SOURCE: “Philosophy as Literature: The Case of Hume's Dialogues,” in Compendious Conversations: The Method of Dialogue in the Early Enlightenment, edited by Kevin L. Cope, Peter Lang, 1992, pp. 34-53.

In a well-known passage toward the close of Book I of his A Treatise of Human Nature,¹ David Hume tells us how, when he reflects on “the condition of the learned world, which lies under such a deplorable ignorance” of the fundamental principles of philosophical learning, “I feel an ambition to arise in me of contributing to the instruction of mankind, and the acquiring a name by my inventions and discoveries.”² The cultivation of a proper style in which to communicate his “inventions and discoveries” was one of Hume's life-long preoccupations but, unfair though it seems to fault any man for such a concern, his public acknowledgement of it seems to have brought him little else than notoriety well into our present century. Ever since the publication of the Treatise over two hundred years ago, Hume's reputation has been beset by critics (not all of them unfriendly to his ideas, by any means) who have questioned the seriousness of his purpose in writing philosophy on grounds that he was more concerned with gaining the quick rewards of a literary renown than pursuing philosophical truth for its own sake. As his latest biographer, E. C. Mossner, points out in a very perceptive article,³ since Hume's death the single most consistently used document to support this scurrilous view has been his own seven-page autobiography wherein he so candidly speaks of “my love of literary Fame, my ruling Passion.”⁴ It is hardly too much to hope that Professor Mossner's forcefully argued and thoroughly documented case to prove that the tradition in which Hume's “ideas are taken as an attempt to do something other than strictly inquire into truth … is based upon distortion of psychology and misrepresentation of fact,”⁵ has forever made it impossible for anyone to lay this old canard at Hume's door again. The truth of the matter is that Hume is that rare phenomenon: an original thinker of great power and subtlety who is also a skillful artist—and he could hardly have succeeded so well in being both of these at once had the driving force behind his work not been a single-minded devotion to the discovery and propagation of truth.

In the posthumously published Dialogues Concerning Natural Religion, composed at a time when Hume's creative powers were at their peak, his talents as a philosopher and writer merged to produce a work of incomparable genius. The subject matter of the Dialogues is what we today refer to as philosophical theology. Their central concern is to debate the cogency of the so-called argument from design for the existence of God. During its heyday in the eighteenth century, this argument was thought by many enlightened minds to prove, by ordinary inductive reasoning, that the order and design evident in the world around us have a supernatural source in a Benevolent Intelligence. The discussion, which is related by a narrator, has three participants: Cleanthes, an empirically-minded theologian; Philo, a skeptic; and Demea, an orthodox theologian of strong anti-intellectualist leanings, who is, however, not averse to putting forward an a priori argument for God's existence. Cleanthes advocates the argument from design, while Philo devotes most of his energies to vigorously attacking it—in the name, he professes repeatedly, of a brand of religious fideism according to which the nature of God is incomprehensible to the human intellect. In his defense of this orthodoxy, we find Philo during much of the discussion forming a rather unlikely alliance with the third participant, Demea. But on the question of the a priori argument, Philo aligns himself squarely with Cleanthes against Demea, in rejecting it.

In his excellent little book on Hume, Terence Penelhum remarks of the Dialogues that “The very fact that the work is presented in dialogue form has itself to be taken with maximum seriousness.”⁶ It would, I am sure, be difficult to find someone who would disagree with this remark. And yet, regardless of the obvious importance of this issue, very little work has been done until quite recently to elucidate Hume's views in regard to the special literary criteria which are involved in the dialogue mode of writing philosophy, and what little there is to be found on this rather special topic in the various commentaries on the Dialogues, is, in my judgment, either inadequate or misguided. It is my intention in this essay to try to correct this situation by attempting an appreciation of the kind of literary enterprise Hume was undertaking when he wrote the Dialogues, while at the same time explaining why, and in exactly what manner, getting clear about this issue is an indispensable prerequisite for arriving at a proper understanding of the Dialogues. Ironically, the very success with which Hume was able to carry out the tasks he set himself in writing his Dialogues, has itself contributed to the controversies that have raged around it.

