Critical Genealogy, Comprehension, and Explanation in Leibniz’s Critique of Bayle on Cosmic Dualism

Abstract

The main aim of this paper is to provide an account of Leibniz’s engagement with the doctrine of cosmic dualism in his Theodicy, i.e., the view that there are two distinct fundamental principles that are responsible for the existence of the created world, one good and the other evil. Leibniz’s discussion is primarily a response to arguments in favour of cosmic dualism that he finds in the writings of Pierre Bayle. However, in addition, he presents a genealogical argument that appears to be intended to provide reasons to reject the view. The paper also contains a critical discussion of Leibniz’s case, and finishes by drawing attention to some issues which arise that are worthy of further consideration.

1. Introduction

In this paper, I will be concerned with the passages in Leibniz’s Theodicy where he rejects a version of cosmic dualism that he finds in the writings of Pierre Bayle—i.e., the view that there are two distinct fundamental principles that are responsible for the existence of the created world, one good and the other evil. The paper will take the following form: I will begin, in Section 2, with an outline of cosmic dualism; in Section 3, I will present Bayle’s case for the doctrine and Leibniz’s response; in Section 4, I will outline additional, genealogical considerations that Leibniz offers to supplement his critique of Bayle; and, finally, in Section 5, I will offer a brief, critical discussion of Leibniz’s case against cosmic dualism.

It is important to note from the outset that the dialectical situation in which Leibniz’s discussion of cosmic dualism is embedded is complex. Bayle believes that there is a compelling a priori argument for the existence of the perfect God of traditional Christian monotheism.1 However, he also argues that cosmic dualism is the most reasonable position to adopt given the existence of evil. Bayle’s resolution of this situation is described by Leibniz as follows in the Preface to the Theodicy: ‘[Bayle] wishes to infer that our Reason is confounded and cannot meet her own objections, and that one should disregard them and hold fast the revealed dogmas, which teach us the existence of one God altogether good, altogether powerful and altogether wise’ (Theodicy Preface/Leibniz 1985, p. 58). Indeed, according to Leibniz, Bayle does not just insist that we should prioritise revealed truth over the deliverances of reason and reject cosmic dualism, he ‘seeks therein a usefulness … [namely] that of displaying the power of faith by showing that the truths it teaches cannot sustain the attacks of reason and that it nevertheless holds its own in the heart of the faithful’ (Theodicy Preface/Leibniz 1985, p. 96).

Leibniz could not be further from this position himself. He observes that ‘since reason is a gift of God, even as faith is, contention between them would cause God to contend against God; and if the objections of reason against any article of faith are insoluble, then it must be said that this alleged article will be false and not revealed’ (Theodicy Preface/Leibniz 1985, pp. 96–97). Given that he agrees with Bayle that God’s existence can be proved, Leibniz insists that any objection must be mistaken. Nonetheless, he thinks there are pressing reasons to rebuff Bayle’s arguments against monotheism. For, given the popularity of Bayle’s writing, Leibniz is concerned that ‘many readers, convinced of the irrefutable nature of his objections and believing them to be at least as strong as the proofs for the truth of religion, would draw dangerous conclusions’ (Theodicy Preface/Leibniz 1985, pp. 96–97).

2. Cosmic Dualism

Dualistic cosmologies can be found in a number of religious traditions. The common denominator is the view that reality includes two fundamental principles. Often these principles are regarded as divinities, where the relationship between the two is antagonistic. For example, in ancient Egypt, the sun god Rē is accompanied by Apophis, who is a serpent and bringer of darkness. However, cosmic dualism is most readily associated with Zoroastrianism, which takes its name from Zoroaster who, on the assumption that he is more than a legendary figure, is taken to have lived not later than 600 BCE.2 In Zoroastrianism, cosmic dualism takes the form of the co-existence of the god Ahura Mazdā, or Ohrmazd, who is the source of all that is good and truthful, and Mainyu, or Ahriman, who is the source of evil and falsehood.3 Reality is then conceived as involving a constant battle between these principles which comes to a conclusion with good triumphing over evil at the end of history. However, Bayle’s discussion of cosmic dualism focuses on other incarnations of cosmic dualism, namely those found among the ‘Manicheans’ and ‘Paulicians’.4

Manichaeism was founded by Mani (216–274 CE) who was from Parthia, which was part of what is now Iran. It was a major world religion for around four hundred years, and, although similar to Zoroastrianism in its foundations, it had a much more elaborate cosmological system. Far less is known about the Paulicians, or ‘followers of Paul’, a sect that arose in Armenia in the seventh century CE that self-identified as Christian. However, they have traditionally been interpreted as committed to a version of cosmic dualism and associated with Marcionism and Manichaeism;5 and Leibniz’s critical remarks are predicated on the fact that he and Bayle regarded them as Manicheans.6

Leibniz engages in an extended discussion of what he takes to be Bayle’s support for cosmic dualism in Sections 144–157 of Part 2 of the Theodicy. He focuses primarily on the article ‘Manicheans’, but also turns briefly to ‘Paulicians’. According to Leibniz, Bayle believes that the dualists of the past—Zoroaster, Marcion, and Manes—had all supported their view with ‘pitiful arguments’ (Theodicy Pt 2, sec. 144/Leibniz 1985, p. 213) but had not ‘known how to apply their principal instrument, which was the difficulty over the origin of evil’ (Theodicy Pt 2, sec. 144/Leibniz 1985, pp. 213–14).7 And Bayle is said to believe that cosmic dualists ‘triumph in their turn when one comes to the a posteriori reasons, which are taken from the existence of evil’ (Theodicy Pt 2, sec. 144/Leibniz 1985, pp. 213–14). I shall group the challenges that Bayle poses under four headings, along with Leibniz’s replies. However, Leibniz’s initial critique of cosmic dualism, which I shall discuss in Section 4, does not involve engagement with Bayle at all.

3. Leibniz’s Critique of Bayle on Cosmic Dualism

Leibniz’s critique of Bayle on cosmic dualism can be productively understood as consisting of responses to four objections, although the first of these consists of two parts: (1) that monotheism cannot explain evil; (2) that cosmic dualism can explain evil; (3) that cosmic dualism is required to account for free will; and (4) that monotheism is in fact a disguised cosmic dualism.

