Hume and Bayle on Localization and Perception: A New Source for Hume’s Treatise 1.4.5

Abstract

Next to the five major areas listed by Norman Kemp Smith in which Bayle had exercised greater influence on Hume’s work, we shall add here a new Baylean source. It is not only the second part of section V (“Of the immateriality of the soul”) of Treatise I, IV that depends on Bayle (for the article Spinoza), but also the first part. The topic at issue is the difficulty of finding a “relation” between “perceptions, which are simple, and exist no where”, on the one hand, and on the other hand some “conjunction in place with matter or body, which is extended and divisible”. A primary source (not listed in Norton & Norton commentary) for this topic is an extensive chapter in Bayle’s Réponse aux questions d’un provincial (III, xv), where the problem of a possible (and impossible) ‘local conjunction’ of the spirit and the body is dealt with in considerable depth.

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Pierre Bayle’s influence on David Hume and the writing of his A Treatise of Human Nature must now be expanded to another field, beyond the five major areas identified by Norman Kemp Smith in the 1940s.¹ The first part of Hume’s “Of the immateriality of the soul”² is influenced, of course, by the “Spinoza” article in Bayle’s Dictionnaire.³ But so too is the second part indebted to Bayle, where Hume presents the difficulty of determining how our “simple” perceptions, which “exist nowhere”, could possibly have any “conjunction in place with matter or body, which is extended and divisible”. We shall now show that a primary source used by Hume for that presentation is a chapter from Bayle’s Réponse aux questions d’un provincial ⁴ which deals with the same problem, and at considerable depth.⁵

1 Localization and Immateriality of the Soul in Bayle’s Réponse Aux Questions D’un Provincial

Of course, the first part of the section is Hume’s detailed refutation of certain orthodox theses concerning the immateriality of the soul. He takes as his starting-point the following apparently closely-defined question: What is the nature of that “local conjunction” presumed to exist between spirit and body, which themselves are of natures so different as to be in many ways opposed?⁶ Remarkably, the nature of immaterial, spiritual substance was a problem that Bayle had taken up in the third part of the Réponse aux questions d’un provincial, analyzing it from exactly the same angle, the issue of its localization in a body. Now, Hume was clearly quite familiar with that work, for he quotes from it several times (although on questions of a different type) in the “Early Memoranda”, most of which he wrote around the same time as he wrote the Treatise.⁷

Let us begin by looking at some of the more important passages in Bayle’s reasoning in the Réponse, and then turn to Hume and his comparable treatments. Bayle actually begins further back, with the immaterial substance par excellence, divine substance. It is not his intention to cast doubt on the existence of this substance, and indeed his next move is to examine the problematic consequences of any dogma that would deny its existence: “we should verify that the consequences of this dogma do not come into conflict with other truths that it is important to maintain.” These consequences include the localization of spiritual substances in general, and of the human soul in particular, and thus the theme of the “local union” of the soul with the body. Typically, Bayle approaches the problem gradually.

First, there is the dilemma surrounding the animal soul, a dilemma that also arises, although in slightly different terms, for nobler substances, like God or the human soul. Either we accept that the soul “does not exist in the body of the animal” – which conflicts with the common conviction that animals are animate beings – or else we say that it does lie in the body, and therefore is “material” too.⁸ The old scholastic distinction of two types of extension, “one indivisible and penetrable, the other divisible and impenetrable” (taken up again later by Cudworth and Le Clerc) proves repugnant to Cartesian clarity, and thus Bayle rejects it. His next step is to extend the problem of “localization” to the case of the human soul, and here, too, his dialectic is stringent. If we say that man’s soul is “diffused throughout his body” and that it “penetrates” it, occupying “the same space as our body”, he argues, we wind up accepting the existence of “two substances that fill the same place, one in an indivisible manner, the other in a divisible manner”. This too can only entail absurdity for a Cartesian like Bayle, for whom material extension and spatiality are identical. And were it replied here that the soul is not extended, then its having any location at all becomes inconceivable, and we are forced to say that souls and bodies have no connection through any shared location.⁹

So one prong of this two-pronged dilemma has us materializing the soul, transforming it into extension, while the other prong has us rendering it impossible that the soul has any location-connection with the body, and thus effectively de-localizing it altogether. Showing proof of great intellectual honesty, Bayle concludes by noting that principles which are evident and shareable in themselves, such as the immateriality of spiritual substance, are nevertheless accompanied by severe and unacceptable “consequences” which keep us from enjoying the evidence of these principles in “tranquility” (“quiétude”).

