Bayle and Panpsychism

1 Introduction

In his writings, Pierre Bayle demonstrates a remarkable skill for combining erudition with topical discussions. Some of the articles devoted to ancient philosophers in the Historical and Critical Dictionary provide him with an occasion to discuss materialism. Bayle makes a surprising suggestion: in order to avoid devastating objections, materialists should postulate that the property of thought does not emerge from certain combinations of matter but is present from the start in any portion of matter. This proposal is strikingly similar to the view recently revived by Thomas Nagel and Galen Strawson, which is termed “panpsychism.”^[1] In fact, Bayle seems to be one of the first philosophers to describe what this kind of panpsychism could be, that is to say, an anti-emergentist theory that is materialist (as opposed to Leibniz’s), non-monistic (as opposed to Spinoza’s), and according to which matter as such, in each of its basic constituents (which Strawson calls “ultimates”, whatever they are), possesses a mental property, or, more simply put, some form of consciousness (generally called in the current discussions “experience” or “what it is like to be-ness”) instead of having to be permeated by a complementary principle (some “soul of the world”, the Stoic pneuma, a vital force, a substantial form, etc., as in earlier theories^[2]) for becoming conscious. In other words, this form of panpsychism is purely “physicalist,” but matter does originally have another fundamental property besides non-experiential properties: that of experientiality.

Is Bayle’s suggestion to be taken seriously and does it really aim at improving the materialist position? Or does Bayle merely intend to reveal the implausibility of materialism? On the face of some of the odd consequences of panpsychism that Bayle points out, such as cognitive faculties in corpses, the second alternative looks obvious. However, Bayle does not shy away from paradoxes and he often advances serious thoughts in a joking way. Furthermore, there is reason for entertaining the idea that Bayle might actually consider panpsychism to be tenable, as he could use the same line of defence that he outlined for another kind of materialism, namely, Stratonism. It is worth, then, taking a closer look at this possibility. If it proved to be a real possibility, a line of interpretation that presents Bayle as a forerunner of eighteenth-century materialism would be confirmed.^[3]

However, accepting the panpsychic hypothesis would lead Bayle to a view similar to Locke’s superaddition theory and I contend that such cannot be his position because he embraces the Cartesian principle that each substance has only one principal attribute. This makes unacceptable, in his eyes, the conjunction of thought with matter in the same being. I will therefore conclude that Bayle considers any kind of materialism to be untenable. By contrast, this will make clear which kinds of metaphysics and epistemology panpsychists need to adopt to defend their view.

2 From Materialism to Panpsychism

3 Is Panpsychism Possible?

3.1 A Respectable Materialism

However, one might ask: why not endorse panpsychism, after all? At times, Bayle himself does not seem to exclude this possibility. In the article Leucippus (main text) of the Dictionary, rather than driving the ancient atomists to panpsychism as to an implausible assumption, he appears to deplore sincerely that they did not take the step of supposing that the atoms are conscious.^[55] With this move, Bayle explains in Remark E of the same article, the atomists would have gained two important advantages.

We are already acquainted with the first advantage through the articles Epicurus and Dicaearchus: the atomists would be in a position to respond to Plutarch’s and Galen’s objection (argument 1). The second advantage is that, according to Bayle, they could dodge an objection that is none other than the “Achilles” (that is, the strongest argument) of rational psychology, as Kant would later call it. In short, this argument rests on the incompatibility between the unity that is characteristic of mental representations and the multiplicity that is characteristic of extension (divisibility being an essential property of matter). The simultaneousness of a (hypothetical) manifold of sensations in different parts of a body (such as the different parts of the brain) would not make one perception; parts thinking together would not produce one thought. Bayle does not explain in detail how the Achilles would be defeated. He only hints at the principle of the answer: since the Achilles’s thrust rests on the indivisibility of that which, in us, has conscious representations, the atomists could claim that this condition obtains in their system, since each atom is indivisible.^[56] Admittedly, it is not clear why the difficulty should not subsist for a collection of atoms. Each of them would be a center of perception, but which one would say “I”? Which one would be the seat of the unity of consciousness?^[57] However, perhaps it suffices, for Bayle’s present purpose, that at least one atom can sense or think. What is in question here is defending against the Achilles the very possibility that matter thinks, not explaining the compounds that actually exist.^[58]