In the following two sections I will develop, using Hume's own Dialogues as the paradigm, some hints Hume has himself provided in that work and elsewhere, as to the test of a good dialogue. The most difficult challenge one faces in attempting to give an adequate theory of the dialogue that would do justice to Hume's outstanding achievement, is this: How to give an account of the distinctive literary features that set Hume's Dialogues apart from a straightforward philosophical treatise, without in the process reversing Hume's priorities, that is, without leaving Hume open to the charge that he was using “ideas … to do something other than strictly inquire into the truth.” It is my view as a matter of fact that not only is it wrong to set Hume's literary aim above the philosophical one, but that it is a fundamental mistake to try to speak of these two objectives as if they could be separated, except at a very superficial level of writing, when analyzing the dramatic qualities of the Dialogues. Hume's Dialogues are neither as obviously didactic as Berkeley's, nor as tangential as Plato's. They contain Hume's own doctrine in regard to natural religion alright, only it would be an error to think that Hume's views can be abstracted from them before their dialectic has been allowed to run its course. I believe (as I will try to show later in this paper) that a disregard of the dialectic nature of Hume's Dialogues has been one of the chief sources of the notorious disagreement that exists among commentators over the question of the message Hume wished to leave in it for posterity.

This notorious disagreement over the interpretation of the Dialogues among Hume scholars, provides the point of departure for an article by Professor John Bricke in which he sets out a “way of approaching the Dialogues … which puts proper stress on Hume's literary objectives in their composition.”⁷ I am in full agreement with Bricke's general “claim that one must look to their literary form if one is to understand the Dialogues”⁸ but I disagree with the specific manner in which he describes the special literary requirements of the dialogue form Hume recognized. In the course of my explication of the special literary criteria Hume employed in the composition of his Dialogues, I will give in some detail my reasons for rejecting Bricke's account. But for the moment let me say that, in my view, Bricke's well-intentioned attempt is vitiated by the fact that he places undue emphasis on a comparatively superficial literary consideration involved in the writing of a dialogue. The result is that we are given a rather facile picture of the structure of the arguments in the Dialogues which is incompatible with the ultimate excellence of Hume's achievement, and comes perilously close to reversing Hume's priorities in the manner referred to above.

I would like to close these introductory remarks by briefly noting an issue which is central to Bricke's interpretation of Hume's literary objectives in the Dialogues, but somewhat peripheral to mine. Typically, most attempts at interpreting Hume's views in the Dialogues have assumed that one of the two main participants in them is Hume's principal spokesman and that he holds a fairly sustained and consistent point of view throughout most of the discussion. We shall see later that Bricke argues that proper attention to the special literary task Hume set out to achieve in the Dialogues, shows that it is fundamentally mistaken to make this two-fold assumption. One of the interpretations that thus gets caught up in Bricke's net, is the well-known Kemp Smith interpretation. In the sequel, I will myself be criticizing some remarks Kemp Smith makes in regard to the dramatic quality of the Dialogues in the course of his classical commentary. I believe, nevertheless, that Kemp Smith's scholarly instincts were very sound and that, when properly understood and with only slight qualifications, the interpretation he supplied is the closest we are ever likely to get to a correct understanding of Hume's views on natural religion. But beyond nullifying Bricke's refutation, any sort of adequate attempt to defend Kemp Smith's interpretation falls outside the very special scope of this paper. To avoid any misunderstanding in this connection, a point I must make clear is this: I do not think my way of looking at Hume's literary achievement in the Dialogues is necessarily committed to Kemp Smith's, or any other specific interpretation as regards who in them speaks for Hume (or even as regards what Hume's views are on the question of natural religion). In other words, unlike Bricke's, my interpretation of the literary criteria involved in the composition of the Dialogues would leave the question whether or not one of the participants in them chiefly speaks for Hume, to be argued for on its own merits.

II

Hume has left us a concise and very clear statement of the literary criteria he set out to meet when he began writing his Dialogues, in the opening passage of this work spoken by its narrator, Pamphilus. In the interest of brevity, I will not append the passage but simply go on to say that the passage shows that there are three important special requirements of the dialogue form that Hume recognized. Unfortunately, most commentators who have had anything to say on this subject (including Bricke, who refers us to this passage in substantiation of his view) pick out only two of these requirements, and (what is even more unfortunate) misconstrue them. These can be easily summed up in the two words: naturalness and philosophical balance. The third essential requirement is not so easily named, but that Hume recognized it is abundantly clear from the passage being referred to and from the fact that Hume satisfied it admirably in the Dialogues. I shall christen this requirement “the dialectic requirement” but say little else about it directly in this section. Later we shall see that there is an intimate connection between the requirement of a proper philosophical balance in a dialogue and the possibility of that dialogue satisfying the dialectic requirement; and that it is the misunderstanding of the kind of balance Hume was aiming at in his Dialogues, which has led to the neglect of their dialectical quality. In fact all three requirements overlap with each other in ways that are complicated. But I will try below to consider them in turn, as far as this is possible, though this will result in some unnaturalness in my exposition.