3.1. Monotheism Cannot Explain Why Evil Exists

The first objection that Leibniz attributes to Bayle is based on the following passage: ‘Every theory has need of two things in order to be considered a good one: first its ideas must be distinct; and second it must account for experience. It is necessary to see if the phenomena of nature can be easily explained by the hypothesis of a single principle’ (Bayle 1702 II, 2025/Bayle 1991, p. 145).8 Although it is far from clear that Bayle had intended it in quite this way, Leibniz interprets these claims as elliptical for a generic challenge, namely that the monotheist cannot explain why there is evil tout court. And he leaps in, insisting ‘I have explained it sufficiently by showing that there are cases where some disorder in the part is necessary for producing the greatest order in the whole’ (Theodicy Pt 2, sec. 145/Leibniz 1985, p. 214). Leibniz recognises that Bayle is unlikely to be satisfied with such a response, and that he ‘wishes for a detailed exposition of how evil is connected with the best possible scheme for the universe’ (Theodicy Pt 2, sec. 145/Leibniz 1985, p. 214). Bayle is not just seeking ‘an explanation’ but ‘a complete explanation of the phenomena’ or ‘comprehension of the issue’ (Theodicy, Pt 2, sec. 145/Leibniz 1985, p. 214), to which Leibniz responds ‘I do not undertake to give it; nor am I bound to do so, for there is no obligation to do that which is impossible for us in our existing state’ (Theodicy Pt 2, sec. 145/Leibniz 1985, p. 214).

As Leibniz observes in the Preliminary Dissertation on Faith and Reason, ‘to “comprehend” something … one must have all the ideas of everything that goes to make it up, and all these ideas must be clear, distinct, adequate’ (Theodicy, Preliminary Dissertation, sec. 73/Leibniz 1985, p. 114). However, even in the case of explaining everyday phenomena, he does not think this can be a requirement, since ‘There are a thousand objects in Nature in which we understand something, but which we do not therefore necessarily comprehend’ (Theodicy, Preliminary Dissertation, sec. 73/Leibniz 1985, p. 114). For example, ‘in natural philosophy we explain up to a certain point sundry perceptible qualities, but in an imperfect manner, for we do not comprehend them’ (Theodicy, Preliminary Dissertation, sec. 5/Leibniz 1985, p. 76).9 Furthermore, Leibniz does not regard the case with which Bayle is concerned as one which is as intelligible as phenomena such as colour and sound, since it concerns the reasons that underlay creation, which involved ‘choice of the order of the universe, which depends upon universal harmony, and upon the clear knowledge of an infinity of things at once’ (Theodicy, Preliminary Dissertation, sec. 23/Leibniz 1985, p. 88).

As we have seen, when it comes to the issue of the relation between God and the existence of evil, Leibniz claims he has ‘explained it sufficiently by showing that there are cases where some disorder in the part is necessary for producing the greatest order in the whole’ (Theodicy Pt 2, sec. 145/Leibniz 1985, p. 214). He does not explicitly indicate what he has in mind at this point. However, examples of this can be found at the beginning of Part 1 of the Theodicy. Here, he suggests that ‘Often indeed two evils have made one great good’ (Theodicy Pt 1, sec. 10/Leibniz 1985, p. 129), and this is accompanied by a quotation from an epigram by Decimius Magnus Ausonius (c.310–c.395CE): ‘And if the fates desire it, they will heal with two poisons’ (Theodicy Pt 1, sec. 10/Leibniz 1985, p. 129)10 as well as a reference to the Exultet, which is part of the Catholic liturgy from the Easter Vigil mass, where the cantor sings ‘Oh truly necessary sin of Adam, blotted out by the death of Christ! O Blessed fault, which won so great a redeemer’ (Theodicy Pt 1, sec. 10/Leibniz 1985, p. 129).11 But Leibniz also lists some more mundane examples. Thus, he notes that ‘a general makes sometimes a fortunate mistake which brings about the winning of a great battle’ (ibid.), and in Part 1, Section 12, he observes

A little acid, sharpness or bitterness is often more pleasing than sugar; shadows enhance colours; and even a dissonance in the right place gives relief to harmony. We wish to be terrified by rope-dancers on the point of falling and we wish that tragedies shall well-nigh cause us to weep. Do men relish health enough, or thank God enough for it, without having ever been sick? And is it not most often necessary that a little evil render the good more discernible, that is to say, greater?

(Theodicy Pt 1, sec. 12/Leibniz 1985, p. 130)

Any such explanation will be ‘imperfect’ and will not provide ‘comprehension of the issue’, but we should not expect or require any more than this, given that we are finite beings in an infinitely complex created world.

However, Leibniz does not restrict himself to the challenge posed by the existence of evil in general. As he points out, although Bayle is willing to allow that ‘the heavens and all the rest of the universe preach the glory, the power, the unity of God’, he claims that this is countered by the fact that ‘Man alone, that masterpiece of his Creator among the visible things … furnishes some very great objections’ (Bayle 1702 II, 2025/Bayle 1991, p. 146).12 More precisely, Leibniz notes that Bayle then claims ‘that man is wicked and miserable; that there are everywhere prisons and hospitals; that history is simply a collection of the crimes and calamities of the human race’ (Theodicy Pt 2, sec. 148/Leibniz 1985, p. 216).

In fact, this is a simplification of the original text from the Dictionary. After his initial claim that ‘man is wicked and miserable’, Bayle adds ‘Everybody is aware of this from what goes on in himself, and from the commerce he is obliged to carry on with his neighbour’ (Bayle 1702 II, 2025/Bayle 1991, p. 146). Indeed, he suggests that being ‘alive for five or six years’ will leave one ‘completely convinced of these two truths’ since one will have ‘done and suffered malicious acts… felt chagrin and pain, [and] sulked many times, etc.’ (Bayle 1702 II, 2025/Bayle 1991, p. 146). And, in addition to ‘prisons and hospitals’, Bayle points to ‘gallows and beggars [and] ruins of a flourishing city’ and cities like Troy whose ruins cannot be found (Bayle 1702 II, 2025/Bayle 1991, p. 147). However, Leibniz does note slightly later that Bayle asks, rhetorically:

If man is the work of a single supremely good, supremely holy, supremely powerful principle, is it possible that he can he be exposed to illnesses, to cold, to heat, to hunger, to thirst, to pain, to vexation? Is it possible that he should have so many bad inclinations and commit so many crimes?13 Is it possible that the supreme goodness would produce an unhappy creature? Would not the supreme power, joined to an infinite goodness, pour down blessings upon its work, and defend it from everything that might annoy or trouble it?

