If we take the considerable literature on Hume's Dialogues and simply weigh it, which I have not done, we would probably discover that the heavier end of the scale would be the pages of those scholars who consider him a skeptic both about knowledge generally and religion particularly.Footnote ² This is by no means, however, the only interpretation – of Hume or of his Dialogues.Footnote ³ In this paper I will take up the very strange – if not thoroughly comic – happenings of the ninth dialogue, treating it as a sort of inset, a play within the play, or literary version of a heraldic mise en abyme of the book in which they are found.Footnote ⁴ My aim in doing so is threefold. First, I wish to show that there is an analogy in part nine – one that no single interlocutor draws, but that arises out of their interaction; this analogy is, or has, a certain comic adequatio to the traditional arguments about proofs for the existence of God such as they may be found in Aquinas. I then show that Philo's seemingly inconsequential example of the strange necessity of products of 9 in section nine is a perfected analogy of the broken arguments actually given in that section, one which allows us to see the decisive factor in the arguments for the existence of God, which have been the matter of discussion there and throughout the book. Finally, I will raise the question of the designer's intent in creating such a humourous piece.

1

The difficulties of part nine. In dialogue nine every character seems to contradict himself. Demea, who in part two had affirmed that no one of common sense had ever doubted the ‘certain and self-evident truth’ of the existence of God, says he will give an argument for this self-evident truth, one which will prove several of the attributes of His previously described ‘altogether incomprehensible and unknown’ essence.Footnote ⁵ Cleanthes, in criticizing Demea's argument, famously lays waste to his own. And, what has heretofore gone unnoticed in the scholarship, Philo's addendum about the products of nine is an analogy which perfectly illustrates the opposite of his point in this section and others that the necessity within a system needs no further explanation for the existence of the whole.

After the previous sections' difficulties with a posteriori arguments about God's existence and nature, Demea proposes, at the beginning of part nine, to deal with the glorious a priori argument, which he had first suggested in part two. To say, as M. A. Stewart,Footnote ⁶ that Hume's treatment of the a priori argument for the existence of God has a perfunctoriness which has carried over to his commentators seems a charitable critique of what goes on in section nine of his Dialogues;Footnote ⁷ but even the most charitable critics fail to pick up on a matter which, comically, seems to prove that however mistakenly his characters understand the matter of proofs about God, Hume's writing itself is more exacting. The critical failure to which I refer concerns Philo's seemingly inconsequential aforementioned addendum – first present, then excised, then reattachedFootnote ⁸ – about the algebra of products of nine. Note that I do not say Hume is more exacting, for, as I will discuss at the close, there is good reason to be unsure how much of what can be seen in part nine that particularly named bundle of perceptions himself intended us to see, or saw himself.Footnote ⁹ That Hume's writing may be more exacting than its author, too, I find comical – whether intended or not. I take it for granted that a person can be ironized by history, perhaps even by a history that precedes him; that the same thing can happen in the history of ideas is no less possible. What to make of these comic riches we will leave to the end of the paper. First we should set out how the comedy appears.

As Stewart says, ‘[t]o readers familiar with the conventional modern distinction between a priori and a posteriori reasoning, Hume's nomenclature may suggest the ontological argument,’Footnote ¹⁰ but that distinction, or something very like it, also appears much earlier than part nine, for several of the characters (Philo and Demea) agree that the existence of God is ‘self-evident’ earlier on in part two, while arguments based on evidence can only be useful to certify something about his nature. One wonders whether, in the mouth of Cleanthes or Philo, self-evidence is merely a rhetorical term for ‘really really evident’ or, perhaps, in the mouth of Philo, itself an ironically pious mouthing for the purpose of leading Demea into either self-cognition or just mockery. We might confess that the distinction between a priori and a posteriori is undergoing some shifting from Descartes through Clarke to Kant, but given all that it still seems fair to conclude that, ‘the ontological argument, as such, is never considered by Hume.’Footnote ¹¹

The inadequacies of the arguments in part nine have other unclarities besides the a priori/a posteriori dysfunction. Both Demea's and Cleanthes' arguments have elements of the then famous argument of Samuel Clarke, but their conflations, vagaries and missing stages or distinctions allow the conclusion that these bear ‘no more relation to the historical Clarke's argument than the ephemeral pamphlets which circulated in criticism of Hume's philosophy during his lifetime bear to his philosophy.’Footnote ¹² Further, without any reference to the contemporary Clarkean argument or the longer history of the distinctions between a priori/a posteriori and self-evident/evident truths, the arguments given by Demea and Cleanthes are themselves sufficiently broken up to inspire a variety of critical menders whose mendings have tended to a variety of judgments about the reformulated arguments (and the author/characters) ranging from utter incompetence to philosophically decisive.Footnote ¹³ The rude mechanicals of Midsummer Night's Dream do as well at classical tragedy as Hume's characters seem to do at philosophy: that is, they are perfit – at comedy.