Nor does the starting-point of the discussion, God, escape this disquieting shadow. For God’s traditional attribute of “immensity”, whereby he is present everywhere “in the infinite spaces”, ends up being incompatible with the immateriality of the divine principle,¹⁰ since the idea that a spiritual substance “[neither] composed of parts nor extensive” could occupy a “three-dimensional location” such as space seems totally incomprehensible. Yet nor does the rejection of the traditional conception of God avoid aporia. Once “the Cartesians” have laid their foundations, Bayle writes, they should accept all the ensuing consequences, including that “God, being spirit, does not exist in any place”, and that “the created spirits are nowhere”. And they should also reject “the greatest of all the chimeras”, the idea “that our soul is locally united with our body, or that it exists in our body”.¹¹ Such consequences also clash with common sense: while the Cartesian line may be superior in clarity, Bayle emphasizes that, nevertheless, it “has no conformity with our way of thinking: it troubles our minds: what grip can our conceptions get upon a substance that cannot be located in any place?” This is why the traditional dogma of divine immensity, and the parallel dogma of the soul’s “local union” with the body, still “hold sway” over our minds, despite the objections of the philosophers.

Incidentally, Bayle’s doubts go well beyond the problem of localization. He raises a doubt which had appeared to lie outside the discussion: the very immateriality of the soul itself. Not only is it clear, he writes, that “the proofs that human reason can give for the immateriality of the human soul are far from convincing”, but, furthermore, the impossibility of knowing “that which constitutes the substance of a spirit and the substance of a body” opens the door to an unsettling hypothesis posited by Locke: that thought is a property of matter. As is well known, Bayle returns to this hypothesis in his article “Dicéarque”, in which, amid a plethora of details, he frees it of the claims represented by the use of the theme of divine omnipotence, which had limited its force in Locke’s Essay. And in that article he also makes reference to “local union”, that starting-point of his Réponse discussion, which only goes to show the centrality of this theme for Bayle, and the danger that he saw it posing for orthodoxy. And we shall see that Hume converges strongly and clearly on this same point, for “local union” would likewise be a starting-point for the Scot as he advances his thesis concerning the material causality of thought.

But let us remain with Bayle’s argument in the Réponse. When arrives at its conclusion, his reasoning makes a sharp revirement. Locke’s argument is suddenly rejected, and by a typically Cartesian argument modelled on the purest of clarity: if we admit that matter thinks, then we are led to admit as well that there is “a single species of substance, which by one of its attributes is joined to thought, and by the other to extension”.¹² And the consequence of this admission, Bayle points out, would be a plunge back into “the ancient chaos of the scholastics”. Substance thus conceived would become an unknown support, distinct from its attributes and its qualities, and thus no longer defined by its “principal” attributes, as the Cartesian conception defines it. It would frankly be something closer to Spinoza’s conception of substance – at least according to the highly polemical and biased interpretation given in the “Spinoza” article of the Dictionnaire, and according to the interpretation adopted by Hume in the latter part of Section V.¹³ Thus, Bayle’s conclusion shows a bit of cautious resignation, plus a sprinkling of scepticism: in spite of all the difficulties, it seems better to hold to the clear Cartesian distinctions, since the alternative conceptions, be they scholastic, Lockean, or Spinozan, would only “throw us into even worse darkness”.¹⁴