We begin with the two requirements according to which a dialogue should be natural, and preserve a proper balance among its speakers. I think we are entitled to say that there is a certain level of writing at which the criterion of naturalness is more straightforwardly literary one of the two. What I mean is that, at this level, the satisfaction of the criterion requires the use of certain obvious devices of the writer's craft which are not distinctive of the philosophical dialogue as such. They include, for example, such things as the proper choice of setting (the creation of the right atmosphere, as we sometimes say); the gradual and believable building up of the characters of the participants (so that, by and large, what persons in the dialogue say, as well as how they say it, matches up with who they are); the throwing in of bits of “small talk’; and so forth. Although the skillful employment of such devices is important in bringing alive the conversation among the dialogue's participants, it has little to do directly with the preservation of a philosophical balance among them. I say “directly” advisedly, because considerations of balance cannot be wholly excluded. For example, what kind of character a person in the dialogue is given may sometimes determine what sort of edge (or otherwise) in regard to the actual soundness of his arguments, he may be permitted over the others, without upsetting the overall dramatic tension in the whole.

But there is also another level to Hume's creative achievement in the Dialogues at which terms such as “natural” and “unaffected” are also appropriate. At this level no such easy distinction can be drawn between what works successfully as literature and what works best as philosophy. It is to this level of writing that Pamphilus is referring when he says that a dialogue-writer who wishes to carry on “the dispute in the natural spirit of good company” will do so by “preserving a proper balance among the speakers” (127). I will devote the rest of this section to an explanation of what Hume's requirement of a “proper balance” in a dialogue entails. We will see that what kind of philosophical balance is achieved in a dialogue—whether, that is, at this level it is natural and unaffected or artificial and stilted—will depend on what kind of literary strategy is used to achieve it; and not all strategies are equally effective or intellectually respectable, by any means. And we will see in the course of this discussion that some otherwise well-intentioned commentators have unfortunately imputed to Hume the deliberate use of certain strategies which, had Hume really resorted to them, would have destroyed the philosophical as well as artistic integrity of his Dialogues.

In regard to the requirement of a proper philosophical balance, two related points must at once be made. The first is that the question of the choice of subject matter is here crucial and the reason it is crucial is that it is this choice that will in large measure control (and this is my second point) what kind and quality of dramatic confrontation is possible between the leading participants in the discussion. That the importance of the first point was not lost on Hume is amply clear from the following remarks the narrator makes as he warms up to the theme of the conversation he is about to report:

Any question of philosophy … which is so obscure and uncertain, that human reason can reach no fixed determination with regard to it; if it should be treated at all; seems to lead us naturally into the style of dialogue and conversion. Reasonable men may be allowed to differ, where no one can reasonably be positive: Opposite sentiments, even without any decision, afford an agreeable amusement: And if the subject be curious and interesting, the book carries us, in a manner, into company; and unites the two greatest and purest pleasures of human life, study and society.⁹

(128)

The simple truth is that it would be impossible to achieve and sustain the right kind of balance in a dialogue unless the subject around which the dialogue is written is one where it is possible to give genuinely weighty arguments on both sides of the issue. Obviously, it would help if the author himself felt some pull in both directions, but this factor can be exaggerated. It is enough that the author be a person in whom the skills of a philosopher and writer are found combined with a lively philosophical imagination and sympathy, so that he is able to successfully alternate between the roles of defender and critic without committing (in Hume's words) “the vulgar Error … of putting nothing but Nonsense into the Mouth of the Adversary.”¹⁰ The essential thing is that the subject of the dialogue be such that the possibility of a credible defense on both sides is inherent in the issue itself. As we all well know, philosophy may provide a goodly number of such topics: but a philosopher, contemplating writing a dialogue, does well to elect a theme where there already exists a well-entrenched and widely held received viewpoint among the informed public of his time. There are many advantages to this. The topicality of the controversy will ensure that a certain amount of interest and involvement of the reader can be taken for granted at the very start. At the same time it will afford the author ample opportunity for invention and originality, and a skillful writer will seize upon these to introduce elements of surprise and suspense which are so essential to making a dialogue come vividly alive for the reader. Must it also be the case that, in order to reach a high level of excellence, such a dialogue (that is, one in which the adversaries are well matched) end in a draw? In any other kind of exhibition of skill pitting two well matched adversaries against each other—say a tennis match between two highly skilled professionals—we do not feel the least bit put out because one of them wins and the other loses. Why must we expect anything different from a clash between two philosophers? Besides, it would be unreasonable to require the philosopher/author of a dialogue to remain perfectly disinterested in its outcome, especially if he uses the opportunity to explore to its limits a highly controversial philosophical issue. The requirement of a proper balance in a philosophical dialogue does not necessarily entail that there should be no winners and no losers. What it does entail is that there should be a good fight, between adversaries who respect each other's quality of mind even when vehemently disagreeing, and who (and this is most important) end up by earning the respect of the reader as well, regardless of the outcome of the discussion.