In concluding his careful commentary on Hume's Dialogues, William Lad Sessions suggests that the “Dialogues as a whole is like Cleanthes's design argument, in that it can work ‘its universal, its irresistible influence’ even though it cannot be reduced to logically tidy form.”Footnote ¹⁴ I propose to read section nine of the Dialogues as a sort of inset, a play within the play, exhibiting the thesis professor Sessions implies is Hume's literary procedure, though I will be considerably more skeptical about Hume agreeing with Cleanthes in the end. Were it not so ungainly I would prefer to title this essay more descriptively as “A (would-be) Thomist Reads (the argument of!) a Fideist in the Book of a (probable?) Skeptic: The Comedy of Part Nine of Hume's Dialogues,” but you see how it is – altogether too busy. It sometimes feels, in reading the dialogues, that Hume is altogether just this busy; the book is a piece of effervescent humours, like that of the above mentioned mechanicals who reach the sublime only under the mode of the ridiculous. But our job as philosophers is to examine the joke. So, let us begin.

2

Hume's characters and the earlier tradition. I am interested throughout in relations of ideas rather than matters of fact and existence; I am skeptical of the value of historicism for understanding either philosophy or literature. Since my first point is to exhibit an analogy between Hume and the medieval tradition that I think thought-provoking I will begin by merely summarizing, without defense or argument, some matters of Thomistic philosophical theology. Aquinas' position on the power and extent of philosophical proof in theological matters is more complicated than the “vulgar theology” Cleanthes confesses he has been espousing.Footnote ¹⁵ Cleanthes' argument can be seen as a modern vulgarization of what Aquinas called the fifth way; it is vulgar precisely in that it presumes that the order of the universe does not merely prove a cause of order, but ‘a mind like the human, and the liker the better;’Footnote ¹⁶ it is modern in its strict empiricism. Cleanthes defends this vulgar anthropomorphism whenever he is not proposing a Great Pumpkin or allowing that there is a kind of order in the rotting of a turnip.Footnote ¹⁷ For the sake of the comedy let us call him the modern Pyramus.

Aquinas considers that while the existence of God is self-evident in itself, it is not so to us.Footnote ¹⁸ By this he means that God's essence includes his existence, but our mind is not capable of seeing God's essence in our present constitution; to do so we must ‘be made deiform’ by grace.Footnote ¹⁹ So a person would be right to say that God's essence includes his existence (for his essence is to exist), but comprehension of this essence is never ours, and knowledge of it is not directly available to us now, so no argument can be made about this essence a priori. Rather, we come to knowledge of God's existence by arguments which begin from the existence of things; thus all proofs are a posteriori. Against the (a priori) argument that the idea of God is the being whose existence is included in the idea of his essence, so he must exist, Aquinas says that even if we admit that the idea of God necessarily includes existence, this does not mean the being has to exist. We find Cleanthes agreeing with Aquinas on this matter when he says, ‘Whatever we conceive as existent, we can also conceive as non-existent.’Footnote ²⁰ Let us call this a priori argument, of which, Aquinas says, we only grasp the cloak but not the real being, Thisbe. Aquinas's distinction between self-evident and evident considers self-evidence as a matter of understanding the essence of a thing: as it is clear that a whole must be greater than its parts to anyone who knows what both whole and part are; or, as ‘spiritual substances do not exist in space’ is self-evidently true for anyone who can understand the idea of spiritual substance. The problem is that with regard to the essence of God our eyes are like those of the bat to the sun; not being able to know his essence (in our present state) his existence cannot be self-evident to us. Since we are unable to grasp the essence of the divine cause it is impossible to argue from our idea of it to any effect – including that essence's existence – as an a priori argument would; we do not descend from cause to effect. We should also point out that while Demea and Philo agree that God's being is ‘incomprehensible,’ it is not clear they are agreeing in their use of that term with Aquinas who himself makes a distinction between comprehension of the divine essence – which is never possible to created being – and (intellectual) vision of it which is possible.Footnote ²¹