In any case, Bayle considers that the “Democritean” hypothesis deserves consideration because it shelters the atomists from devastating objections. Moreover, Bayle points out that another problem which used to plague atomism, namely, the non-existence of vacuum,^[59] has been dismissed by recent physics.^[60] Furthermore, referring to Maimonides’ Guide of the Perplexed, he points out that some Arabic thinkers, the “speakers”, that is to say, the mutakallimun (the proponents of kalam), have in effect sustained the thesis of sensitive atoms.^[61] Finally, he rehabilitates atomism by separating the question of the existence of atoms from that of their eternity and that of their chance encounters. Once they ruled out that atoms are uncreated and randomly combined (which would never yield regularity and suitable design in the effects),^[62] modern philosophers, such as Gassendi, have been able to devise “a very nice system.” In fact, Bayle adds, all modern philosophies, even those which are not strictly speaking atomistic but corpuscularian in a looser sense, try to explain natural processes according to the principles of mechanism, that is to say, through the diverse combinations of parts of matter, by contrast with the useless Aristotelian forms.^[63]

Thus, with some alterations, atomism has become a decent theory. It might be, then, that in the article Leucippus Bayle earnestly suggests to the atomists a third improvement, which would render their philosophy perfect: postulating that each atom is always animated (P1 and P2).

This would not necessarily mean that Bayle rallies to atomistic panpsychism. In fact, he has serious objections against the idea that any part of matter could be indivisible. He borrows one of Aristotle’s classical arguments, namely, that the atoms cannot be strictly speaking indivisible and simple. Since they are supposed to hook together, one must be able to distinguish in an atom the part by which it is in contact with another atom from its other parts.^[64]

However, building on Bayle’s suggestion, it seems that other kinds of materialism, namely, non-atomistic ones (such as Dicaearchus’s, or modern corpuscularianism), could derive similar benefits from the move that Bayle recommends to the atomists, and as a result they might have a stronger appeal.

First, non-atomist materialists too would not face the impossible task of showing how thought can emerge from extension, as we already know (argument 4). But there is more. They could also repel the “Achilles of rational psychology”, although in a different way than the atomists (since they could not appeal to the indivisibility of atoms). When he was writing the article Dicaearchus (before 1697), Bayle still believed that the “Achilles” was unobjectionable. This proof, he says, “has always seemed to me very proper to show the impossibility of joining together the three dimensions and thought in the same subject.”^[65] However, as Todd Ryan has emphasized,^[66] Bayle realized, at the latest in the Réponse aux questions d’un provincial (1704), that the “Achilles” may not work if, as Locke contends, it is not evident that extension is the essence of matter. In fact, Locke claims, we do not know what matter really is, because a substance is the unknown substrate of some of the properties we know. Therefore, we cannot be sure that thought is incompatible with extension. They may very well have opposite characteristics, but both could belong to the same unknown substrate that is neither one nor the other. In that case, the divisibility which characterizes extension would not have to be attributed to thought nor to the very subject which thinks. Consequently, Locke says, one may hypothesize that God, by virtue of his omnipotence, could make some matter think, that is to say, could add, as a property among others, the property “thought” to a substance that possesses the property “extension.” In the same way, a real panpsychist (which Locke is not)^[67] could contend that matter is a substrate that eternally carries the properties of extension and thought, which are not reducible one to the other, or to matter. As a result, the divisibility of extension would not affect thought and the “Achilles” would be repelled.

Last but not least, it seems to me that the panpsychic hypothesis could bring another major improvement to materialism, namely, a solution to the problem of “blind causality.” A usual objection to materialism, as we just saw, is that complex, organized effects require a causal agent capable of reflection and intention. Moreover, in the second half of the 17^th century, Geulincx (and, to a certain extent, Malebranche) imposed on causality an epistemic condition: in order to qualify as the cause of an effect, a thing must know how to bring about this effect.^[68] In agreement with them, Bayle often emphasizes that if secondary causes are to be efficient causes, they should be cognizant of the way of producing their effects. This condition would be fulfilled in panpsychic materialism. If matter is endowed with thought, it can produce complex effects such as living beings.