2 Localization of Perception and Thought in Hume’s Treatise: The Implicit Reply to Bayle’s Aporias

It is not at all improbable that this passing allusion to Spinoza’s system in the Réponse attracted Hume’s attention, and that it acted as a run-up toward the themes dealt with in the much more substantial “Spinoza” article of the Dictionnaire, which, as we know, the entire second part of the section “Of the immateriality of the soul” is centered on. But there are other, even more telling clues revealing the undoubtable influence of this chapter of the Réponse on Hume’s Treatise discussion of the localization of perception and thought. The equivalence of “being extensive” with “being divided into parts”, writes Hume, is an “argument commonly employ’d”¹⁵; the alternatives which he then develops clearly indicate Bayle’s influence on the topic. Let us suppose, Hume writes, that the soul is conjoined with extension; if so, “it must exist somewhere within its dimensions”. And

If it exists within its dimensions, it must either exist in one particular part; and then that particular part is indivisible, and the perception is conjoin’d only with it, not with the extension:

Or if the thought exists in every part, it must also be extended, and separable, and divisible, as well as the body; which is utterly absurd and contradictory.¹⁶

The mark of Bayle’s reasoning is easy to recognize here. And although the dilemma appears unavoidable, Hume does admit, just as Bayle had before him, that the question does not directly lead to the theme of the soul’s substance, but rather to the (apparently more limited) theme of “its conjunction in place with matter or body”,¹⁷ which is precisely the side-issue that Bayle had chosen to deal with first (rather than immediately tackling the larger question of immateriality). Thus it is obviously not only a reception of Bayle’s arguments and (denounced) aporias that we see in Hume’s Treatise, but a reaction to those arguments, too, and indeed an attempt at their solution. But before we examine those replies, let us first follow the steps of Hume’s analysis, so as to fully appreciate his indebtedness to the Réponse.

Hume knew full well that he was walking through a minefield of controversies, and indeed that he was pronouncing “a maxim which is condemn’d by several metaphysicians, and is esteem’d contrary to the most certain principles of human reason”. We are told about the conclusion to which Bayle had implicitly been led. “This maxim,” he continues, “is that an object may exist, and yet be no where: and I assert that this is not only possible, but that the greatest part of beings do and must exist after this manner.”¹⁸ According to Hume, we can consider an object to be “nowhere” when its parts “are not so situated with respect to each other as to form any figure or quantity”, or when “the whole” is not situated “so as to answer to our notions of contiguity or distance”.¹⁹ The points on which Hume’s approach differs from Bayle’s have more to do with method than with content; whereas Bayle had appealed mainly to metaphysical considerations (the incompatibility between thought and localization), Hume proceeds to the ground of experience and records its features almost descriptively, without invoking abstract categories. Indeed, he finds the soul’s lack of location rather typical of perceptions other than those of sight or touch, and so he counters the argument that had denounced “the absurdity of supposing them to be nowhere” (the outrage that Bayle had attributed to scholastic philosophers) with the simple remark that it is impossible to derive the idea of extension from “passions and sentiments” like we can from tactile or visual perceptions. More generally, a sort of equivalence holds in Hume’s philosophy between appearing and existing, and between the possibility of conceiving and the possibility of existing: “If they [the perceptions of passions and sentiments] appear not to have any particular place, they may possibly exist in the same manner; since whatever we conceive is possible.”²⁰ That which appears not to be localized, may thus exist in this same manner.

Therefore, the dilemma is the same one displayed by Bayle: On the one hand, some kinds of thoughts or perceptions seem incompatible with having a location; on the other hand, it is difficult to conceive of an existence that does not literally have a place, and thus to conceive of a spiritual nature that has no “local union” with the body and extension. However, whereas Bayle had drawn sceptical conclusions from this, or had fallen back, faute de mieux, upon a dualism of the Cartesian type, Hume, as a good empiricist, stops at simply noting the fact, and warning metaphysicists not to place too many arbitrary or a priori restrictions on reality. That which appears in one manner – that which appears as localized, or as unlocalized – may, indeed, really exist that way.