We can, on the other hand, argue from effects to causes, and Aquinas turns next to the five ways of a posteriori argument, among which ways of proving there is a certain order.Footnote ²² The argument from motion, which is the easiest way, is a special case of the argument from efficient causality, and both of them depend in the end upon the fact that potentiality can only be reduced to actuality by something already in act (of moving or of causing respectively). The third way begins from the fact that all the beings we experience come into existence and fail of existence; the most basic kind of change (of which the first two arguments are special cases) is from not being to being. From not being in the side pocket to being in the side pocket is a motion of the 8 ball, which motion needs a tracing of possible and actual motions back to a first cause; from being a tree to being a cue stick needs a tracing of efficient causality in which actual lathes and other actual causes reduce possible cue sticks to actual ones; the third way tells us that the reduction of any and all beings which can possibly not-be to actually being requires an actual being who cannot be like that, a being who does not have an existence that is received from another being, for an infinite regress among possibles still leaves us only a possible, not an actual being. But we have an actual being before us; therefore, there must be some being whose actuality is not received. This is the Prime Existent; it has existence of itself, or, its existence is necessary. Aquinas considers this same issue from the other direction (coming forward rather than tracing backward) as well: if all those beings could possibly not-be then at some time there would be nothing, and from nothing nothing comes. Yet here we are. Either way of thinking about it arrives at the same conclusion: that there are actual beings which are able not-to-be requires a fully actual being; a being without possibility or potentiality, one who's being is not possible not-to-be, or received from another: a being who's being is necessary.

The fourth and fifth ways depend upon this same consideration of metaphysical dependency – the existence of beings more and less good etc. (way four), depends on the existence of a being who is the actuality of good (etc.), for the good which can come to be and fail, which admits of more and less – that is, admits of possibility – can only be reduced to actuality by a being which is good – the actuality of the quality we see coming to be. Similarly, things which are ordered, but could not be and could be disordered must be ordered (finally) by an actual orderer; an orderer without possibility of disorder or non-existence – since we have order. Contemporary philosophers consider the pivotal argument – way three – as turning on the principle of sufficient reason.Footnote ²³ Aquinas is perfectly clear in all this that we get a very skinny God from these proofs; for, when we do not know the cause, the middle term in our demonstration is always “cause of effect x” which we substitute for the essence, which is the true middle term of a demonstration. This argument, skinny though it be, is the traditional Pyramus.

An analogy might help understand Aquinas' last point, about the permanent skinniness of Pyramus. If we are doing a physics experiment with a bubble chamber and photographic plate, we would see certain kinds of sparks on the plate and certain bubble tracks, and we postulate that they differ according to a number of variables. We could name the tracks neutrino, electron, positron, etc. but what we really mean is “cause of bubble track one” or “cause of spark a” which track turns differently than track two or three, or which spark has a different intensity or color from b and c. We mean by these names to distinguish “the cause of” one sort of effect from another. We understand what “the neutrino” effects, not what it is. So, what we get in the proofs is First Actuality, Uncaused Cause of order, good, motion, etc. And Aquinas thinks that Aristotle's God is about as far as philosophy alone gets: a First Cause of motion and all effects, but not necessarily a creator;Footnote ²⁴ an eternal source of order, but not necessarily an anthropomorphic designer, certainly not a watchmaker; a source of good who has no moral – in the sense of willing – relation to any being, but who by his being inspires every being to achieve what good its nature can. What we get is a simple, completely actual first cause of being who cannot be otherwise, and causes all order and good that exists, and it does this (in Aristotle, for example) merely by thinking itself.

It is curious that all of these characters – except the skinny God – appear in part nine of Hume's Dialogues, and what is more curious is how they appear. In fact, the traditional arguments (of Aristotle or Aquinas) seem sawn and joined about as strangely by the characters of part nine as Peter Quince and his crew's carpentering of the traditional tale of Pyramus and Thisbe. Let us first speak of Demea, who wishes to present Thisbe – an a priori argument. As brevity is the soul of wit, I will be brief: the argument he eventually gives runs thus:

But this appears to be either Aquinas' second way (efficient cause) or third way (possible being needs unconditionally actual being), or some jointure of them (which jointure is recognized by Aquinas – but he is able to distinguish them first); neither of these were thought of as a priori by Aquinas, or by Aristotle, since they start from the fact and experience of existing things; further, Demea himself notes that he is mounting up from effects to causes, not the converse. So, though Demea wraps himself Thisbe's cloak, he is an altogether traditional Pyramus.