All that is needed to reap these benefits is to suppose that the two properties of thought and extension, independent from each other, eternally coexist in the same subject (without, therefore, thought being derived from extension). This move is possible if one adopts an ontology that allows a plurality of attributes in the same substance, much as Spinoza or Locke did each in their own fashion.^[69]

3.2 Stratonism and Panpsychism

However, such a straightforward assumption may seem quite arbitrary. What gives a materialist the right to postulate that matter originally has the property of thought? Well, Bayle himself might provide some justifications for that claim. Referring to the genuine, ancient atomistic position, for which atoms are uncreated and eternal, Bayle concludes Remark F of his article Epicurus in an ambiguous manner:

Moreover, it is not more absurd to suppose that atoms are essentially animated, than to suppose they exist and move of themselves.^[70]

What does this statement mean? Should we read this sentence as an antiphrasis? Does it signify that it is no more absurd, but also not less absurd, to ascribe a soul to atoms than to imagine them in eternal movement? Or does it mean, rather, that it is legitimate to grant thought to the atoms, after their non-created existence is presupposed? In other words, according to this second reading, the atomists make the big leap when they postulate that atoms possess, by nature, the properties of existence and motion, and it would only be a small leap further to claim that they possess the property of thinking as well.

Admittedly, the mere fact that the second assumption is not more arbitrary than the first one would be a poor reason for endorsing it. Yet, we should take a closer look at this possible reading of Bayle’s sentence. To suppose that atoms exist by themselves, that is to say, without a cause, is tantamount to affirming that they have the same property of being a se that God has in Christian theology. This property of aseitas (this is not the word that Bayle uses here, but the idea is in effect present) should draw our attention, because the thrust of the famous Stratonician retort, in the Continuation des Pensées Diverses, hinges on that notion.^[71]

In the Continuation, Bayle attempts to demonstrate that rational theology is not able to refute atheism. For this purpose, he imagines a dialogue between, first, theistic philosophers and disciples of Strato of Lampsacus, later respectively replaced by Christian theologians and more or less fictional Chinese philosophers to whom Bayle ascribes the same staunch materialist stance as the Stratonicians defend. It is Cudworth who chose the figure of Strato to epitomize the materialist-atheistic view of uncreated matter which possesses the dynamic resources necessary to engender all living beings, including thinking ones (Cudworth also forged the term “hylozoism” to summarize this view).^[72] Bayle is happy to use the label “Stratonician” to tag this position, but, contrary to Cudworth, he thinks that the Stratonicians, or their modern avatars, the Chinese philosophers, are able to resist their opponents.

The theologians’ strongest argument is the following. Matter, as hypothesized by the Stratonicians, is not absolutely perfect, since it does not have every possible property, according to their own terms.^[73] But a being that exists by itself cannot be imperfect, by virtue of the principle that nothing is limited except by an external agent. There is no such thing as self-limitation because there must be a reason for every state of affairs, and the only reason for a being having these or those limits is that its cause has shaped it in this or that way. Therefore, a being that has no cause cannot be limited.^[74] In other words, a being that exists by itself is necessarily infinite in all respects, and, conversely, a being that is limited cannot be a se.

The Stratonicians cannot but concede the point. They have to admit that they are unable to explain why their Nature, which exists by itself, is limited and imperfect. They can only say state that it is just as it is, for no other reason. But this type of response: “such is the nature of things”, is “the last asylum of ignorance”, as Bayle says, since it appeals to an inexplicable state of affairs.

Yet, the Stratonicians are poised to offer a retort to the theologians’ objection. Let me recall first that the notion of “retort” (antistrophe, in ancient rhetoric) has a technical sense. You retort when you prove that your opponent is plagued by the same difficulty he or she points out in your position. Your move, therefore, is merely a negative one. You do not solve the difficulty objected to you; yet, you obtain a tie game. You are entitled not to modify your position, given that the opponent’s is no better.

In retort, then, the Stratonicians just have to ask the theologians why their God can do certain things and not other things. As a matter of fact, it is commonly agreed in Christian theology that that which contains an intrinsic contradiction, such as squaring a circle, cannot be done, even by God. Therefore, God’s power is limited to non-impossible essences, and the theologians are obliged to concede that there is no reason for this state of affairs, except that essences are as they are: some are possible, others are not. In other words, they can only invoke the nature of things, exactly as the Stratonicians do.^[75]

Thus, since the theologians take the liberty to affirm that God, who exists by himself, exists with such and such limits, the Stratonicians are entitled to require equality of treatment and claim the right to suppose that their matter is imperfect and nevertheless eternal, uncreated, and endowed with certain properties – the properties that are necessary to produce the variety of beings and the order of the world.