(Bayle 1702 II, 2025/Bayle 1991, p. 147)

Leibniz’s response begins with the suggestion that Bayle’s claims about the parts of the universe other than mankind should have been enough to lead to the conclusion that God alone exists. For it contains objects which, when taken on their own, are found to be ‘so perfect that we must admire the artifice and beauty in [them]’ and ‘recognize therein the wonderful artifice of the author’ (Theodicy Pt 2, sec. 146/Leibniz 1985, p. 215). Here, Leibniz has in mind such things as ‘our planetary system … each plant, each animal [and] each man’ (Theodicy Pt 2, sec. 146/Leibniz 1985, p. 215). By contrast, Leibniz suggests that ‘when we do not see an entire work … it is no wonder if the good order is not apparent there’(Theodicy Pt 2, sec. 146/Leibniz 1985, p. 215), and he takes this to be the case when we consider ‘the City of God’ or ‘the republic of Spirits’ (Theodicy Pt 2, sec. 146/Leibniz 1985, p. 215). For, although we are aware to some extent of the nature of the humans that make up the City of God, this ‘is only a fragment, only a small portion’ and ‘we know too little, to be able to observe the wonderful order therein’ (Theodicy Pt 2, sec. 146/Leibniz 1985, p. 215). Furthermore, Leibniz suggests that the fact that everything else is harmonious ‘allows of a strong presumption that it would exist also in the government of men, and generally in that of Spirits, if the whole were known to us’ (Theodicy Pt 2, sec. 146/Leibniz 1985, p. 215). And he appeals to Socrates, saying we should judge God’s works ‘as wisely’ as the former did when he said of Heraclitus: ‘What I have understood thereof pleases me; I think that the rest would please me no less if I understood it’ (Theodicy Pt 2, sec. 146/Leibniz 1985, p. 215).

However, Leibniz also explains at least some of the phenomena to which Bayle is referring, namely that ‘in giving [man] intelligence, [God] has presented him with an image of the Divinity’ and that ‘free will plays its game’ (Theodicy Pt 2, sec. 147/Leibniz 1985, p. 215). As a result, whilst man ‘sometimes performs wonders … he also commits great errors’(Theodicy Pt 2, sec. 147/Leibniz 1985, p. 215). But Leibniz does not rely on this claim to offer a version of what is usually referred to as ‘the free will defence’, namely the view that God’s desire to produce free creatures explains the creation of a world in which the evil resulting from human action exists. Instead, he claims that God is ‘like a father or tutor, training or chastising children’ and on other occasions ‘like a just judge, punishing those who forsake him’ (Theodicy Pt 2, sec. 147/Leibniz 1985, p. 216). And, although this means that ‘Man finds himself the worse for this, in proportion to his fault’, God ‘turns all the errors of these little worlds to the greater adornment of his great world’ (Theodicy Pt 2, sec. 147/Leibniz 1985, p. 216). Here, Leibniz draws an analogy with anamorphic paintings, and observes that ‘the apparent deformities of our little worlds combine to become beauties in the great world, and have nothing in them which is opposed to the oneness of an infinitely perfect universal principle’ (Theodicy Pt 2, sec. 147/Leibniz 1985, p. 216). Thus, far from detracting from God’s goodness, ‘on the contrary, they increase our wonder at the wisdom of him who makes evil serve the greater good’ (Theodicy Pt 2, sec. 147/Leibniz 1985, p. 216).

Finally, Leibniz turns to the particular examples that he quotes from Bayle which he regards as containing ‘exaggeration’ (Theodicy Pt 2, sec. 148/Leibniz 1985, p. 216). Indeed, he claims that rather than the misery that Bayle claims, we find that ‘there is incomparably more good than evil in the life of men, as there are incomparably more houses than prisons’ and he suggests that rather than thinking of men as ‘wicked’ it should be noted that ‘with regard to virtue and vice, a certain mediocrity prevails’ (Theodicy Pt 2, sec. 148/Leibniz 1985, p. 216). Furthermore, he claims that Bayle’s reliance on historical accounts should be tempered by the recognition of ‘a great fault in historians’, namely ‘that they keep their mind on the evil more than on the good’ (Theodicy Pt 2, sec. 148/Leibniz 1985, pp. 216–17).

3.2. Cosmic Dualism Provides an Explanation of the Existence of Evil

Leibniz presents Bayle’s second objection in Theodicy Part 2, Section 152. Here, he draws attention to the fact that Bayle claims not merely that monotheism cannot explain the phenomena of evil, but that cosmic dualism can. He points to the place in ‘Manicheans’ where Bayle considers an imaginary dialogue between the Eleatic philosopher Melissus (5th century BCE) and Zoroaster, who represent monotheism and cosmic dualism, respectively. Zoroaster observes:

I surpass you in the explanation of phenomena and a posteriori reasons. And since the chief characteristic of a good system is its being capable of accounting for experience, and since the mere incapacity of accounting for it is a proof that a hypothesis is not good, however fine it appears to be in other respects, you must grant that I hit the nail on the head by admitting two principles and you miss it by admitting only one.

(Bayle 1702 II, 2025/Bayle 1991, p. 148)14

Bayle invites us to accept that insofar as evil exists in the world it can be explained on the assumption that there was an evil creative force. Whilst the argument is not spelled out, presumably Bayle’s thought is that it makes sense to think that an evil creator would express their evil nature through the production of evil.

In reply Leibniz claims that the hypothesis attributed to Zoroaster is ‘not a very good explanation’ since it involves ‘assigning it to an ad hoc principle’ (Theodicy Pt 2, sec. 152/Leibniz 1985, p. 218). To explain the phenomenon of evil by ‘a principium maleficum’ would be the same as explaining the phenomenon of cold by ‘a primum frigidum’ (Theodicy Pt 2, sec. 152/Leibniz 1985, p. 218). In both cases, Leibniz claims there would be ‘nothing so easy and nothing so dull’ and draws a comparison with those who would claim that ‘the Peripatetics surpass the new mathematicians in the explanation of the phenomena of the stars, by giving them ad hoc intelligences to guide them’ and ‘Some old Galenist’ whose medical explanations appeal to ‘a chylific, a chymific and a sanguific [faculty]’ (Theodicy Pt 2, sec. 152/Leibniz 1985, pp. 218–19). However, Leibniz does not merely accuse the cosmic dualist of introducing a principle which does no explanatory work, he returns to the issue of how evil can be explained by the monotheist. But, it is important to remember that where Leibniz talks of offering an explanation, this should not, as we have seen, be understood to imply that the account is supposed to render the existence of particular evil phenomena comprehensible.