Demea is not a complete idiot.Footnote ²⁶ He seems even to go so far as to agree with Aquinas' belief that the whole order of causes and effects in existing things could exist from all eternity, for he follows his first turn as the traditional Pyramus with what looks to be another – a version of proof which nods at the fifth way, and returns to its dependence on the third:

Now if Demea is reasoning, his last phrase must mean that we cannot suppose this necessary cause of order not to exist without expressly contradicting the reasoning which we have just gone through. That reasoning is a posteriori, starting from the existence of a particular order of causes (the actual universe), which, even if eternal, needs a cause of its being, for there are many orders that could have been; the possibility here (and perhaps eternally) actualized needs a perfectly actual cause, a cause which cannot be otherwise. His last line may look like Thisbe, but so far as it reasons the argument can only be the traditional Pyramus repeating himself.

Cleanthes, however, first attacks the name Demea had given his argument – a priori, saying ‘[n]othing distinctly conceivable implies a contradiction. Whatever we conceive as existent, we can also conceive as non-existent.’ This, we have seen, Aquinas himself is willing to say against the idea of God being self-evident to us. In fact, a bit later Cleanthes will nearly quote Aquinas' thought on the matter of an a priori proof when he says,

Aquinas says we shall see the essence of God through being made deiform, by His joining of Himself to us.Footnote ²⁹ All we see in this world is Thisbe's cloak, not the real Thisbe. Cleanthes follows this attack upon Demea's title with the presumption that Demea's last phrase – ‘who cannot be supposed not to exist, without an express contradiction’ – is a stand alone ontological argument. To it he counterposes that ‘[t]here is no being, therefore, whose non-existence implies a contradiction. Consequently there is no being whose essence is demonstrable. I propose this argument as entirely decisive.’Footnote ³⁰

It appears, then, that Demea has given a traditional Thomistic argument (or two or three) for the existence of God, which argument(s) depend upon noticing that things (or orders of them) exist, that things are brought into existence, and that there are particular orders of succession rather than other possible ones (or none) – arguments which are, in other words, a posteriori. Cleanthes rebuts him by criticizing the kind of argument Demea's title gave his argument rather than the argument given, and stating that the argument which has that title cannot be valid for us as we are now constituted. This point is also Thomistic. More comically, we have a traditional Pyramus who is calling himself Thisbe and a modern Pyramus dressed as a traditional Lion who is tearing at Thisbe's cloak (which our traditional Pyramus has been dressing himself in). Demea (the traditional Pyramus) scampers away thinking himself Thisbe. Considering his mauling a disaster, he gives up not only his beloved's cloak, but any semblance of human heroism, by which I mean he gives up not only the title of his argument (Thisbe), but the argument he really is (Pyramus) – against which Cleanthes' tearing has (so far) done nothing; in fact, he gives up all argument itself and ceases reasoning altogether. He reduces himself to Moonshine, ‘and, from a consciousness of his imbecility and misery rather than from any reasoning, is led to seek protection from that Being on whom he and all nature are dependent.’Footnote ³¹

The lion, however, is really no one other than the modern Pyramus momentarily in another costume. His tearing at Thisbe's cloak has done nothing to the traditional Pyramus and little to Thisbe, who in any case has not appeared; our modern Pyramus is equally far from rescuing anything, rather, he now removes his costume and feeds himself to the Lion. Cleanthes – a modern Pyramus attempting to defend an anthropomorphic designer, who must be considered to exist behind all of nature's various machines, which, even in ‘their most minute parts, are adjusted to each other with an accuracy which ravishes into admiration all men who have ever contemplated them’Footnote ³² – now considers that the material universe itself might be the necessarily existent being. The universe, he says, ‘for aught we can determine, may contain some qualities which, were they known, would make its non-existence appear as great a contradiction as that twice two is five.’Footnote ³³ But this argument agrees precisely with one of the many suggestions the real skeptical Lion, Philo, has already produced: ‘Chance has no place on any hypothesis…  Everything is surely governed by steady inviolable laws…  Instead of admiring the order of natural beings, we should clearly see that it was absolutely impossible for them, in the smallest article, ever to admit of any other disposition:’ There is ‘an eternal inherent principle of order in the world.’Footnote ³⁴ Cleanthes clearly abandons the central issue of the traditional proofs (as well as his own), which depend upon our noticing that both the order and the existence of everything in the universe gives every indication of being able not-to-be; such (possible not to be) actualities cannot explain their own existence, and to explain the existence of something rather than nothing we need a being who is perfectly actual, and has no possibility of not being so.

