as to perceive: that therefore wherein colour, figure, and the like qualities exist, must perceive them; hence it is clear there can be no unthinking substance or substratum of those ideas” (PR 7). The allusion to the argument of PR 49 resides in the fact that Berkeley does not conclude here that ideas must be in a mental substratum. He seems deliberately to conclude only that they cannot be in a material substratum¹⁴ and that their mind-dependence rests rather on the fact that “to have an idea is all one as to perceive.” Indeed, this last statement itself alludes to PR 49, where Berkeley says sensible qualities “are in mind … only by way of idea”; and his emphasis on ‘substratum’ in PR 7 is amplified at PR 49 when he asserts, not with respect to bodies but, rather, with respect to minds: “As to what philosophers say of subject and mode, that seems very groundless and unintelligible.” As Berkeley well knows and as the remainder of PR 49 makes clear, ‘subject’ refers to the linguistic correlative of the ontic ‘substance’, and he is thus saying that talk of mental substances and their modes is “groundless and unintelligible.”

We can turn now to the second important section, PR 91. In this section, Berkeley gives what appears, at first glance, to be an argument for the mind-dependence of sensible objects based on the premise that sensible qualities must inhere in mental substances. That is, he seems to argue in this way: “Qualities need a support, a substance in which to exist. But the only substances available are minds. Hence, qualities must be supported by minds, they must be in minds.”¹⁵ Here, from PR 91, is the passage in question:

It were a mistake to think, that what is here said derogates in the least from the reality of things. It is acknowledged on the received principles, that extension, motion, and in a word all sensible qualities, have need of a support, as not being able to subsist by themselves. But the objects perceived by sense, are allowed to be nothing but combinations of those qualities, and consequently cannot subsist by themselves. Thus far it is agreed on all hands. So that in denying the things perceived by sense, an existence independent of a substance, or support wherein they may exist, we detract nothing from the received opinion of their reality, and are guilty of no innovation in that respect. All the difference is, that according to us the unthinking beings perceived by sense, have no existence distinct from being perceived, and cannot therefore exist in any other substance, than those unextended, indivisible substances, or spirits, which act, and think, and perceive them.

Beyond that first glance, a careful reading of this passage reveals no argument, implicit or explicit, based on a Scholastic-Cartesian inherence pattern. The passage suggests in fact that Berkeley is making a note here only of his agreement, in respect of “the reality of sensible things,” with the “received principles.” He is saying that, in his philosophy, the dependence of sensible things on minds does not strip sensible things of their reality; and he suggests that it does not do so, because “it is agreed on all hands,” first, that sensible qualities “have need of a support, as not being able to subsist by themselves.” Let us call this, the requirement demanded by the dependent nature of sensible qualities, the support condition. The first point of agreement, then, between Berkeley and his adversaries, is the support condition.

On the “received principles,” the support condition is satisfied within a substance-mode account of bodies. On this account, sensible qualities are modes; they are dependent items that, if they are to exist, must inhere in existentially independent items or substances. On the “received principles,” as Berkeley well knows, it is in that sense that “sensible qualities … have need of a support.” But in what sense of ‘support’ does Berkeley hold that sensible qualities have such a need? In other words, how does Berkeley, within his own philosophical system, satisfy the support condition?

There is no direct answer to that question in PR 91, but Berkeley does answer it directly in PR 49. Before turning to that, however, it is worthwhile to pause briefly and examine the second point of agreement between his and the “received principles.” This point is that “the objects perceived by sense, are allowed to be nothing but combinations of those qualities, and consequently cannot subsist by themselves.” But in point of fact not every materialist would “allow” this second point. It is certain that no Aristotelian-Scholastic philosopher would agree that a sensible object is nothing but a collection of sensible qualities. Moreover, as Berkeley well knows, such philosophers would certainly also disagree with his implication here that the sensible qualities we normally attribute to a body must be in a substance simpliciter: they hold that these sensible qualities must be in a material substance. Berkeley's intentions in this passage are thus unclear: one cannot help feeling that irony is at play in PR 91 (not an uncharacteristic Berkeleian strategy) or that Berkeley is deliberately obscuring his own position. In view of what we have seen to this point, the latter alternative is at least as likely as the former.