Leibniz claims that ‘Evil itself comes only from privation [and] the positive enters therein only by concomitance’ (Theodicy Pt 2, sec. 153/Leibniz 1985, p. 219) and reprises the analogy with cold, observing ‘Water in freezing is capable of breaking a gun-barrel wherein it is confined; and yet cold is a certain privation of force’ which comes from ‘the diminution of a movement which separates the particles of fluids’ (Theodicy Pt 2, sec. 153/Leibniz 1985, p. 219). Leibniz then offers a physical explanation of how this occurs in terms of the way in which trapped air particles are able to become more active. In the case of evil, the reader is told that Leibniz has ‘already shown how privation is enough to cause error and malice’ (Theodicy Pt 2, sec. 153/Leibniz 1985, pp. 219–20). Whilst he does not indicate what he is referring to explicitly, Theodicy Part 1, Sections 32–33 contain a discussion of the view that ‘not only ignorance but also error and malice consist formally in a certain kind of privation’ (Theodicy Pt 1, sec. 32/Leibniz 1985, p. 142). Here, Leibniz begins with an account of what the activity of a creature consists in, namely that it is ‘a modification … flowing naturally from it and containing a variation not only in the perfections that God has communicated to the creature, but also in the limitations that the creature, being what it is, brings with it’ (Theodicy Pt 1, sec. 32/Leibniz 1985, p. 142). These limitations entail that what is brought about by a creature involves a lack or privation of perfection. One might at this point wonder why it is that creatures must be limited. But Leibniz has an answer. Given his commitment to the identity of indiscernibles, God could not have made anything distinct from himself unless it was imperfect. As he puts it in Theodicy Part 1, Section 31 ‘God could not give the creature all without making of it a God; therefore, there must needs be different degrees in the perfection of things, and limitations also of every kind’ (Theodicy Pt 1, sec. 31/Leibniz 1985, p. 142).

As an example of error, Leibniz turns to the case of ‘a tower which from a distance appears round although it is square’ (Theodicy Pt 1, sec. 32/Leibniz 1985, p. 142). Here the error is a ‘false judgment’, which arises from two things: (1) ‘the thought that the tower is what it appears to be’; and (2) ‘dwell[ing] on this thought’ (Theodicy Pt 1, sec. 32/Leibniz 1985, p. 142). Leibniz notes that were we to ‘pursue the examination’ then the error would not be sustained. Where this does not occur, it is because we ‘abide in a certain place, or not to go further, not to espy some landmark’; and these ‘are privations’ (Theodicy Pt 1, sec. 32/Leibniz 1985, p. 142). In the case of ‘malice or ill will’, Leibniz claims that ‘The will tends towards good in general, it must strive after the perfection that befits us, and the supreme perfection is in God’ (Theodicy Pt 1, sec. 33/Leibniz 1985, p. 142). However, there are cases in which ‘one is limited to the pleasures of the senses, or to other pleasures to the detriment of greater good, as of health, of virtue, of union with God, of felicity’; and here ‘the privation’ is the lack ‘of a further aspiration’ (Theodicy Pt 1, sec. 33/Leibniz 1985, p. 142).

3.3. Free Will Requires Cosmic Dualism

For the third objection, Leibniz turns his attention from the article ‘Manicheans’ to ‘Paulicians’, observing that Bayle attributes to the Paulicians the argument that ‘free will must come from two principles, to the end that it may have power to turn towards good and towards evil’ (Theodicy Pt 1, sec. 154/Leibniz 1985, p. 220). Leibniz suggests that this is to be found on p. 2323 of the Dictionary. However, free will is not discussed at this point and it seems more likely that he is thinking of the following passage from p. 2326:

It is inconceivable that the first man could have received a faculty for doing wrong from a good principle. This faculty is vicious and everything that can produce bad is bad, since evil can only arise from a bad cause; and thus the free will of Adam is the result of the action of two contrary principles; insofar as it was able to move in the right way it depended on the good principle; but insofar as it was able to embrace evil, it depended upon the bad principle.

(Bayle 1702 III, 2326/Bayle 1991, pp. 179–80)

Leibniz interprets Bayle’s claim that the will can produce good and evil, where these are considered to be distinct kinds of effects, as implying that the will is ‘simple in itself’ (Theodicy Pt 1, sec. 154/Leibniz 1985, p. 220). In other words, he takes it that Bayle conceives of the will as neither inclined to good nor evil intrinsically. As Leibniz notes, he does not endorse this view himself, but rather the view that ‘free will tends towards good’ (Theodicy Pt 1, sec. 154/Leibniz 1985, p. 220), and his response is to claim that Bayle’s inference is ‘non-conclusive’ (Theodicy Pt 1, sec. 154/Leibniz 1985, p. 220). Apparently relying on the assumption that like causes like, which is invoked by Bayle, Leibniz claims the will as Bayle conceives it ‘should rather have come from a neutral principle’ (Theodicy Pt 1, sec. 154/Leibniz 1985, p. 220).

3.4. Monotheism Is Really Dualistic

Leibniz attributes one final objection to Bayle, based again on ‘Paulicians’, namely that ‘the orthodox seem to admit two first principles, in making the Devil the originator of sin’ (Theodicy Pt 1, sec. 156/Leibniz 1985, p. 221).15 His response is to acknowledge that ‘the Devil is the author of sin’ but not ‘the origin of sin’ (Theodicy Pt 1, sec. 156/Leibniz 1985, p. 221). Leibniz elaborates on the second of these claims in now familiar terms claiming that ‘its source is in the original imperfection of creatures: that renders them capable of sinning, and there are circumstances in the sequence of things which cause this power to evince itself in action’ (Theodicy Pt 1, sec. 156/Leibniz 1985, p. 221). He also gives a brief characterisation of what he takes to be the orthodox conception of Satan, namely that he is like other ‘devils’ who were ‘angels … before their fall’ (Theodicy Pt 1, sec. 157/Leibniz 1985, p. 221).