At this point one might expect the skeptical lion to declare victory, but he had already done so – with an appropriate lion-like metaphor – just before dialogue nine began:

Thus Lion. Dialogue nine seems merely to re-enact his victory – without the necessity of his least activity.

We can see that section nine, to this point, has a certain comic adequatio to the traditional Thomist ideas about the proofs; Demea speaks of the glories of the a priori argument, which never appears to us here – exactly as Aquinas said it could not; Cleanthes attacks the idea of the a priori argument exactly as Aquinas attacked it, which attack touches Demea's actual argument at no certain point; Cleanthes then destroys his own earlier anthropomorphic arguments by presuming, first, that the world might always be in being (which his usual anthropomorphic design argument requires to be false) and, second, that “to remain always in being” is equivalent to “exists necessarily,”Footnote ³⁶ unaware that it was once possible to consider the world both eternal and contingently actual, for Demea has said exactly that is the question, and Aquinas considered the thesis possible though divinely revealed to be false. Cleanthes (or Hume) may see that a distinction can be made between eternal and necessary, but Cleanthes does not admit this distinction has being in the dialogue, Demea does not return to defend it, and Hume (as deus absconditus) lets this lack of distinction be. So, by default, the possibly eternal is equated with the necessarily actual. How does one make a lack be? By not bringing something into existence. What power could have brought this (lacking) distinction into existence here in the dialogue? Only the deus absconditus of the dialogue: Hume. But he actually absconded, and so we have a failure in the discussion between these characters to consider one of the distinctions on which the traditional proof depended – even though it has been mentioned. Now whether this lack of coming to be of the important distinction is due to a lack of intelligence in the all too human designer,Footnote ³⁷ or is itself part of the humours these characters express is a well too deep for my dowsing.

3

Philo's addendum: the qualities of nine. Well then, rather than ‘starting any further difficulties,’ or making a distinction which might clarify the preceding inconcinnities – of which I have only brought forward those which allow the comic analogy to the medieval tradition, Philo finds himself unable to forbear ‘insisting still upon another topic’ (9, 56). He introduces the strange mathematical fact that products of 9 always add up to 9, or some lesser product of 9. And he tells us that this wonderful regularity might seem a matter of chance or design, but really is a necessity in the number system itself. He uses this as an analogy, not for a new topic, but for precisely the one by which the modern Pyramus had just fed himself to the lion: the whole economy of the universe might operate on a like inbuilt necessity.Footnote ³⁸ This necessity comes not from some deus, feeble, aged, decaying or unpracticed, but lives (as it were) in the number system itself, as necessity and design – so goes the analogy – might inhere in ‘the whole economy of the universe’ itself ‘though no human algebra can furnish a key’ which explains it (9, 57). Just so, the universe itself may be the being which cannot not be otherwise than as it is.

Here too, a distinction is not made, and it is precisely the distinction that the Thomistic argument would insist upon. It is also illustrated precisely in the example Philo gives. The distinction is that between the necessity within an order (which order may or may not exist) and the necessity for that order itself to exist. For the necessity Philo points out within the system of base 10 algebra says nothing at all about the existence of base 10 algebra; just as the necessity within the laws of the physical universe explains nothing about the existence of this particular physics to which we are actually subject. In fact, an infinite number of possible number systems are available, and each of them would have their own like internal necessity. For example, if we counted in base six, we would discover that products of 5 would add up to five or lesser products thereof, and similarly for every other possibility of counting in base n, the products of n-1 would always be found to add up to n-1 or lesser products thereof, presuming we used single characters for each number until n. This being so, we might ask, “why this particular system of necessary connections rather than any of those infinite others?” That each system has internal necessity is true, but only one of them IS the system we experience and use. Why that? The explanation for our particular system out of all the possible systems has to be (according to the traditional argument) a being already actual. The human being, the actual being who makes our system of enumeration come to be from all those n possibilities, has ten fingers; these beings cause one system to be used – base ten – and the necessity about 9s which is possible comes to be seen in the world of this actual maker. For it is perfectly possible that we could count in base six, and perhaps the three-toed sloth does so, it being so much easier than 10 – for them. If the three-toed sloth does mathematics, it experiences a different system of numerical necessities: the products of 5 necessarily add up to 5 or lesser products of 5.