If we turn from the second point of agreement and examine the first more carefully, however, that second alternative becomes a virtual certainty: in PR 91 Berkeley is deliberately concealing the congeries analysis of mind. At PR 49 he declares as “very groundless and unintelligible” the materialists' talk of “subject and mode”; and the sentences preceding this declaration make it clear that he is talking about minds. The example he uses to illustrate this, however, is not a mind but a body: “For instance, in this proposition, a die is hard, extended and square, they will have it that the word die denotes a subject or substance, distinct from the hardness, extension and figure, which are predicted of it, and in which they exist. This I cannot comprehend.” Berkeley then presents us with his own analysis of a die: “[T]o me a die seems to be nothing distinct from those things which are termed its modes or accidents. And to say a die is hard, extended and square, is not to attribute those qualities to a subject distinct from and supporting them, but only an explication of the meaning of the word die.” These passages and numerous others (PR 1, to cite just one example) make it clear that, for Berkeley, a sensible object is just a collection of sensible qualities: an inventory of the sensible qualities of bodies exhausts their analyses. Berkeley holds that the support condition is satisfied within his account by requiring that sensible qualities always be members of such combinations as constitute individual bodies.¹⁶ But this means, of course, that the sense in which sensible qualities “have need of a support” is quite different from the sense of ‘support’ countenanced by the materialists—quite different from the sense of ‘support’ that Berkeley finds unintelligible and groundless. Thus, again, in PR 91 we find Berkeley craftily disguising his real position.

His real position is that sensible qualities—and those collections apart from which they are inseparable—need to be “supported” by minds because they, in addition to being inseparable from collections or combinations, are also inseparable from sensations.¹⁷ This position is suggested by what Berkeley says at PR 5; it is greatly amplified in the first of his Dialogues (D 176-78 and 191-92); and although he does not say as much in PR 91, what he does say presupposes either that sensible qualities are sensations or, at least, that sensible qualities depend for their existence on being perceived.¹⁸

At the end of the long passage from PR 91 cited above, Berkeley calls attention to a difference between the position that he advocates and the position of those accepting the “received principles”; and this difference is crucial: sensible things have no existence apart from being perceived and therefore cannot exist apart from spirits. In other words, Berkeley explicitly argues that because the esse of sensible things is percipi, sensible things cannot exist apart from minds. The argument of PR 91, then, is not an argument for the mind-dependence of sensible things based on the inherence pattern. It is not, to say the same thing differently, an argument for idealism based on the support condition that the materialists would find agreeable. Rather, if there is an argument here at all, it is an argument that presupposes idealism, an argument that assumes idealism has been antecedently established.

Now, I do not mean to suggest here that any of this proves that Berkeley has not introduced a Scholastic-Cartesian mental substance into his analysis of mind by the time he publishes the Principles; indeed, it seems clear that he has. Although this introduction generates difficulties, is embarrassing, and is inconsistent with several key sections, and although Berkeley has gutted it of much of its ontological significance—for he certainly believes he can establish idealism as well as his theory of causation without it—Berkeley nevertheless has not only introduced mental substances and emphasized their distinctness from ideas, but he has also taken great pains to conceal his underlying congeries analysis of mind. Why?

As I mentioned at the outset, this is really two questions: Why does Berkeley introduce mental substances? And why does he conceal the congeries analysis? Part of the answer to the first question is, I think, that Berkeley needs a way to distinguish minds from sensible objects if he is to vouchsafe his commitment to common sense, in particular if he is to secure the bare possibility¹⁹ that the plain person's bodies can exist independently of the plain person's mind. Moreover, the required method must, at the same time, not compromise his idealism. This is a very tall order. Berkeley certainly thought it could be filled; and, indeed, he came very close to filling it.

Having nearly completed his notebooks, Berkeley reasoned, I submit, in the following way. On the one hand, if my mind is nothing but a congeries of ideas and volitions, and if some of these ideas are bodies—that is, “houses, mountains, and rivers” (PR 4) or “the furniture of the world” (PR 6)—how can any sense be given to the notion that such bodies might exist apart from my mind? To be sure, such bodies can exist apart from my volitions, since I am not the causes of them. Moreover, since causes are distinct from their effects (PC 780), even if my volitions did produce bodies, it would be logically possible for bodies to exist apart from me. But if a mind is constituted, even in part, by its ideas, it would seem that my ideas cannot exist apart from my mind.²⁰ On the other hand, if my mind can be described as “entirely distinct” from its ideas,²¹ then it is at least logically possible that some of my ideas, those which constitute bodies, might exist apart from my self.²² Thus the attraction of a substance analysis of mind.