Although Leibniz is providing a plausible account of orthodox revealed Christian theology here, as a response to the challenge it is somewhat misleading. It is true that Bayle claims that insofar as ‘he is the one who made Eve and her husband fall from the state of innocence’ the Devil ‘is the first principle of evil’ (Bayle 1702 III, 2330/Bayle 1991, p. 186). And it is presumably in this sense that Leibniz wishes to say that Satan is ‘the author of sin’ (Theodicy Pt 1, sec. 156/Leibniz 1985, p. 221). However, Bayle adds that ‘since [the Devil] is neither eternal nor uncreated, he is not the basic wicked principle in the Manichean sense’ (Bayle 1702 III, 2330/Bayle 1991, p. 186). Thus, Leibniz presents and defeats a straw man. Bayle’s real challenge here derives from the fact that God created the Devil in the first place.16

4. Leibniz’s Genealogy of Cosmic Dualism

Whilst Leibniz’s discussion of cosmic dualism emerged from his reading of Bayle’s Dictionary, only around a third of the space that he devotes to criticizing the doctrine consists of a direct engagement with Bayle. Leibniz alludes to this briefly in the Preface, where he observes that ‘The ancient error of the two principles, which the Orientals distinguished by the names Oromasdes and Arimanius, caused me to explain a conjecture on the primitive history of peoples’ (Theodicy Preface/Leibniz 1985, p. 71) and goes on to present a very truncated version of an argument that recurs in Theodicy Part 2, Sections 136–43 (Leibniz 1985, pp. 208–13).

In Section 136 Leibniz places the ancestry of Manichaeism in Zoroastrianism and claims that ‘Plutarch, in his treatise On Isis and Osiris, knows of no writer more ancient than Zoroaster the magician, as he calls him, that is likely to have taught the two principles’ (Theodicy Pt 2, sec. 137/Leibniz 1985, p. 208).17 He then observes that ‘The ancient Greeks and the modern Orientals agree in saying that Zoroaster called the good God Oromazes, or rather Oromasdes, and the evil God Arimani’ (Theodicy Pt 2, sec. 138/Leibniz 1985, p. 210). Next, Leibniz makes the following observation:

When I pondered on the fact that great princes of Upper Asia had the name of Hormisdas and that Irminius or Herminius was the name of a god or ancient hero of the Scythian Celts, that is, of the Germani, it occurred to me that this Arimanius or Irminius might have been a great conqueror of very ancient time coming from the west, just as Genghis Khan and Tamburlaine were later, coming from the east.

(Theodicy Pt 2, sec. 138/Leibniz 1985, p. 210)

To accompany this account of the origin of the name ‘Arimani’, Leibniz suggests that ‘Arimanius would therefore have come from the north-west … to raid the dominions of one Ormisdas, a great king in Upper Asia’ (Theodicy Pt 2, sec. 138/Leibniz 1985, p. 210). And he supports this suggestion by appealing to Herodotus’ account of the much later invasion by the Scythians of the Medes, an area of ancient Iran, in the 7th century BCE (Theodicy Pt 2, sec. 138/Leibniz 1985, p. 210).18

Leibniz then offers the hypothesis that ‘The monarch governing civilized peoples, and working to defend them against the barbarians, would have gone down to posterity, amongst the same peoples, as the good god’ and at the same time ‘the chief of these devastators will have become the symbol of the evil principle’ (Theodicy Pt 2, sec. 138/Leibniz 1985, p. 210). The final component in Leibniz’s account of the origin of cosmic dualism in Zoroastrianism is his suggestion that ‘these two princes contended for long, but that neither of them was victorious. Thus they both held their own, just as the two principles shared the empire of the world according to the hypothesis attributed to Zoroaster’ (Theodicy Pt 2, sec. 138/Leibniz 1985, p. 210).19

Up to this point, Leibniz has presented a speculative genealogy: Two princes fought for a long period of time with no resolution, one of whom, an invader, was mythologised as ‘the evil God Arimani’, the other of whom, the defender of the local population, was mythologised as ‘the good God Oromazes, or rather Oromasdes’ (Theodicy Pt 2, sec. 138/Leibniz 1985, p. 210). However, he is not happy with a merely speculative account, and he observes that ‘It remains to be proved that an ancient god or hero of the Germani was called Herman, Arimanius or Irminius’ (Theodicy Pt 2, sec. 139/Leibniz 1985, p. 210). He finds this supposed proof in Chapter 2 of Tacitus’ Germania, which was written around 98 CE, in which it is claimed that there were ‘three tribes which composed Germania, the Ingaevones, the Istaevones, and the Herminones or Hermiones’ (Theodicy Pt 2, sec. 139/Leibniz 1985, p. 210). Whilst Leibniz indicates that he is not concerned with the historical accuracy of Tactitus’ account, it is important to him because he takes it as evidence that Tacitus wanted to ‘indicate that there was a hero named Herminius, from whom he was told the Herminones were named’ (Theodicy Pt 2, sec. 139/Leibniz 1985, p. 210).20 According to Leibniz this name, which means ‘soldiers’ (Theodicy Pt 2, sec. 139/Leibniz 1985, p. 210), then provided the basis for the fact that ‘all the Teutonic peoples were named Hermanni or Germani’ (Theodicy Pt 2, sec. 140/Leibniz 1985, p. 210), which he regards as unsurprising given that it is ‘very usual for one part of a nation to give the name to the whole’ (Theodicy Pt 2, sec. 140/Leibniz 1985, p. 210).

But Leibniz is not content to rely on Tacitus, he also notes that ‘Charlemagne found and destroyed near the Weser the column called Irminsäule, erected in honour of this god’ (Theodicy Pt 2, sec. 141/Leibniz 1985, p. 211). The entry for the year 772 CE in the Annales Regni Francorum states that ’He [i.e., Charlemagne] conquered the Eresburg and found the place which is called Ermensul, and set these places on fire’ (Pertz and Waitz 1826–1876, vol. 6, pp. 33–34),21 and a similar view is found in the early 20th-century historian Charles Henry Robinson’s discussion of the destruction of Irminsäule. Robinson claims that it was a ‘celebrated Saxon idol, near Eresburg on the R. Drimel’ which ‘was a lofty pillar, or the trunk of a gigantic tree, which had been consecrated by immemorial reverence.’ And he suggests that ‘Irmin appears to have been the name of a national god or demi-god’ (Robinson 1917, p. 389).

Robinson in turn supports his view by appealing to Sharon Turner’s The History of the Anglo Saxons, which discusses the nature and role of the Irminsäule at some length by drawing on Heinrich Meibom’s (1555–1625) Opuscula historica ad res Germanicus spectantia, which was edited and published in 1660 by his grandson, who was also called Heinrich Meibom (1638–1700). According to Turner, the question of whom the Irminsäule was dedicated to is ‘a question full of uncertainty’ (Turner 1799–1805, p. 544). However, Turner reports that Meibom’s own view was that the similarity of the word ‘Irminsäule’ to ‘Ares’ and ‘Hermes’ suggests that Mars or Mercury might have been the relevant god. Leibniz does not cite his source for his discussion of the ‘Irminsäule’. But it seems likely that he is drawing at least in part on Meibom, given that in Section 142 he adds that: ‘It is not beyond belief that the Hermes (that is, Mercury) of the Greeks is the same Herminius or Arimanius’ (Theodicy Pt 2, sec. 142/Leibniz 1985, p. 211). However, to the extent that he embraces this idea, Leibniz again wishes to identify ‘Herminius or Arimanius’ with ‘an inventor or promoter of the arts and of a slightly more civilized life among his own people and in the countries where he held supremacy, while amongst his enemies he was looked upon as the author of confusion’ (Theodicy Pt 2, sec. 142/Leibniz 1985, p. 211).