Now, panpsychists could make the same claim. True, the “Democritean” hypothesis is not exactly identical to the Stratonician one, since it presupposes the presence of mind in matter ab initio, while a strict, reductionist materialist theory, as Stratonism is supposed to be, would have to show how the mind is a product of matter (which is, for Bayle, an unfeasible task, as we know).^[76] Yet, Strato could inspire Democritus, if I dare say so. That is, a defense of panpsychism might operate along the same lines as the Stratonician retort. A panpsychist materialist would be entitled to say: I too have the right to suppose that matter not only exists by itself, although imperfect, but also is endowed with certain characteristics (and not with some others), among which is the property of thinking.

Such, then, could be the meaning of Bayle’s sentence that we are scrutinizing: it is not more absurd for materialists to postulate that matter by nature thinks than to postulate that it is a se, because they are as much entitled to endorse both these assumptions as theologians are entitled to affirm that God’s power has certain limits and that God has no cause. We may wonder, therefore, if Bayle is not hinting at one of the “tie games” he is fond of – a tie game, that is, between panpsychism and theism.

3.3 Descartes versus Democritus

Yet, is panpsychism really viable, to Bayle’s mind? The end of Remark E, in the article Leucippus, allows doubt, I believe, that he considers this solution as a real possibility. After having once more advised the atomists to adopt the “Democritean” hypothesis, Bayle reduces this assumption to the same degree of probability, or rather, improbability, as the very first assumption in which the atomists have indulged, namely, the aseitas of atoms:

They had no less right to suppose atoms animated than they had to suppose that they were uncreated, possessing the attribute of the power of self-motion.^[77]

They had no less right, but one can say as well that they had no more right to do so. In effect, Bayle is here more explicit:

It is as difficult to conceive of this power [of self-motion] in an atom as to conceive of its having sensation. Extension and solidity make up the whole nature of an atom according to our ideas of it. The power of self-motion is not contained in that idea. It is something that our ideas find alien and ‘extrinsic’ to bodies and extension, in the same way that knowledge does not seem to belong to them.^[78]

Bayle unquestionably speaks here as a Cartesian. By means of clear and distinct ideas, he notices no other properties of a body than extension and impenetrability, and these properties make the essence of bodies. A motive force inherent in matter would be a hidden potency comparable, and indeed compared by Bayle, to the moving force attributed to the soul by Aristotle. The same goes for thought.

As we saw earlier, Bayle’s Cartesianism is also manifest in the article Dicaearchus, Remark L (argument 4, above). After having affirmed that, up to now, no one has dared to say that “to make a substance pass from the absence of all thought to actual thought, it sufficed to move it”, Bayle additionally insists: “And even if some people should boast that they clearly conceived this, they do not deserve to be believed.”^[79] To silence these braggarts, one simply has to point, says Bayle, to a certain passage in Aristotle’s Metaphysics. This passage is none other than the one formulating the principle of non-contradiction: the same thing cannot at once be and not be.^[80] This comes down to adding a new, quintessentially Cartesian argument by which Bayle attempts to reduce to outright nonsense, not only the claim that matter produces thought, but also the weaker claim that one can clearly conceive that an extended thing thinks:

[5] To suppose that mind and matter coexist in the same subject entails a contradiction.

This is tantamount, Bayle contends, to affirming that the same thing simultaneously possesses and does not possess a certain set of properties.^[81] In fact, the characteristics of matter and thought are mutually exclusive: occupying space and being divisible on the one hand, not occupying any space and not being divisible on the other hand. Therefore, the two sets of characteristics cannot be conjoined in the same substance. They can only be found in numerically distinct subjects.^[82]

This implies that Bayle necessarily rejects the panpsychic claim that the two properties of thought and extension might coexist in a material substrate. It also implies that Bayle rejects Locke’s claim that God could implement thought in a material being. If the notion of thinking matter is intrinsically contradictory, thinking matter is therefore an impossibility. This conclusion exceeds the incomprehensibility that Locke acknowledges, but that he judges surmountable by God’s omnipotence.^[83] That which is incomprehensible for us is not necessarily impossible (in other words, God can do things that we do not understand, but that are possible), whereas that which plainly contains a contradiction is absolutely impossible, even for God.^[84] Therefore, Locke argues, in order to exclude the possibility of thinking matter, one ought to prove positively that this idea contains a contradiction. But this is exactly Bayle’s point: he finds an evident contradiction in the notion of thinking matter; thence, even God cannot make matter think.^[85]