All the features necessary for bringing about a proper philosophical balance in a dialogue, are present in Hume's Dialogues. They deal with a theme in which the current interest ran very high and Hume was well-aware when he began work on the Dialogues that he was putting to the test a whole contemporary tradition (that of “experimental theism”) that most of the best intellects and the educated public of his time took for granted.¹¹

Now it happens to be the case that I agree with those who say that by the time Hume came to write the Dialogues he had completely emancipated himself from believing in anything that can seriously be termed a religion. Unfortunately, however, it is true of too many of us, that by the time we first approach the Dialogues, we have already become jaded by textbook readings and hearsay evidence into (to adopt Anthony Flew's apt words) “taking for granted that total emancipation which Hume achieved only by labor and genius.”¹² The result is that we often fail to see that the Dialogues, despite their occasional levity and the bantering tones in which issues are frequently discussed in it, constitute a serious record of Hume's intellectual struggle on the way to achieving this emancipation. The obvious casualty of this obtuseness is the intellectual character of Cleanthes who in some commentaries is reduced to the status of a mere strawman. But the more serious casualty of this way of looking at the Dialogues is not Cleanthes but rather the quality of the confrontation between him and Philo that Hume so “artfully” managed to achieve.

In an earlier article I tried to show that Cleanthes is a more considerable philosopher than some commentators give him credit for.¹³ But for the present I need only say that there is no finer testimony to the quality of the confrontation between his two adversaries that Hume was able to sustain throughout the Dialogues than the fact that astute commentators continue to debate to this day the extent to which Hume's views on religion should be identified with those of Cleanthes.

I will now give two illustrations of the kind of dramatic hash that is made out of the Philo/Cleanthes confrontation by commentators who downgrade Cleanthes' contribution. My first example is from Kemp Smith's work. Perhaps no commentary has contributed more to enhancing our understanding of Hume's Dialogues than the one Kemp Smith supplied in his irreplaceable edition, but Kemp Smith's incomprehension of Cleanthes' intellectual abilities is the one regrettable consequence of the zeal with which he went about the task of proving that Philo is Hume's chief spokesman. According to Kemp Smith, Cleanthes, no less than Demea, is simply a pawn in the hands of Philo, but the “dramatic balance” of the Dialogues required that “Cleanthes must never be represented as having been definitely and finally refuted.”¹⁴ Then follows an account of the exclusively literary expedients employed by Hume “to preserve Cleanthes' dignity and to cover over his failure to make any effective reply.” But I find that account most unconvincing. Is it seriously possible to believe that in a report of philosophical debate, the natural imbalance resulting from a grossly unequal matching of adversaries, can be redressed simply by allowing a show of exaggerated civility and deference toward the weaker party? We know quite well what Hume would have thought of a philosophical dialogue that merely used literary ruses to bring about a balance among its speakers. In a letter dated 1751, Hume makes this revealing suggestion;

I have often thought, that the best way of composing a Dialogue, would be for two persons that are of different Opinions about any Question of Importance, to write alternately the different Parts of the discourse, and reply to each other. By this means, that vulgar Error would be avoided, of putting nothing but Nonsense into the Mouth of the Adversary: And at the same time, a Variety of Character and Genius being upheld, would make the whole look more natural and unaffected.¹⁵