We should note that if this interpretation is correct, Berkeley has introduced mental substances for a purpose that has nothing directly to do with the analysis of mind. He introduces them, rather, to shore up the claim that his philosophy is not in opposition to the plain person's “vulgar” realism.²³ The upshot is astonishing: the revolutionary departure from his adversaries' general notion of substance is that Berkeley introduces mental substances to secure the reality of material things!²⁴ Of course, from the perspective of “the modern philosophers,” Berkeley's maneuver here is, far from being revolutionary, completely absurd. But Berkeley is nothing if not bold.

Yet although the problem of securing commonsense realism is almost certainly on his mind, Berkeley has another reason for introducing mental substrata—one that, in his view at least, is even more important than saving realism: the substance analysis of mind is required to placate the theologian-philosophers. This second answer to my first question also provides an answer to the second question: the congeries analysis must be concealed because it cannot be easily squared with theological dogma. The evidence for this is, in the nature of the case, rather thin. Nevertheless, I think it is convincing.

Return, once again, to Berkeley's notebooks. There is a “memo file” in these notebooks, and toward the end one finds the following entries:

I must not Mention the Understanding as a faculty or part of the Mind, I must include Understanding & Will etc in the word Spirit by w^ch I mean all that is active. I must not say that the Understand differs not from the particular ideas, or the Will from particular Volitions.

(PC 848)

The Spirit, the Mind, is neither a Volition nor an Idea.

(PC 849)

I must not give the Soul or Mind the Scholastique Name pure act, but rather pure Spirit or active Being.

(PC 870)

In these late entries Berkeley has reached the decision that will shape his description of mind in the Principles. Despite all of the entries arguing against the existence of mental substances, Berkeley now reminds himself that he “must” downplay the congeries analysis (PC 848) and incorporate a substance, “the Mind” (PC 849) or an “active Being” (PC 870), into his ontology.

There is no entry in the immediate context that explains why Berkeley thinks he must do this. From a series of entries occurring just a bit earlier, however, the right explanation nearly leaps off the page: Berkeley is here reminding himself to avoid theological difficulties when he publishes the Principles, difficulties that are sure to follow from any explicit endorsement of the congeries analysis. At the beginning of this series (PC 713-15), Berkeley has written the following memo, similar in content to the ones above: “The Concrete of the Will & understanding I must call Mind not person, lest offence be given, there being but one volition acknowledged to be God. Mem: Carefully to omit Defining of Person, or making much mention of it” (PC 713). In the next entry (PC 714) he gives voice to his rejection of mental substrata: “You ask do these volitions make one Will. w^t you ask is meerly about a Word. Unite being no more.” And this is followed immediately (PC 715) by an urgent memo: “N. B. To use utmost Caution not to give the last Handle of offence to the Church or Church-men.” Berkeley describes himself as having been “distrustful at 8 years old and Consequently by nature disposed for these new Doctrines” (PC 266). Having so boldly insisted at PC 714 that we should eschew the temptation to “unite being”—that we should eschew the temptation to postulate a single item to which individual episodes of volition are connected and from which they emanate—a Berkeley with that eight-year-old's temperament might well have ignored this urgent memo. It is no “juvenile” Berkeley who writes the notebooks, however, but the mature man of twenty-two that boy became. And this man, the notebooks' Berkeley, is fully prepared to heed those less bold but more discretionary memoranda. Indeed, since he is himself grooming to become one of these very “Church-men,” the memo at PC 715 is not only prudent, it is understandably urgent.