Surely, Locke would defend his view by saying that extension is not the essence of matter, as we saw earlier. That is why, for him, extension and thought do not exclude each other. These two properties could coexist in a substance of which we do not know the essence. In other words, the duality of properties that are not reducible one to the other does not necessarily imply a duality of substances. However, Bayle is not willing to allow this possibility, even in such a late work as the Réponse aux Questions d’un Provincial:

[…] we know that the essential attributes of a substance are not numerically different, and thus we may not believe that it is possible that matter relates to space by one attribute, and to thought by another.^[86]

The fundamental reason for Bayle’s disagreement with Locke, then, is that Bayle unshakably adheres to two Cartesian theses:

[6] There is only one principal attribute per substance, and this attribute constitutes the essence of the substance.^[87]

 

[7] Extension is the principal attribute of matter, that is to say, is the essence of matter.^[88]

This is why Bayle maintains that there would be a contradiction in imagining something that simultaneously possesses and does not possess the characteristics of extension. Since extension constitutes the essence of matter, a material thing cannot have properties which are incompatible with extension.^[89] This is exactly what Descartes replied to Regius, who contended that:

[…] if we are to follow some philosophers, who hold that extension and thought are attributes which are present in certain substances, as in subjects, then since these attributes are not opposites but merely different, there is no reason why the mind should not be a sort of attribute co-existing with extension in the same subject, though the one attribute is not included in the concept of the other.^[90]

First, Descartes answers, Regius fails to distinguish modes from attributes (in the sense that Descartes gives to these terms).^[91] An attribute is not in a substance as if the latter were a subject that is different from this attribute. The principal (or essential, as Bayle says) attribute constitutes the essence of the substance and is inseparable from it, contrary to modes. Second, thought and extension are not just different from each other, they are opposites. This is because they each constitute the nature of a substance, and a simple substance cannot have two natures. If a simple substance is X (its essence), it is not-Y (another essence), and if it is Y, it is not-X. So a substance cannot be X and Y, that is, X and not-X. In the case of principal attributes, difference necessarily means contrariety and mutual exclusion.^[92] And we know that a material substance is a simple subject, because, upon inspection, all the properties we notice in it are reducible to extension and are mere modifications of it.^[93] Therefore, extension is the essence of that substance, and there is no room for any other essential attribute.^[94]

3.4 Possibility and Compatibility

Further, Bayle rejects Locke’s recourse to God’s omnipotence on the grounds that, should God decide to exert his power, he would nevertheless do so within the limits of certain conditions. Locke’s hypothesis, i. e. God giving matter the faculty of thinking, would be a case of what Scholastics call potentia obedientialis: because they are created, all things remain in the dependence of their creator, have a “disposition to obedience”, and therefore can be reshaped at will by God, which means that God always can make creatures have properties it is not their nature to possess, or produce effects other than those their natural powers can produce. But for Bayle God’s actions, even those carried out by his absolute power, are always met by a corresponding possibility on the side of the creatures.^[95] As a result, a miracle never is against nature.^[96] Indeed,Bayle criticizes the distinction between natural potency and “obediential” potency on the ground that “natural” means: according to a certain law of nature that has been freely chosen by God among other possible laws that were no less convenient for God to create a world and no less suitable to the respective essences of the creatures. So when God decides to break a law of nature, he does not act in a way that is less in phase with the essences of creatures.^[97]

True, in Bayle’s conception the range of what is compatible with an essence is rather broad. He affirms that the nature of a stone does not any more entail per se the power of breaking a glass when thrown at it than it entails the power to convert water into blood. The causality relation is not of itself a necessary connection, because it does not depend on the intrinsic natures of the cause and the effect, but on the laws of motion that God has chosen (occasionalist insights evidently underpin Bayle’s argument).^[98] If God, then, had a stone transform water into blood, this would not be against the nature of the stone.

However, for Bayle, shattering a glass and changing water into blood both pertain to the same basic kind of reality and explanation: fundamentally, it is all a question of how parts of extension are moved around. This is why one of these effects is not more per se natural than the other and why in none of these cases God infringes upon the nature of a creature. On the other hand, a philosopher, Bayle says, should never affirm anything that is not distinctly conceivable. We distinctly see that corporeal substances are only capable of receiving an impulse and of the consequences that follow: changes in shape, size, position, etc. On the contrary, one cannot distinctly conceive how, even by God’s intervention, a stone could be made capable of a cognitive act.^[99] For God miraculously to endow matter with thought, as Locke contends is possible, matter would have to be capable of receiving thought.^[100] But for matter to have this capacity is, according to Bayle, strictly impossible, given the contradiction between the properties of the mind and the properties of extension – a contradiction that God himself cannot override.