A demonstration then ensues that “this question of the local conjunction with objects” is present not only “in metaphysical disputes concerning the nature of the soul… even in common life we have occasion at every moment to examine it” – for example, when we localize tastes in those extended objects to which they appear to be connected.²¹ (It is worth noting that Hume refers indifferently to “objects” and to “perceptions” as though they were synonyms.) Thus, Hume’s (implicit) reply to Bayle’s subtle arguments is backed by an analysis of experience. Certain perceptions do give rise to spatial location (i.e., visual and tactile perceptions), whereas others are incompatible with this manner of existing in loco (tastes, smells, feelings, passions). Despite this, Hume’s well-known excursus on the “places” of tastes (recall his fig or olive situated at the two ends of a table) shows that, at least when it comes to common sense, men reason differently, and wind up falling back on that very principle which Bayle had criticized as being obscure and confused.

Furthermore, Hume, too, casts ridicule on the typically scholastic claim that an immaterial substance is present “entirely at every part of space”, as though it could “penetrate” that same space occupied by the body.²² And, as had Bayle, he uses a scholastic phrase in describing the conviction that a perception, or a soul, or indeed any spiritual entity is “totum in toto & totum in qualibet parte”. He, too, finds that this principle, “when crudely propos’d, appears so shocking”,²³ since it is the equivalent of saying “that a thing is in a certain place, and yet is not there”.²⁴ Even though it “suits our most familiar way of thinking”, the proposition seems such an “absurdity”²⁵ that our reaction is to locate in space something that actually has no place (“to bestow a place on what is utterly incapable of it”).²⁶ We are led, in fact, to complete the union of ideas founded on “causation” and “contiguity of time” with “conjunction in place”. We give a taste, for example, which is not something extended, a local union with an extended body. The perception, then, (in this case a perception of taste) ends up being identified with the whole extended body yet without its being divided into parts.²⁷ Thus common sense appears to be split between two contrasting principles: the “inclination of our fancy”, on the one hand, which pushes us to give locations to perceptions that, in reality, are “nowhere”,²⁸ and “reason”, on the other, which denounces the possibility of such a conjunction.²⁹ Having made the appropriate distinctions, common sense is an interpretation of that confused dynamic attributed by Bayle to the “chaos” of scholastic thought, while reason, by contrast, represents those certainties which Bayle had described in philosophical terms as Cartesian. Of course, the long détour through the analysis of experience is typical of Hume, and has no equivalent in Bayle; yet some of the alternatives outlined by Hume had already been described by Bayle, even though this latter had settled for a compromise between his underlying scepticism and his moderate sympathy (or perhaps nostalgia) for the clear distinctions of Cartesian dualism.

3 Bayle’s Dilemma and Hume’s Trilemma

In fact, it is Bayle’s scepticism that seems the more radical, for it is directed at philosophy itself, and not just common sense. Those aporias that Hume would (more descriptively) grasp at the level of perceptions, Bayle found also to be true of the soul, as well. And Hume’s Treatise is leavened considerably with irony, such as when he asks himself about the spatial relationships that ought to exist between one perception and another (e.g., between the taste and the place of the tasted object), or about the local relations that exist between “the indivisible subject, or immaterial substance” and the sensations that carry extension.³⁰ For Bayle, the worst outrage is simply the incomprehensibility of the soul-body relationship. If the soul is not extended, he reasons, then it follows that “the soul cannot find itself in any space, nor be united with any matter.” His conclusion is clear: “it is therefore false that it exists in the body of a man”. The fact that thought is incompatible with divisibility into parts (another fact which would be repeated by Hume)³¹ may be a Cartesian tenet with a priori validity, but for Bayle this validity cannot eclipse the “embarrassment” of the consequence into which it precipitates us, again that “there is no relation of place between souls and bodies”. For Bayle, neither the “evidence” nor “any distinct notion” in reality accompany this type of discourse.³²