It seems clear, then, that Berkeley's introduction of mental substances is motivated primarily by the desire to account for personal identity in a way consistent with religious dogma. If a mind is simply a collection of mental episodes (of episodes of perception, passion, and volition), it is hard to see how this can account for the theological concept of a soul—a concept that seems to require an ontological ground in a continuously existing item. Berkeley says about ideas that they are “fleeting indeed, and changeable” (D 258), but if the same be said of the contents of a substanceless mind, then—notwithstanding that he can give an account of the continuance or identity of minds similar to the one he advances for bodies (at, e.g., D 247-48), that is, a relational account in which there is no item that endures or is “strictly” identical through time—he will indeed have given the “Handle of offence to the … Church-men”

I conclude that Berkeley makes no use of the substance analysis of mind in securing any of his central metaphysical doctrines, that the principal use to which he puts it resides in vouchsafing commonsense realism, and that the even more pressing need for it derives from extraphilosophical prudence. Berkeley speaks freely of minds as substances, despite the inconsistency between this analysis and the preferred congeries account of mind—an inconsistency so cleverly concealed—in order to avoid exposing an offensive opening to the theologians. What is most remarkable in all of this is Berkeley's mastery of the arts of deflection and camouflage. Indeed, he succeeded so exceedingly well that generations of theologians, as designed, but also generations of commentators on his philosophy, have been thoroughly misled.

Notes

1. The central entries are PC 577-81 (it is here that Berkeley describes a mental bundle as a “congeries”), but there are supporting entries scattered throughout the notebooks. In the recent secondary literature, see I. C. Tipton's ground-breaking paper, “Berkeley's View of Spirit,” in New Studies in Berkeley's Philosophy, ed. Warren E. Steinkraus, (New York: Holt, Rinehart & Winston, 1966), pp. 59-71. See also A. C. Lloyd, “The Self in Berkeley's Philosophy,” in Essays on Berkeley, ed. John Foster and Howard Robinson (Oxford: Oxford University Press, 1985), pp. 187-210. In neither of these articles does the author devote sufficient attention to the development of Berkeley's account of mind within the notebooks, nor do the authors place such developments as are discernible within the broader context of his two major works. If these developments are exposed and placed in that broader context, it becomes clear that Berkeley's nominalism (and, consequently, his antiabstractionism) is fatal to substances, whether material or mental. It becomes clear, too, that the notebooks' Berkeley not only fully appreciates the significance of this, he also repeatedly and enthusiastically exploits it. Satisfying the antecedent of the preceding conditional is, of course, well beyond the scope of the present essay. I make an effort along these lines in Berkeley's Ontology.

2. Berkeley's effort to deflect this objection is dissected by Phillip D. Cummins in “Hylas' Parity Argument,” in Berkeley: Critical and Interpretative Essays, ed. Colin M. Turbayne (Minneapolis: University of Minnesota Press, 1982), pp. 283-94.

3. An excellent discussion of Berkeley's theory of causation can be found in Kenneth P. Winkler's Berkeley: An Interpretation (New York: Oxford University Press, 1989), pp. 104-36.

4. Why does Berkeley hold that the existence of any abstract idea is an “impossibility” (IN 21)? In Berkeley's Ontology, pp. 24-76, I argue that the answer to this question lies in his implicit nominalism.

5. In “Substance, Reality, and the Great, Dead Philosophers,” American Philosophical Quarterly 7 (1970), pp. 38-49, Michael R. Ayers insists that while PR 27 appears to be an objection to all talk of substance and support, it cannot be an objection to substance “per se,” since “Berkeley has just categorically stated that spirit is an incorporeal, active substance, applying to it a traditional definition” of “one simple, undivided, active being” (47). I agree. See note 18.

6. John Locke, An Essay Concerning Human Understanding, ed. Peter H. Nidditch, (New York: Oxford University Press, 1975), book II, chap. xxvii, sec. 3.

7. In the first of his Dialogues, in a context in which Philonous is discussing qualities—both determinate: “some degree of swiftness or slowness, some certain magnitude or figure peculiar to each [sensible thing]”; and determinable: “extension in general, and motion in general”—Berkeley has Philonous emphatically declare that “every thing which exists, is particular” (D 192). Winkler, Berkeley: An Interpretation, p. 30, is thus mistaken when he says that “by ‘things’ in this context [Berkeley] means all entities capable of independent existence” (emphasis added). For more on this, see Berkeley's Ontology, pp. 33-36.