Furthermore, when Locke states that we are not certain whether thought could not in fact be compatible with matter, he does not tell us what the meaning of “being compatible” is. Bayle is much more specific. Even granted that there is no obvious contradiction, it does not suffice to ask: “why not?” and then float the gratuitous hypothesis that a property could be conjoined with a subject whose essence is unknown. The compatible property cannot just be arbitrarily juxtaposed with this essence. For a property to be attributed to a subject, there is a more restrictive condition: one ought to show that the property in question and the subject that is supposed to receive it are conformable to each other. In Bayle’s own words:

[8] That which is compatible^[101] with something else must be based on the essence of that thing and presuppose this essence.^[102]

When, on the contrary, we establish that this connection is absent, we may be certain that we are dealing with two different natures: “we conceive that two beings belong to different species when the essential attributes of one cannot be conceived to be compatible with the other.”^[103] Bayle states that, for example, qualities such as being colored, round, translucent “are founded on extension, and presuppose it,”^[104] inasmuch as they cannot be conceived but as modes of extension. This is why they can be attributed to what has extension. On the other hand, it is not the case for matter with regard to thought, or for thought with regard to matter: “extension, figure and divisibility are not founded on the essence of the soul, nor do they presuppose it.”^[105] By virtue of [8], Bayle can conclude: “therefore, they cannot be compatible with the soul,” that is, cannot be attributed to it.^[106] Conversely, thought modifications such as affirming, negating, wanting, reasoning, do not logically presuppose extension. Therefore, they cannot be attributed to matter. There must be a reciprocal correspondence between the substance and its properties. If this correspondence is lacking, there is no foundation in the subject for receiving the property, even supernaturally. As a result, thought can never be conjoined with matter in a simple substance.

Thus, beyond the discussion with Locke, Bayle’s position against materialism in general is supported by an argument which is, in fact, the reverse of the argument that should steer materialists towards panpsychism. To be consistent, materialists should say: since certain beings obviously think, but are nevertheless merely material, and since matter is everywhere the same, therefore the whole of matter thinks (argument 2, above). On the contrary, Bayle contends: since there are beings that do not think, “it is certain that the figure and extension” that we find in these beings “do not suppose the faculty of thinking”^[107]; thence, extension is not attributable to thought, because if it were, extension would conceptually imply thought, and therefore would always require it. Conversely, since God is a thinking being, there is at least one case where it is clear that the faculty of thought does not suppose extension, from which Bayle can conclude that thought is never attributable to things that are extended, because if it were attributable, it would conceptually imply extension, and therefore would always require it.

Consequently, if materialists can attribute thought to matter, it is only because ex falso sequitur quodlibet.^[108] From the moment one allows the existence of hidden potentialities or properties in bodies, why not grant also that of feeling and thought? But this does not make them intelligible properties, that is to say, compatible with the notion of matter.^[109]

3.5 The Baylean Strategy

To conclude, I think it safe to say that when Bayle advises the materialists to postulate that matter always thinks, he does not believe he is turning materialism into a flawless system. He is too much of a Cartesian to believe that materialism, even panpsychist materialism, is an option. Consequently, it is wrong to see Bayle as a forerunner of eighteenth century materialism. A contrario, it is clear which sort of background theory panpsychists must have: a metaphysics that allows several essential attributes for a simple substance, and an epistemology that shows we have not an exhaustive knowledge of what matter is.

But, then, what do Bayle’s reflections aim at? Why does he make this suggestion to the materialists? In order to understand this, we must place Bayle’s reflections in the context of his usual strategy when addressing philosophical issues.

First, Bayle likes to pose as an arbiter of the internal coherence of systems. If one chooses to be a materialist, one ought to uphold certain premises, Bayle explains, so as to avoid some obvious difficulties. However, when Bayle points out what a philosopher or theologian could and should have said, this does not mean that he comes round to their opinion. Even when a theory is corrected or complemented according to Bayle’s suggestions, it may remain exposed to objections which he deems insurmountable, such as, in the present case, the mutual incompatibility of thought and extension. Pure matter cannot engender thought. Materialism, then, has to be a panpsychism. But, in its turn, panpsychism stumbles against an external objection, namely, the impossibility that a substance has several principal attributes. Matter and thought cannot coexist in the same being; such is the bottom line for Bayle.