The analogousness of Bayle’s and Hume’s respective criticisms extend also into certain specific and not unimportant areas. The first is evident. Hume attacks the belief whereby taste, for example, “exists within the circumference of the body, but in such a manner that it fills the whole without extension, and exists in every part without separation.”³³ This belief is analogous, at the level of common sense, to the philosophical subterfuge criticized by Bayle when he mocks the scholastics. In distinguishing between « occuper un lieu circonscriptivement » and « l’occuper définitivement », he says, the scholastics were trying to assign a “local presence” to “immaterial natures” in order to make them wholly to be “at every point in space, so that, without being composed of parts nor being extensive, they occupy a three-dimensional place.”³⁴ Clearly this is the same principle attacked by Hume with his “totum in toto & totum in qualibet parte” reference,³⁵ and which for both philosophers is the height of absurdity.

From a more general standpoint, both Hume and Bayle fully understand that their respective discussions of “local union” are just prologues to the much more serious problem of the nature of the substance of the soul, and its relationship with perceptions.³⁶ Indeed, the entire second part of Hume’s section is dedicated to this theme, and here the contribution from Bayle’s “Spinoza” article was huge, as has clearly been demonstrated by Kemp Smith.

To complete the examination of Hume’s “Of the immateriality of the soul” we must also add, however, that his exposition contains not only an echo, but also a reply to Bayle’s aporias. At a superficial reading, it may appear that most of the problems get resolved when we give up trying to assign a place to something that absolutely cannot have one; this would halt the inclination to add the relationship of “local conjunction” to those of causality and temporal contiguity. In fact, what Hume outlines is not just Bayle’s two-pronged dilemma, but a trilemma:

For we have only this choice left, either to suppose that some beings exist without any place; or that they are figur’d and extended; or that when they are incorporated with extended objects the whole is in the whole, and the whole in every part.

The second of these three terms – having extension and shape – Bayle had ruled out for immaterial substances, and Hume rules it out too for simple perceptions, since that which is not spatial cannot be divided into parts. The third term falls under the same objections that Bayle had reserved for the scholastics, and which Hume, in almost the same terms, moves against common sense; the scholastics and the common man are brought together under the “absurdity” of imagining a whole that is simultaneously present in every individual part. Thus there remains only one possibility, the first term of the trilemma: qualities that spatially exist nowhere. “The absurdity of the last two suppositions proves sufficiently the veracity of the first.”³⁷ Hume appears to reach the same results attained by Bayle: that substances or qualities exist that are not capable either of location or of local conjunction with extended substances.

However, the conclusion of the Treatise differs in its overall sense from the outcome of Bayle’s reasoning in the Réponse. The latter had related the above conclusion to the theses of Cartesian dualism, though with all the limits and difficulties that surround them. But Hume, for his part, keeps his analysis at the level of a description of experience, without digressing into metaphysical declarations, not even by default. This prudent empirical approach translates into a declaration of apparent equidistance from the opposed theses of the materialists and the immaterialists. Moreover, this prudence has overtones of open scepticism, since the impossibility of reaching a satisfactory notion of substance represents “a sufficient reason for abandoning utterly that dispute concerning the materiality and immateriality of the soul”; indeed, it inclines Hume to “absolutely condemn even the question in itself.”³⁸

4 “Theologians” and “Free-Thinkers” Are Only Apparently Equidistant

Yet as Kemp Smith made clear,³⁹ Hume is only apparently equidistant from the materialists and the immaterialists in his analysis of the nature of the soul. While it is true that he condemns the materialists (“the materialists who conjoin all thought with extension”) for having related the properties of extension to qualities belonging neither to sight nor to touch, Hume also says that “a little reflection will show us equal reason for blaming their antagonists, who conjoin all thought with a simple and indivisible substance.”⁴⁰ And indeed the entire second part of “Of the materiality of the soul” (again, which is based on Bayle’s “Spinoza”) is dedicated to this latter aporia. Although the indivisibility of thought is the common starting point for both Bayle and Hume, ultimately the dualists, which Hume calls more bluntly “the theologians”, suffer a reversal of fortune against the free-thinkers.