In presenting this account, Leibniz is offering an example of what is traditionally referred to as ‘Euhemerism’ after the Greek mythographer Euhemerus (late 4th C CE), or what Thomas Bulfinch termed the ‘historical theory of mythology’ (Bulfinch 1913, p. 301). According to this view, myths are explained as originating from historical events and exaggerated over time. In this case, Leibniz offers what he claims is a historically grounded account of the emergence of cosmic dualism in Zoroastrianism as the mythical reification of two protagonists in an ancient conflict over territory. Perhaps surprisingly, however, Leibniz does not say anything explicitly about why he has provided this account. I shall return to the question of what the purpose of the genealogy might have been in Section 5, though given Leibniz’s own silence on the issue what I will say will be rather speculative and preliminary.

5. How Successful Is Leibniz’s Refutation of Cosmic Dualism?

I now want to turn to a critical discussion of the arguments as I have presented them above. I will pass over the third of Bayle’s objections quickly, namely the argument that free will requires dualism. Since it is hard to know just what Bayle is claiming, it is hard to know how to assess the adequacy of Leibniz’s response. However, I take it that Leibniz is trying to carve out a causal role for the will, and, given that Bayle suggests bad outcomes are caused by a bad cause and good outcomes by a good cause, he assumes that these must be external to the will on pain of it having a contradictory nature. Thus, to the extent that Bayle relies on a like-causes-like principle, Leibniz’s employment of it in his response seems permissible dialectically.

Central to Leibniz’s response to Bayle’s claims that traditional monotheism is unable to explain evil, and in particular the evils that humans undergo or are responsible for producing, is the distinction that he draws between ‘explanation’ and ‘comprehension’. Leibniz suggests that comprehension is something that will only be present where one has a ‘complete explanation’, something which requires that the explanation is couched in terms of ideas that are ‘clear, distinct, [and] adequate’ (Theodicy Preliminary Dissertation, sec. 73/Leibniz 1985, p. 114). The typology of ideas that Leibniz employs here is one that he appeals to throughout his corpus, and which made its appearance publicly in the Acta Eruditorum of 1684 in his article ‘Meditations on Knowledge, Truth, and Ideas’ (Leibniz 1989, pp. 23–27). In essence, to satisfy the criteria that Leibniz mentions above, the conception of something would need to reveal all of its complexity transparently. As we have seen, Leibniz himself was committed to the claim that the content of God’s choice was infinitely complex, and it seems likely that Bayle would have agreed with this. However, even without a commitment to the infinite complexity of the facts determining why God chose the actual world, it is hard to see that anyone would object to Leibniz’s claim that the actual world is incomprehensible for human beings in his sense.

Thus, to the extent that Leibniz is correct in thinking that Bayle would only really be satisfied if Leibniz were to present an account that made the existence of evil comprehensible, his failure to provide such an account appears defensible. If explanations are supposed to render their explananda comprehensible, then failure to explain why evil exists is not a genuine failure. However, Leibniz is not content with this response, insisting that he is able to provide ‘imperfect explanations’—i.e., the only kinds of explanations that we ever really have.

Leibniz’s basic explanation for the existence of evils is that they are ‘necessary for producing the greatest order in the whole’ (Theodicy Pt 2, sec. 145/Leibniz 1985, p. 214), something that we have seen on a more local scale, such as when a mistake can lead to something fortuitous such as the winning of a battle. When it comes to the particular evils that Bayle’s observations on mankind bring to light, Leibniz does not offer additional explanations. Rather he insists that closer examination will reveal that the good outweighs the evil, and that we should hold the partial and distorting narratives provided by historians responsible for our thinking otherwise.22 Furthermore, Leibniz insists that we should attend to the fact that we are only aware of a small portion of the universe and not extrapolate, especially given that observations of things other than man allow a ‘strong presumption’ that were we to have access to all the phenomena involving humans and other ‘spirits’ then we would find order there as well. And he also suggests that insofar as any given individual suffers, God will bring something greater out of it for that person.23

Although Leibniz’s response ultimately rests on his claim that the evils that in fact exist exist because they are components of the best choice that God could have made, it is hard to deny that he has offered an explanation for them. Furthermore, Bayle accepts that there is a compelling argument for traditional monotheism, and it is unlikely that he would resist Leibniz’s additional claim that traditional monotheism provides the grounds for a demonstrative proof of the claim that whatever God created was the best option available. Thus, given our lack of comprehension of that which makes the actual world better than all the other possible worlds, it seems that there is no equally compelling reason to conclude that the existence of evil is incompatible with the claim that the actual world was created by God. And we need to remember that this is the position that Bayle wishes to occupy: that there are equally good reasons to think that God exists and that God does not.

The third of Bayle’s objections, that cosmic dualism can explain evil, is a companion to the previous one. Leibniz’s response is to claim that this is to invoke an ‘ad hoc principle’ (Theodicy Pt 2, sec. 152/Leibniz 1985, p. 218), as if one were to explain the phenomenon of coldness by invoking something that one could refer to only as that which produces cold. The strategy employed here is more familiar from Leibniz’s criticism of Newton for introducing an ‘occult quality’ when he employed gravity to explain the motion of the planets. Thus, Leibniz tells Clarke,

This overthrows attractions, properly so called, and other operations inexplicable by the natural powers of creatures; those who assert these kinds of operations must suppose them to be affected by miracles, or else they have recourse to absurdities, that is, to the occult qualities of the schools, which some men begin to revive under the specious name of forces, but which bring us back again into the kingdom of darkness. This is inventa fruge, glandibus vesci [To feed on acorns when corn has been discovered].