So, that it is ‘human algebra’ is precisely the reason why we note this mysterious necessity among products of nine, and see it on such a regular basis. It is the actual existence of the algebraic human (using Arabic numerals) which causes the necessity we see to come into existence. That we have this necessity rather than a similar one with 5s (as the sloth would note) is accounted for by another being's actuality – it is not in the numbers, but in us; if ET has 12 fingers (and a sufficiently Arabic-style system of enumeration) he will notice the same necessity pervades his algebra at the digit that stands for 11. All of these are possible systems of algebraic necessity; the cause for the existence of the algebra we actually have is not itself a necessary being – for we are not that – but it is an already actual cause. There must be some already actual cause for the one rather than the infinite others, and that this cause must have its actuality in a quite different way than the received actuality it gives to that one system is the distinction the traditional proofs depend on. By analogy then, for all of the possible skeins of universes with all their necessary laws and actual connections, there must be an already actual cause for the one that exists. That cause must be actual, and its actuality is not an element within that skein of possibility, but rather makes that particular skein come to be. That causing actuality must have its actuality of itself, or else have its possible actuality be actualized by something else (as is true of the human cause of base 10 algebra), and these something elses cannot go on to infinity, for in the dialectic of existence the denial of a first existence denies the existence of all the intermediates. The first being's existence must be an existence which not only does not suffer being brought into existence, but can suffer no further actualization at all for it is perfectly actual. To be possible depends on to be and This be must be actual, or all other actualities, and all other skeins of necessity, fail to be. This is what we have called the traditional Pyramus – one who does not err with the lion, or mistake his beloved for a cloak.

3²

The designer's purpose? And again, I must say that I do not know whether Hume was aware of the way the algebra his character considers works, and how infinite the possibilities it (does not) refer to are; so I do not know whether Philo considers this algebraic analogy as a hint to the true argument which is also lacking from this section, or whether both Philo and Hume are giving silent testimony to an actuality whose providence actually escapes them, without letting them escape. Whatever the case is, it cannot be denied that the example Philo gives illustrates precisely and exactly the point of the traditional arguments: what needs explanation is the being of the world (or system). A cause of its being is necessary for any system of world to come into being, since it is possible for it not to be and it is possible that another system (in the case of algebra, an infinite number of other systems), or none at all could be the case. The order must have a cause of its being and of its being the order it is. This cause of order must have its actuality of itself; et hoc dicimus deum.

That the mystery of the products of nine appears in section nine of his dialogues, suggests a distance between Philo and his cause, a distance which allows Philo's intelligence has a cause plausibly more subtle, which makes his subtlety both be and be given away. We should note again that Demea's main point seems to be that even ‘a whole eternal chain or succession, taken together, is not determined or caused by anything: And yet it is evident that it requires a cause or reason,’Footnote ³⁹ which Philo's ‘products of 9’ example, rather than destroying, exemplifies – to the nines. We should remember that Cleanthes had first introduced mathematical analogies; he had said that if we knew God's essence we should see that ‘it is as impossible for him not to exist as for twice two not to be four,’ while the universe might have some quality which would ‘make its non-existence appear as great a contradiction as that twice two is five,’Footnote ⁴⁰ but this introduction of algebraic contradiction (so ostensibly of relations of ideas) is (as we have seen) an introduction of an existentially dependent necessity: it occurs in the particular algebra human beings use. And such necessities need causes of their existence that are more absolute – or else we must think it chance or nothing which gives rise to the necessity (among all of those possible) that is actual. (In base 4, by the way, twice two would not be five but ten – or so it would look using our present system of notation.) Something must BE an essence which is not so dependent, for possibilities do not explain their own being – not even if they are possible skeins of absolute necessity like algebra; a necessary actuality is required if one of them is actual, this actuality is not a member of the system but the cause of its being, et hoc dicimus deum.