The free-thinker may now triumph in his turn; and, having found there are impressions and ideas really extended, may ask his antagonists how they can incorporate a simple and indivisible subject with an extended perception. All of the arguments of theologians may here be retorted upon them. Is the indivisible subject, or immaterial substance, if you will, on the left or on the right hand of the perception? Is it in this particular part, or in that other? Is it in every part without being extended? Or is it entire in any one part without deserving the rest? ’Tis impossible to give any answer to these questions but what will both be absurd in itself and will account for the union of our indivisible perceptions with an extended substance.⁴¹

Using the same method of retort, Hume actually went a lot further than the sceptical results achieved by Bayle, and to which he himself appeared to return when he wrote that “the final decision upon the whole” is that “the question concerning the substance of the soul is absolutely unintelligible”. In truth, this admission of ignorance is merely a backdrop for the step taken in the Treatise well beyond both scepticism or dualism; there, a materialist perspective is outlined which extends the conclusions of reason beyond strictly empirical data (the recognition of “the constant conjunction”) and into matter as the cause of thought:

And as the constant conjunction of objects constitutes the very essence of cause and effect, matter and motion may often be regarded as the causes of thought, as far as we have any notion of that relation.⁴²

Thus Hume replaces the metaphysical, Cartesian categories with a different type of analysis, that of experience, in terms of “the constant conjunction of objects”, as they are experienced. Furthermore, he asserts that this kind of analysis can decide questions such as the cause of thought. Whereas Bayle had oscillated between sceptical desperation and a retreat, faute de mieux, to Cartesian dualism, Hume conjures a new empirical method that would join the deconstruction of the category of cause, and its reduction to the constant conjunction, with the description of experience. Cartesian metaphysics had insisted on the unbridgeable gap between the properties of matter and those of thought, denying that the latter could be derive from the former. Hume replies in a way that is quite original, even in comparison with Bayle. Since we generally do not perceive any “connexion betwixt causes and effects”, says Hume, and since it is only by our experience of their constant union that we can know something of that relation, so too in the case of thought and matter the only possible way to know the cause is to register the “constant conjunction of thought and motion”, a conjunction of which we are aware in “the operations of the mind”.⁴³

With these observations, the author is fully and declaredly aware that he is shifting from “hypotheses concerning the substance and local motion” to another hypothesis, one that is “more intelligible than the former, and more important than the latter, viz. concerning the cause of our perceptions.”⁴⁴ Whereas in the first direction the analysis unavoidably stumbles into the well-known incompatibilities between thought and extension, in the second direction, on the contrary, the research reaches affirmative conclusions. At this point it is hardly necessary to point out the distance of this outcome from the dualism with which “Of the immateriality of the soul” begins; its distance from Bayle’s default Cartesianism is even wider.

5 Bayle as a Source and the Debate on the “New Hume”

How might this further Baylean influence, which we have just outlined, fit into recent literature on Hume? As Luigi Turco has pointed out, whereas in the past “interpretations of Hume revolved around the dilemma introduced by Kemp Smith, naturalism versus scepticism”, the horns are now “realism and scepticism”.⁴⁵ To echo the terms used in Kenneth P. Winkler’s highly critical essay on the “new Hume”,⁴⁶ according to this new trend Hume would be a “causal realist”: in reality, for the “new Humeans” the appeal to more or less necessary objective links is to be meant much more as an assumption than as a belief, as P. Kail has explained. However, with some approximation, and with all the necessary distinctions, this description of the “causal realist” might be adapted both to The Sceptical Realism of David Hume by John P. Wright,⁴⁷ and to The Secret Connexion by Galen Strawson.⁴⁸ As Antonio Santucci has pointed out,⁴⁹ Wright and Strawson both distinguish the Scot’s sceptical epistemology from his realistic ontology; and both (Wright above all) stress that the former does not rule out the latter, and indeed may profitably co-exist with it. Strawson, in particular, distinguishes the ontological thesis (the real existence of causal power) from the epistemological thesis (the knowledge of causality as a mere regular succession). Incidentally, Strawson and Wright differ widely in their approaches and styles, Strawson being more rigorously analytical and highly sober in his use of “sources”, while Wright is much more the historian, and very attentive to the use made by Hume of his preferred “authors”.