(Letter 5, sec. 113/Leibniz 1956, p. 92)24

Given the way in which Bayle presents his position, Leibniz’s objection clearly has some prima facie force. Whatever limits there may be to the explanatory power that Leibniz claims for the monotheist’s position, there is a characterisation of the divine nature that ensures that his argument for the existence of the best possible world, and the accounts of why evil exists have content. Bayle offers nothing. Here, Leibniz’s position finds echoes in contemporary discussions. For example, in The Science of Evil: On Empathy and the Origins of Cruelty, Simon Baron-Cohen offers a theory that is supposed to account for the most deplorable ways in which humans behave which is grounded in lack of empathy. At the beginning of the book, Baron-Cohen contrasts his account with the ‘standard explanation’ which takes the following form: ‘Why did the murderer kill an innocent child? Because he was evil. Why did this terrorist become a suicide bomber? Because she was evil’ (Baron-Cohen 2011, p. 5). Baron-Cohen’s concern is that in such cases evil itself ‘is treated as incomprehensible … [and] no explanation at all’ (Baron-Cohen 2011, p. 6).25

However, I want to suggest that a counter to this line of thinking can be found if we turn to the well-known discussion of the problem of evil by Dostoevsky, using the voice of Ivan Karamazov, in The Brothers Karamazov.26 In the passage in question, Ivan discusses the problem of evil with his brother Alyosha and contemplates the torture of innocent children—focusing in particular on the case of an eight-year-old deliberately torn apart by hunting dogs in front of his mother on the command of a nobleman for a minor misdemeanour (see Dostoyevsky 1994, p. 304). Ivan grants the following:

I understand how the universe will shake when heaven and earth shall unite in a single paean of praise, and all Rebellion that lives and has lived will cry out, ‘You are just, O Lord, for your ways are revealed to us!’ When the mother embraces the murderer whose dogs tore her son apart, and all three shall cry out weeping, ‘You are just, O Lord’—that, of course, will be the summit of all knowledge, and all will be explained.

(Dostoyevsky 1994, pp. 306–7)

Here, the existence of the evil under consideration is not just presented as explicable, it is comprehended as just by all those who were involved in the initial evil. Those who have been harmed are reconciled to the one who harmed them. Thus, Ivan allows for an even better position than the one that Leibniz is offering.

However, Ivan is not content. He worries that this might come to pass and that he ‘will cry out in unison with them, “You are just, O Lord”’ and protests ‘While there’s still time I want to guard myself against this, and therefore I absolutely reject that higher harmony’ (Dostoyevsky 1994, p. 307). Ivan’s concern is that the situation described is one where the ‘tears have remained unexpiated’ (Dostoyevsky 1994, p. 307). Indeed, he goes as far as to claim that the mother ‘has no right to forgive the suffering of her mutilated child’ (Dostoyevsky 1994, p. 307). Even if we acknowledge the harmony that Leibniz suggests God had in mind when creating the best of all the imperfect worlds available, ‘the price of harmony has been set too high’ (Dostoyevsky 1994, p. 307).

Ivan’s challenge is difficult to pin down, but an earlier passage suggests one interpretation. Here, Ivan insists ‘I don’t want to understand anything. I want to stick to facts. I gave up trying to understand long ago. As soon as I feel I want to understand something I immediately have to renounce facts, whereas I have decided to stay true to facts…’ (Dostoyevsky 1994, p. 307). On the reading I am suggesting, what is appalling to Ivan is the idea that one would respond to the horrors of the kinds of evils he describes by attempting to explain them at all. Even if it brings consolation and a sense of divine justice, this is not the appropriate way to respond, since it is to allow one’s mind to turn away from what has truly happened. Indeed, this seems to be close to the sentiment that is famously found in Elie Wiesel’s response to the evils of the holocaust, namely, to use Inga Clendinnen’s words, that ‘the only possible and proper stance for the observer is one of awed incomprehension’ (Clendinnen 1999, p. 21).27

Perhaps Leibniz is correct to say that postulating an evil principle as the cause of evil does not explain evil. However, a more charitable reading of the dualist’s intent might be that talk of an unintelligible evil principle that causes evil is supposed to serve as an obstacle to our attempts to take solace in explanation; that it brings us face to face with evil as fact, and the realisation that the only genuine ‘explanation’ of such evils is that they cannot be explained if one is to do justice to them. They must be regarded as the unintelligible manifestation of pure evil.

As I noted above, Leibniz misses the mark when he responds to Bayle’s final argument, namely, that ‘the orthodox seem to admit two first principles, in making the devil the originator of sin’ (Theodicy Pt 1, sec. 156/Leibniz 1985, p. 221), given that Bayle was not equating the Devil with the evil principle, but suggesting that the commitment to cosmic dualism was implicit in the fact the God created the Devil in the first place. As such it seems to rely on two points that we have already considered, namely Bayle’s insistence that monotheism cannot explain the existence of evil whereas cosmic dualism can. But, insofar as this is the case, it seems likely that Leibniz would invoke the same responses as we have seen already.

However, this is not the only thing of interest that arises at this point in the text in connection with this issue. For, despite his defence of monotheism, Leibniz himself admits that ‘There are in truth two principles’ (Theodicy Pt 2, sec. 149/Leibniz 1985, p. 217). This is accommodated by his further insistence that ‘they are both in God, to wit, his understanding and his will’ where ‘[t]he understanding furnishes the principle of evil’ (Theodicy Pt 2, sec. 149/Leibniz 1985, p. 217). The crucial difference according to Leibniz is that it does this ‘without being evil’ (Theodicy Pt 2, sec. 149/Leibniz 1985, p. 217). The view that Leibniz tries to defend here is that the source of evil is intrinsic to the possible creatures that God conceives, rather than God himself. As he puts it in connection with human beings: ‘Man is himself the source of his evils: just as he is, he was in the divine idea’ (Theodicy Pt 2, sec. 151/Leibniz 1985, p. 218).

However, one might wonder how plausible a response this is. Leibniz accepts that there could not be divine ideas of beings that are less perfect than God unless there were something in God that constrained divine perfection, and he identifies this with the creatures themselves insofar as they have being as ideas in the divine intellect. However, an opponent might suggest that the ideas of creatures must result from a constraining of the divine perfection by a principle of limitation which opposes perfection; and that on pain of God being both perfect and opposed to perfection, this principle must reside in another being. In any event, greater clarification of Leibniz’s position seems to be required if we are to determine how adequate it is in the context of his critique of Bayle.

Finally, I want to turn to Leibniz’s genealogy of cosmic dualism. As I noted above, in trying to assess this strategy, we face a significant difficulty insofar as Leibniz himself fails to say anything explicit about why he presents the discussion. As a result, most of what I have to say at this point will be an attempt to make sense of what the purpose of Leibniz’s genealogy might be.