To suggest that Cleanthes is a mere strawman, and that, moreover, he is so by Hume's deliberate intention, would be to raise a serious question about Hume's capacity for fair-mindedness, and thus compromise the reputation of his Dialogues for having (in Kemp Smith's own words) subjected “the design argument to a passionless and searching criticism” that was at the same time “final and complete.”¹⁶ But what is of equal importance to notice here is that, if Kemp Smith's premise were valid, the Dialogues would be a failure as a literary venture as well. The true heroes of a dialogue are arguments, and it is only when forceful arguments in it join horns with each other that the true drama unfolds. If one side in it were as heavily weighted to win from the start as Kemp Smith would have us believe it is in Hume's Dialogues, there is simply no way in which an author, however skillful he may be in the use of dramatic devices, could succeed in sustaining the illusion of dramatic balance. Better to leave the imbalance, than to attempt to create a bogus balance that is all style and no substance.

Thus if a commentator can see nothing worthwhile in Cleanthes' case, inflating his contribution cannot restore the balance. But such a commentator might be tempted to say that Hume achieved the wanted balance by using a different strategy, viz., by equivocating on Philo's contribution. This is in effect what happens in Professor Bricke's article to which I referred earlier. But Bricke's account emerges quite incidentally out of what I consider to be a wrongheaded approach to understanding Hume's literary achievement in the Dialogues.

Bricke begins his argument by drawing our attention to “the striking fact” that the Dialogues have been “subject to so many mutually contradictory interpretations”¹⁷ in each one of which one of the participants is singled out for the honor of being Hume's primary spokesman, with no effort spared to show that he holds a consistent and well-argued position. Another “equally striking fact about the Dialogues,” it seems to Bricke,

is the fact that while each of the proposed interpretations has some plausibility, none of them is satisfactory. Each can be defended by the citation of passages from the Dialogues, by reference to Hume's other philosophical writings, by the introduction of remarks from his letters, and by the discovery of corresponding passages in works which we have reason to believe Hume was familiar with. But each interpretation fails to convince because … typically, the defence … involves a very cautious selection of those passages in the Dialogues which suggest the preferred view, together with a vigorous attempt to explain away, in a more or less ad hoc way, the many passages which seem not to fit.¹⁸

I think Bricke is exaggerating somewhat, but that’s not the issue I wish to go into now. The fact remains that much of what he says in the above passage is true of many (perhaps most) interpretations of the Dialogues. And he is correct in thinking that the situation is curious enough to stand in need of explanation, and that it must be something about the Dialogues themselves which calls forth such remarkable diversity in their interpretation. On the way to showing what he believes this feature of the Dialogues to be, Bricke first argues that the fundamental error the various interpretations make is to assume that one of the two main characters in the Dialogues is Hume's primary spokesman. Bricke's demonstration of his view that neither Philo nor Cleanthes (Demea being too obviously a non-candidate) could be speaking for Hume, takes the form of adducing evidence from the text of the Dialogues to show that both (a) are often seriously and, as a matter of fact, blatantly, inconsistent, (b) often hold quite unHumean theses, and (c) commit other lapses which make them unpromising standard-bearers for Hume. I will be returning to this aspect of Bricke's case shortly, but for the moment, I want to continue with the main line of Bricke's argument in his paper.

Once we abandon the vain search for Hume's spokesman in the Dialogues, Bricke goes on, we are in a position to see them for what they truly are: “the Dialogues are, and were intended to be, a work of literature, with a particular rather complicated, literary form.”¹⁹ Bricke quotes the hackneyed passages from Hume's autobiography and other sources to show how anxious Hume was to establish his literary reputation and, in particular, “about the quality of his Dialogues as a piece of literature.”²⁰ He opines, quite correctly, that “Hume appears to have given some quite serious thought to the special difficulties, and special literary requirement, of the dialogue form,” quoting from the narrator's opening speech and the letter of 1751 in confirmation. And Bricke puts forward this suggestion (which I think is the undoing of his thesis):

The principal literary problem to be solved, at least in a philosophical dialogue, is that of combining the naturalness of an ordinary conversation with the precision and more systematic character of a written philosophical argument.²¹

Bricke, of course, thinks that in making this suggestion he has put his finger on the key to explaining what is really going on in the Dialogues, and, given his view of the matter, it must be confessed his solution is a rather neat one. We have only to remind ourselves, says Bricke, about what happens in the course of a “real life” philosophical conversation (“even at rather lofty levels”²²) to appreciate Hume's artistic achievement in reproducing his imaginary conversation. Once we see what Hume was getting at,

it should come as no surprise that the argument takes a somewhat erratic course, that issues are raised which are not treated exhaustively, that the most fundamental assumptions of the participants do not become perfectly explicit … In short, once one takes seriously the literary criteria Hume employed in writing the Dialogues, the inconsistencies and other lapses of both Philo and Cleanthes can be taken at face value, and need not be explained away. This, I submit, is confirmation of my claim that one must look to their literary form if one is to understand the Dialogues.²³