In looking at Bayle as one of those sources, it is this second aspect, the historical aspect, that is of most interest here.⁵⁰ The reference point is a letter written by Hume in August 1737 to Michael Ramsay. There Hume lists the works he has used as a guide in the “metaphysical parts” of his reasoning. It is well known that there are four of them: Recherche de la vérité by Malebranche, Principles of Human Knowledge by Berkeley, the more metaphysical articles in Bayle’s Dictionnaire, and lastly Descartes’s Méditations. Now, if we look at the two representatives of the “new Hume” mentioned above, we will see that Strawson quotes Berkeley extensively, yet Malebranche and Descartes hardly at all, and he never mentions Bayle. On the other hand, in Wright’s book Malebranche becomes the author of reference thanks to his theories on cerebral traces, animal spirits, and “natural judgements”. His “science de l’homme” provides the mechanistic and realistic framework for Hume’s analysis.⁵¹

What has happened, in this new perspective, to Bayle? Fundamentally, the passages indicated by Wright are the same ones already pointed out by Kemp Smith and Popkin: the articles “Pyrrhon”, “Zénon d’Elée”, and “Spinoza” plus the arguments of the Stratonician atheist of Continuation des pensées diverses.

Turning to other studies, there too we find Malebranche having had better luck than Bayle. In his careful analysis of “Of the immateriality of the soul”, John Yolton stresses the fact that Hume apparently had Malebranche chiefly in mind, although with an ironic intent: his main objective appears to have been to “satirize” the representative theory of ideas.⁵² More recently, P. Russell has expressed the conviction that the principal philosophical context of this section is given by the dispute between Collins and Clarke on the possibility that matter can think, in the wake of the hypothesis put forth by Locke in his Essay.⁵³ In sum, Bayle appears to have lost his appeal for Hume scholars – or perhaps it is thought that the work done by Kemp Smith and Popkin has definitively exhausted this vein.

The analysis we have provided here shows this perception to be misguided. Indeed, to have identified Bayle as the implicit source of the first part of the section, just as his article “Spinoza” is the explicit and recognized source of the second part, is not just to have made a philological contribution, one as limited as it is valuable. In fact, the predominance of Bayle as the “source” of the entire fifth section of the Treatise significantly displaces the terms of its interpretation.

6 Three Conclusions: Hume’s Post-Baylean Philosophy

Let us now briefly sum up the main points of our interpretation:

• (a) It is not the Cartesian principle of substantiality that enables Hume to resolve the dispute between “free-thinkers” (or materialists) on the one hand, and “theologians” (or immaterialists) on the other, as Turco maintains,⁵⁴ but just the opposite: the very doubt cast by Bayle, and cast again by Hume in the first part of the section, on the “hold” of Cartesian dualism with regard to the problems of localization, compels Hume to abandon keeping a notion of substance that we do not actually have, and which is for this very reason a source of paradoxes and other insurmountable difficulties. The pars destruens of scepticism (its critical function) is taken entirely from Bayle in the first part of the section, as it is in the second part (although from different works, respectively the Réponse and the Dictionnaire), and contributes greatly to the section’s overall thematic unity. Despite the Cartesian overtone to the arguments with which it begins, Section V is largely a reflection on the crisis and the dissolution of metaphysical dualism, and in general of all metaphysical approaches. (This includes the approach of Spinoza and of the theologians attacked in the second part.) The trajectory of the argument toward a sceptical crisis is also typical of Bayle: it begins with notions that in themselves seem evident to arrive at unacceptable consequences which rebound against the assumptions; or it shows that the evidence of a priori notions is incompatible with equally certain a posteriori ones.⁵⁵