Before the genealogy is given, Leibniz observes that Zoroaster is generally taken to be the first person to have preached cosmic dualism and that the names that Zoroaster gives for the principles coincide with the warring princes for whose existence Leibniz presents historical evidence. It seems then that it is implicit in Leibniz’s discussion that this is the appropriate explanation of why Zoroaster was committed to cosmic dualism in which the principles bore the names Oromasdes and Arimanius. Whether in good faith or otherwise, he propagated a tradition, which at some point had falsely elevated good and bad princes to the status of gods. One might speculate that Leibniz believed that, in contrast, adherents of cosmic dualism such as the Manicheans and Paulicians assumed that cosmic dualism was true and that the divine status of Oromasdes and Arimanius had been revealed, whether to Zoroaster or to someone whose views Zoroaster preached. On this reading, Leibniz can be taken to be offering a competing explanation for Zoroaster’s commitment to the reality of the cosmic principles Oromasdes and Arimanius, which does not depend on the genuine revelation of their status.28 In doing so, he offers a version of what Amia Srinivasan has termed a ‘critical genealogy’ (Srinivasan 2019, p. 127).

Drawing on the work of Bernard Williams (Williams 2002), Srinivasan contrasts critical genealogies with ‘vindicatory genealogies’. Paradigm cases of the latter include ‘Locke and Hobbes’s accounts of the emergence of the state’, which are said to vindicate insofar as they ‘are meant to not only explain but moreover justify their authors’ favoured political arrangements’ (Srinivasan 2019, p. 129). By contrast, critical genealogies provide accounts which are seen as ‘undermining the epistemic standing of our representations—the justification of our beliefs, the aptness of our concepts, and so on’ (Srinivasan 2019, p. 128). As an example, Srinivasan presents the famous fragment from Xenophanes:

Mortals suppose that the gods are born (as they themselves are), and that they wear man’s clothing and have human voice and body. But if cattle or lions had hands, so as to paint with their hands and produce works of art as men do, they would paint their gods and give them bodies in form like their own—horses like horses, cattle like cattle.

(Xenophanes 1898, fragments 5–6)

The point that Xenophanes appears to be making is that the Greek gods are taken to be human because of a narcissistic desire on the part of humans to make gods in their own image rather than because this is an accurate portrayal based on revelation. Similarly, Leibniz seems to be suggesting that rather than a revelation, the belief in cosmic dualism that became embedded in Zoroastrianism and was picked up by the Manicheans and Paulicians resulted from an only too human desire to consolidate the power of Ormisdes by portraying him as a good god whilst vilifying his rival Arimanius as an evil one.

As Srinivasan notes, when presented with such accounts we feel as though the genealogy has some ‘negative bearing’ on the theologies in question, but must recognise that this alone does not entail that they are false (Srinivasan 2019, p. 128). To do so would be to commit what Ernest Nagel and Morris Cohen called a ‘genetic fallacy’, namely the error arising from ‘the supposition that an actual history of any science, art, or social institution can take the place of a logical analysis of its structure’ (Cohen and Nagel 1934, pp. 389–90). And relatedly, Hans Reichenbach insisted that it was important to distinguish the ‘context of discovery’ from the ‘context of justification’ (Reichenbach 1938, pp. 6–7) of any given view, and to recognise that the epistemic status of a view is independent of its provenance. With this in mind, we might worry that even if Leibniz has provided a plausible historical account of the genesis of Zoroaster’s commitment to his version of cosmic dualism, this does not suffice for concluding that Zoroaster’s views were incorrect.

At this point, I think we need to return to the dialectical situation within which Leibniz’s argument emerges. Both Leibniz and Bayle claim that the existence of a traditional monotheistic God is demonstrably true and Leibniz has also argued that Bayle has not presented any arguments that require belief in an evil principle. Thus, Leibniz takes himself to have defeated cosmic dualism insofar as the debate involves paying attention to the local ‘context of justification’. However, Leibniz cannot deny that cosmic dualism has a long history and that it established itself in the form of a religious tradition that had many adherents. And, whilst Bayle himself does not rely on this to mount an argument, one might be tempted by the thought that the popularity of Zoroastrianism and Manichaeism confers at least some positive epistemic status on the cosmic dualism contained within them.

My suggestion then is that when he gives his genealogy, Leibniz is offering an argument which is supposed to undermine this limited consideration in favour of the doctrine by explaining how cosmic dualism might have become embedded in the consciousness of a group of people on the basis of nothing more than the mythologizing of aspects of the history of that group. In doing so, he shows that the truth of monotheism is compatible with the emergence of widespread commitment to cosmic dualism in a way that renders that commitment intelligible. On this reading, there are still questions to be pursued about the strength of the argument, given that Leibniz does not provide reasons to think that his genealogy provides a more adequate explanation than the hypothesis that the truth of cosmic dualism was at one time revealed. However, he is at least not guilty of a simple genetic fallacy, and it should be remembered that among the audience for whom the Theodicy was written, those sympathetic to the existence of revealed truth would have all agreed with Bayle that monotheism had been revealed rather than cosmic dualism.29

6. Conclusions

In this paper, I have offered an account of the way in which Leibniz engages with Bayle’s defence of cosmic dualism. For the most part, it seems to me that his responses are dialectically adequate. However, I have also pointed to places in which one might raise further questions for Leibniz if one did not share Bayle’s commitments. There is much more that could be said about all of these matters, and in closing I want to draw attention to some interesting questions that the debate over cosmic dualism throws into relief.

We have seen that Leibniz insists that we can prove that the world that God created is the best possible world and yet must admit that the reason that it is the best possible world is incomprehensible to us. By his own lights, this is ultimately enough to rebuff worries that God’s choice of a world with evil cannot be explained. And yet, Leibniz goes on both to insist that there are explanations and that he can give them. It is natural to wonder at this point why Leibniz does not simply bite the bullet and fall back on his claim that the reasons for God’s choice of the actual world rather than other possible worlds are incomprehensible. Indeed, this question is all the more pressing insofar as we pay attention to the kinds of evils that are presented by Dostoyevsky and holocaust testimonies. Secondly, there is the question of what justifies the particular explanations that he gives given that he has insisted that the explanandum is incomprehensible. Finally, there is more to be said about Leibniz’s appeal to genealogy. Although I have focused here on the critical genealogy of cosmic dualism, the Preface to the Theodicy also contains what appears to be a vindicatory genealogy of Christian monotheism (See Theodicy Preface/Leibniz 1985, pp. 50–51). A more detailed comparison of these genealogies, alongside a consideration of Leibniz’s philosophy of historiography more generally, would surely be a worthwhile project.30