8. Berkeley does appeal to Ockham's razor as part of his attack on the supposition that there are substantial material causes. At the end of PR 19 he says that it “must needs be a very precarious opinion; since it is to suppose, without any reason at all, that God has created innumerable beings that are entirely useless, and serve to no manner of purpose.” Berkeley does not appeal to Ockham's razor to dispose of Aristotle's materia prima—unless we read his comment at the end of PR 19 as a blanket complaint—because, as the context (PR 11) suggests, he finds this notion beneath contempt.

9. Cf. the coach example at the end (D 204) of the first dialogue.

10. In “Hylas' Parity Argument,” p. 291, Cummins holds that Berkeley's attempt here to derail the parity argument can be defended against its principal criticism, namely, that Berkeley “argues from what is experienced (the empirical self and its ideas) to an unexperienced entity (the individual active principle or substance) on the basis of a premise (one idea cannot perceive another) whose truth cannot be established from the data of experience.” But while Cummins is right to point out it “is a profound mistake to represent Berkeley as permitting no metaphysical premises,” this does not alter the fact that Berkeley never explains why this particular unexperienced entity, introduced by a different metaphysical “premise,” must exist—why the “furniture of the world” must include individual active mental substances in addition to “their” volitions and perceptions. While the empirical self, invoked in that passage from the third dialogue, can be described as “one and the same self,” what is the point of describing the unexperienced substance in this way? The answer is surprising. But wait.

11. In the body of his Treatise Hume found no difficulty in advancing a substance-free account of mind. See A Treatise of Human Nature, ed. L. A. Selby-Bigge and Peter H. Nidditch (New York: Oxford University Press, 1978), pp. 251-63. For a response to many of the standard objections to this account, see Nelson Pike, “Hume's Bundle Theory of the Self: A Limited Defense,” American Philosophical Quarterly 4 (1967), pp. 159-65. To be sure, Hume himself found something troubling in his account (see Treatise, appendix, pp. 635-36), but no satisfactory explanation has been found for his lament that “my account is very defective.” For a critical review of some attempts to do so, as well as an intriguing new attempt, see Wayne Waxman, “Hume's Quandary Concerning Personal Identity,” Hume Studies 18 (1992), pp. 233-53.

12. For the Aristotelian-Scholastics, a mind is part of the substantial form, the substantia secondae, of a body, but for Descartes, a mind is a substance distinct from a body. When I speak, then, of the Scholastic-Cartesian account of mind, my intention is to stress that the account of mind Berkeley attacks is a Cartesian hybrid growing out of a crack in the Aristotelian-Scholastic bedrock.

13. In “Berkeley's Idealism,” Theoria 29 (1963), pp. 229-44, Edwin B. Allaire gives an ingenious reading of PR 49 in which he tries to do just this. However, in “Berkeley's Idealism Revisited” (in Berkeley: Critical Essays, ed. Turbayne, 197-206, esp. 198-200), he retracts this reading of PR 49 while nevertheless defending the inherence interpretation as a valuable analytical tool. See also Allaire's contribution to this volume, Essay 1, as well as Fred Wilson's, Essay 4.

14. Cf. Phillip D. Cummins, “Berkeley's Manifest Qualities Thesis,” Essay 6 in this volume (esp. 123). In “The Role of Perceptual Relativity in Berkeley's Philosophy,” Journal of the History of Philosophy 29 (1991), pp. 397-425, I argue in detail that Berkeley makes a similar move in the first of his Dialogues: he holds, as he does three years earlier in the Principles, that the argument from perceptual relativity shows only that sensible qualities cannot be in material substances. To put it differently, I argue—pace the standard interpretation—that Berkeley does not use the fact of perceptual relativity to establish the mind-dependence of sensible qualities. Cf. also Alan Hausman and David Hausman, Essay 4 in this volume (esp. 54).

15. Allaire, “Berkeley's Idealism,” p. 235.

16. More accurately, Berkeley's position is that sensible qualities cannot exist apart from the proper objects of immediate perception.

17. Historically, this interpretation is first advanced by Thomas Reid. See Phillip D. Cummins, “Berkeley's Ideas of Sense,” Nous 9 (1975), pp. 55-72.