• (b) Both Wright and Anderson⁵⁶ argue against Yolton, who attributes a “satirical” or chiefly ironic intent to Hume’s fifth section. Indeed, Yolton states that “Hume is clearly interested in satirizing both positions”, those of the materialists and of the immaterialists alike. In reality, throughout this section Hume goes well beyond the sceptical equidistance that seems implicit in Bayle’s final impasse in the Réponse, which he echoes in the first part. Furthermore, it must be stressed that Hume is only able to do so because he has fully assimilated Bayle’s lesson. To use Hume’s own words, which go right to the kernel of Bayle’s reasoning, “the question concerning the substance of the soul is absolutely unintelligible.” Though he starts from Bayle, he goes well beyond him, since he shifts the discourse from the Cartesian analysis of substances (whose attributes are mutually exclusive, or remain “distinct and separable”, as Hume puts it at the start of the section), to concentrate on the “constant conjunction” of impressions and ideas as we actually find them in experience. From this standpoint, as Yolton points out at the end of his analysis, the relationship between matter and movement poses no more of a problem than does the relationship between matter and thought. If we admit a relationship of causality in the first case, then it is not clear why we should not do so in the second case as well. To quote Yolton, who sums up Hume succinctly: “the condition necessary for assigning cause with respect to matter and motion is also met with motion and thought.”⁵⁷ Only the principle of causality in Hume’s formulation enables us to discover that there is effectively a causal relationship between the movements of the body and the thoughts of the mind.

• (c) It is true that certain passages of Bayle (such as Ch. XV of the Réponse) may be interpreted, and indeed have been interpreted, as the expression of a species of desperate dualism, faute de mieux. However, leaving aside problems of interpretation, which are always complicated in the case of Bayle, most scholars do recognize that the general tone of his reflection is clearly sceptical and anti-metaphysical, and that he insists much more on the pars destruens than on the pars construens. Cartesianism comes out of Bayle’s analysis in much worse condition than when it entered. For this reason, too, and for the role that Bayle implicitly plays in the first part of Hume’s Section V, it is very difficult to believe that Hume was able to hold together “dualism” and “scepticism” in the terms in which Bricke presents them: “His [Hume’s] brand of scepticism commits him to the plain man’s metaphysics, or possibly to the metaphysics of enlightened common sense… The metaphysics of the plain man or enlightened common sense is, in Hume’s view, a dualist metaphysics. It is also an interactionist one.”⁵⁸ While I greatly doubt that so pacific a position could be obtained from so tormented source as Bayle, I am completely certain that one can rule out the correctness of such a description applied to Hume. Effectively creating a connection between the implicit Bayle of the first part (the one which is informed by the Réponse) and the explicit Bayle of the second part (the part referring to the article “Spinoza”), Hume overcomes the dualism of the Cartesian type, which in any case would imply the availability of the notions of the two metaphysically distinct res. Indeed, with his criticism Hume attacks the very category of substance itself. In this connection, the conclusion that Yolton draws is correct: “Hume sought a way around these problems by rejecting some of the standard concepts. Other eighteenth-century writers struggled to find a solution within the traditional concepts of matter and mind.”⁵⁹

Naturally, it would be wrong to attribute to Bayle tout court what Yolton attributes to Hume in his first statement; but certainly Bayle does not fall under the category of thinkers described in the second statement; and it is for this reason, too, that his lesson could be so stimulating and influential for the author of the Treatise.

In a well-known book, Donald W. Livingston described Hume’s general approach as “post-Pyrrhonian” philosophy.⁶⁰ Going further, we might well ask ourselves whether “Of the immateriality of the soul” in particular is not, above all, “post-Baylean” philosophy. Hume’s great merit is that, proceeding from Bayle’s penetrating analyses, he opened up new paths, going well beyond the impasse at which the latter had gotten deadlocked.