Like Shakespeare, Hume has reached this sublime result in an utterly ridiculous manner; perhaps, considering the already noted fact that Hume questioned whether this part of Philo's part in part nine should be in the text, there is an actual providence more providential than the author's intent operating in the text. Perhaps, on the other hand, there is such a thing as pure blind chance falling magically into significant order, for that the magic of the products of nine appears in section nine of the dialogues does seem to signify. There is, in any case, this further analogy to consider, namely the analogy between the one who made this particular comedy come to be and the one who made the universe come to be. Demea considers Cleanthes' earlier book analogy to be dangerous, for

I wonder if reading Hume is so easy; if he is a skeptic, and if Philo speaks for Hume, then he has given a very foolish example – as foolish an argument as those self-destructive arguments given in this same section by Cleanthes, who feeds himself to Lion. The example Philo gives demonstrates exactly what he has been arguing against: the necessity for a cause of being of the particular order of results we have, even given that the necessity of each individual result in the order may be explained and predicted. On the other hand, if our author knows what he is doing, then he knows that his skeptical lion is merely a costume, and the being beneath the comedy of these characters is the traditional Pyramus, who stands nakedly before us in the algebraic example, but whose presence is not noted by anyone – except the author who (secretly) calls this part of the dialogue by Pyramus' lucky number, nine.

Given the undecidability of authorial intent in this section, what should we think of the book as a whole? For surely it is a strange book on natural religion in which no one accurately gives the best and traditional argument of natural theology in any clear way.Footnote ⁴² Demea has considered natural theology, like all of reason's works, a questionable undertaking from the first; Cleanthes' natural theology has been far more vulgar than Aquinas' or Aristotle's metaphysical reasonings; and Philo has, by his own admission, been ever on the attack, with not only no permanent abode, but, it seems, no direction home either. Hermippus – the receiver of all these speeches, named after the god of secret messages – is sealed at the mouth; we cannot even read his lips; and Pamphilus seems merely a blank slate – or pamphlet – most deeply written on by his guardian, Cleanthes, but carrying along the experience of other tablet carvers too.

Perhaps then, that is the point. Perhaps if one is an eighteenth century empiricist, as all of the interlocutors are, there are some things one cannot get clear on, some things the mind cannot come to, much less grasp, limited as it is to the deeper and shallower groovings of its waxy tablature. Could such a mind even count? What is it that stays constant in the move from one to two to three? Nine is infinitely far away. What is it that achieves the whole of nine, not one and one and one and one…  What is the being that actually unites these possibles? This act of unification too, contrary to Cleanthes, needs a cause.Footnote ⁴³ But let us leave the problems of the empiricist philosophy of mind to one side.Footnote ⁴⁴ The “natural theological” arguments of the Dialogues seem to show, as Cleanthes half summarizes, that ‘the more we abandon all human analogy’ (11, 67), the more we open ourselves to every form of blind dogmatism, a wide variety of which it is Philo's pleasure to consider – from the infant deity, to the committee of usual idiots, to the animal, vegetable, or pure blind force which settles eventually into form. Philo himself will conclude that there is some analogy between the workings of a mind and the rotting of a turnip, so the order in the universe probably has some remote analogy to this (12, 81), and the quarrel can only be one over degrees. There is truly not much to choose from among these;Footnote ⁴⁵ but these are, we MUST admit, the hyperbolic graph of possibilities for the empiricist: our analogy, if not to the human, must be to some other thing we experience – its form, matter or operations. But the traditional argument does not depend upon an analogy to a human maker, or a vegetable, the traditional argument depends upon no analogy at all, but on the fact that beings which can not-be need a cause which cannot be that way. The dependent being must have its being from a being which is independent; and this reception of being will allow an analogia entis between the dependent and the independent being, but such an analogy is not an analogy of form or of matter or of operations. All the options of empiricism suffer comic limitation to the form, matter and operations of a more or less warm piece of wax. But this no one calls God. Hume thereby proves natural religion cannot amount to much for an empiricist.Footnote ⁴⁶ So perhaps by reducing (empiricist) natural religion to absurdity Hume means to reduce empiricism to absurdity? In any case, if we are not to be left with varieties of blind dogmatism – all we are permitted under empiricist theories of mind – we shall have to abandon or strictly limit empiricism. Kant takes this hint; the medieval philosophers, as well as Plato and Aristotle, never needed it.