As Bayle’s piece of advice leads to no viable solution, we may uphold our first impression, namely, that Bayle only wants to show the unbearably high price that materialists have to pay for being consistent. This matches another pattern of Bayle’s usual strategy, which is to highlight the unwanted consequences of the theories he scrutinizes. Some side effects of panpsychism, such as thought in corpses, are not easy to come to terms with. Bayle’s intention, beyond his posture of arbiter, is undoubtedly to contrast these odd consequences with the simplicity of Cartesian dualism.

However, in line with the “skeptical” interpretation of his philosophy that I follow,^[110] Bayle systematically points out the aporias of reason. For him, no philosophical system is satisfactory on all points. Accordingly, Bayle does not unconditionally support Cartesianism either. Although he endorses the idea that human beings have an immaterial soul, he highlights in the article Epicurus the problems that Cartesianism faces. Animal life and animal sensibility too must be accounted for. Within the framework of Cartesian dualism, only two solutions are possible. One can either, as Descartes himself, deny animals any psyche and reduce them to being mere machines; or, one can attribute to them minds, but these minds must, for the reasons we have just seen, be distinct from their bodies, that is to say, be “immaterial souls” similar to those of human beings.^[111] Descartes’s radical solution leaves Bayle extremely doubtful.^[112] The second solution is a “dangerous” hypothesis. As it grants animals sensibility and therefore the experience of pain, it fuels doubts about the Creator’s goodness: why do animals suffer?^[113]

Thus, it turns out that Cartesian dualism faces a formidable problem. However, this does not mean that materialism is in a better position regarding the issue of animal life. We saw what Bayle thinks of the “Democritean” hypothesis.^[114] As to non-panpsychist theories, the difficulty that Epicurism encounters with its non-animated atoms (argument 1) affects all non-atomist philosophies that try to explain how an animal psyche comes from the composition of non-sensitive principles.^[115] Such is the case with the four Aristotelian elements. They may associate each to the others in the so-called “mixtures”, but in order to confer the properties of life and of sensitivity to these compounds, the Aristotelians have to suppose that a substantial form called “material soul” or “animal soul” (as opposed to the human, rational soul, which is not reducible to the function of substantial form organizing the body) is added to the mixtures. But, Bayle remarks, “it is a vain subterfuge that is, no less than the Epicurean atoms, devastated by Galen’s objection.”^[116] If one element is not apt to sense (that is to say, to receive a certain soul), there is no reason that a compound of elements should be more capable of doing so.

Thus, all the explanations of consciousness that have been submitted so far are exposed to objections. There are, for Bayle, some bits and pieces, or “islets”, of indubitable truth, such as the certitude that mind cannot be reduced to anything material, but these scarce truths do not solve all the difficulties and reason is perpetually assailed by doubts and aporias.^[117]

1. Descartes, R. 1996. Œuvres de Descartes. 13 vols. Eds. Ch. Adam / P. Tannery. – P. Costabel / B. Rochot. Paris.

2. Descartes, R. 1985. The Philosophical Writings of Descartes. 2 vol. Trans. J. Cottingham / R. Stoothoff / D. Murdoch. Cambridge.

3. Lucretius. 1992. De rerum natura. Eds. W.H.D. Rouse / M. F. Smith. Cambridge, MA.

4. Leibniz, G. W. 1875–90 (repr. 1960). Die Philosophischen Schriften. 7 vols. Ed. C. Gerhardt. Berlin (Hildesheim / New York).

5. Epicurus. 1926 (repr. 1970). Letter to Herodotus. In: Epicurus. The Extant Remains. Ed. C. Bailey. Oxford (Hildesheim / New York).

6. Epicurus. 1926 (repr. 1970). Letter to Pythocles. In: Epicurus. The Extant Remains. Ed. C. Bailey. Oxford (Hildesheim / New York).

7. Bayle, P. 1734–38. Historical and Critical Dictionary. 5 vols. London.

8. Bayle, P. 1727–31 (repr. 1964) Œuvres Diverses. 4 vols. La Haye (Hildesheim / New York).

9. Bayle, P. 1965. Historical and Critical Dictionary. Selections. Transl. R. H. Popkin. Indianapolis/New York.