18. In “Substance, Reality, and the Great, Dead Philosophers,” Michael R. Ayers points out that Berkeley “is liable to be misread as if he objects to the concept of substance per se, although his language makes it plain that he is not in fact doing so” (p. 49). It is, indeed, plain that Berkeley uses ‘substance’ in talking about minds, and plain too that he describes mental substances in the traditional ways (PR 27 and 135). But let me emphasize once again that my argument is not directed at demonstrating a Berkeleian rejection of “the concept of substance per se”; rather, I am raising the question about what role, beyond this merely nominal one, the concept of mental substance plays in Berkeley's philosophy. My conclusion, that it plays none of the traditional roles, is not a possibility Ayers considers, even though it is consistent with his observations that in the Aristotelian tradition “ontological and causal priority go together” and that the causal role played by Berkeley's mental substance is not “the usual explanation of activity by essences” (p. 48). While Ayers tells us what the causal role of mental substance is not, he does not explain what it is; and as for its ontological role (the role of “immaterial support”), while he alleges that it “can be made clear and intelligible by identifying” supporting with perceiving (p. 47), Ayers in fact provides no explanation how the Berkeleian conversion of ‘support’ (or ‘inherence’) into ‘perceiving’ is to be understood. In Berkeley's Ontology I have argued, in effect, that the right explanation resides in Berkeley's construing an object of immediate perception as a sensation (his argument for this is the first major argument presented in the Dialogues: the heat-pain argument) and then relies on the assumption, shared by all his contemporaries and predecessors, that sensations are mental, that is, that sensations cannot exist apart from sensory awareness.

19. I stress this because the distinction between a mental substance and an idea of sense (an object) can provide only a necessary condition of the truth of realism. (That Berkeley is fully aware of this is clear from his response, at PR 48, to the intermittency objection of PR 45: the response is simply that the “foregoing principles” do not entail intermittency.) To provide a sufficient condition, Berkeley needs an additional argument, an argument he thinks he finds in his proof of God's existence.

20. This appearance is deceptive, as Hume recognized. See Treatise, p. 207.

21. Berkeley never says that a mind is distinct from its volitions.

22. While Berkeley several times asserts that minds are entirely distinct from their ideas, he never asserts that ideas are entirely distinct from minds. How could he? The latter is inconsistent with his idealism. This fact is connected to a controversial point of interpretation: are Berkeley's ideas acts of perceptual awareness (or features of those acts), or are they objects of awareness? Winkler examines this controversy with some care in Berkeley: An Interpretation, pp. 291-300; I examine Winkler's argument in Berkeley's Ontology, pp. 229-33; and the issue comes up again in Charles J. McCracken's quarrel with Margaret Atherton (Essays 13 and 14 in this volume).

23. According to the inherence account, advanced by Edwin B. Allaire more than two decades ago in “Berkeley's Idealism,” Berkeley requires a mental substance in order to vouchsafe idealism. For more on this, see the editor's Introduction and Essays 1 and 4 in this volume.

24. It might be urged that in Berkeley's ontology only God is a mental substance in anything like the traditional sense of ‘substance’. I agree. Since God is required to secure the reality of bodies, however, the result is the same: mental substances are introduced to secure the reality of material things.

Bibliographical Note

Cross-references to this anthology are by parenthesized page numbers: (#) or (#-#). Except for those to Berkeley's writings, all other (external) page references are preceded by p. or pp.

For Berkeley's writings, we have used A. A. Luce and T. E. Jessop, editors, The Works of George Berkeley, Bishop of Cloyne, 9 volumes (London: Nelson, 1948-57); hereafter referred to as Works. More specifically, references to Berkeley's writings in this volume include the following abbreviations:

A Alciphron (Works III). References are by dialogue, section, and page number.

D Three Dialogues Between Hylas and Philonous (Works II); hereafter, Dialogues. References are by page number.

DM De Motu (Works IV). References are by section number.

IN Introduction to the Principles (Works II). References are by section number.

NTV An Essay Towards a New Theory of Vision (Works I). References are by section number.

PC Philosophical Commentaries (Works I). References are by entry number.

PR A Treatise Concerning the Principles of Human Knowledge (Works II); hereafter, Principles. References are by section number.

S Siris (Works V). References are by section number.

TVV The Theory of Vision Vindicated and Explained (Works I). References are by section number.