The story improves with the inclusion of the requirement of philosophical balance (though I don’t know if Bricke regards it as a separate requirement or thinks his way of understanding the naturalness requirement entails it). “Hume insists that the participants in the discussion be properly balanced” and “that no one … have a clear-cut advantage over the others. Each will, presumably, make a reasonably substantial contribution to the discussion, but each will also make his mistakes.”²⁴ Bricke considers it to be a further confirmation of his view of the Dialogues that the two requirements he mentions when taken together would explain how it comes about that commentators who “fail to grasp Hume's literary purposes, and make the seemingly natural assumption that one of the characters must be Hume's spokesman,” end by giving “a series of plausible though contradictory readings of the text, together with unconvincing attempts to explain away what does not fit.”²⁵

The seductiveness of Bricke's viewpoint on the Dialogues, especially for those who have “had it” with the numerous interpretations that have appeared at various short intervals ever since their first publication, cannot be underestimated. But for this very reason it cannot be emphasized enough that, despite its reasonable sounding appearance, Bricke's view is treading on dangerously treacherous ground and to take it seriously would be to totally subvert the philosophical integrity of the Dialogues and in the process undermine their literary integrity as well. As I have already indicated, I believe Bricke is mistaken in thinking that the naturalness requirement, as understood by him, is what sets the chief problem to be solved by the writer of a philosophical dialogue. I do not wish to underestimate the importance of conveying the special flavor of an actual conversation in a dialogue, but the fact remains that this is not the deepest level at which a philosophical dialogue can affect us. But, furthermore, Bricke simply misrepresents what the naturalness requirement entails even at this comparatively superficial level, thus compounding his error. I agree that it was not Hume's intention to write a dialogue in which the participants never stumble, never hold inconsistent views, where, that is, there are “no false starts, no blind alleys.”²⁶ But surely the naturalness criterion does not require the participants in a philosophical dialogue to commit anywhere near the number and variety of blatant contradictions, palpable errors in argument and other mindless lapses that Bricke's theory requires to be attributed both to Philo and Cleanthes. And let there be no mistake about this: the list of Philo's and Cleanthes' shortcomings as philosophers that Bricke ticks off is depressingly long and formidable.

Cleanthes emerges a good bit like he did in the portrait of him we found in Kemp Smith: “Despite his occasional acuity, Cleanthes does not seem to be a very good philosopher.” Where contradictions and unHumean thesis are concerned, however, he seems to have a distinct edge over Philo (though he is said to be guilty of committing one big discrepancy with Hume's outlook in supporting the design argument.²⁷)

The much longer account of Philo's failures is not so easily summed up. At various times he is caught “foolishly using an inappropriate argument,” being “not as careful and reflective as he might be” and at one time a piece of evidence is brought forward as “further indication of Philo's weaknesses as a philosopher.”²⁸ And the list of Philo's inconsistencies takes four full pages²⁹ to relate (whereas Cleanthes' needed hardly a paragraph). Every view that ever any commentator, however farfetched, saddled Philo with is paraded. Philo is said at times to follow “a confident and straightforwardly theistic way,” but at other times his “theism is neither confident nor straightforward nor unequivocal,” and at still other times “most minimal and equivocal”; like a scatter-brain, he is said to move from “the direction of deism” to speak “in more naturalistic ways,” not to mention his many skeptical moods, both pyrrhonist and moderate. And we are asked to take this “textual evidence” at “face-value” and “adopt the straightforward course of recognizing that Philo is, and does not merely appear to be, seriously inconsistent in his views,”³⁰ all in the interest of enhancing Hume's literary reputation!

It’s difficult to tell, once Bricke is through with them, who comes off worse—the mediocre Cleanthes or the muddle-headed Philo. But Bricke apparently thinks he has shown how well-matched the two adversaries are, and thereby hangs a tale. The truth is, of course, that (though he does not explicitly say so) Bricke believes that Philo also has an obviously disproportionately larger share of “the most persuasive and compelling arguments” (that is, the sort that at his article's end Bricke agrees “are probably to be ascribed to Hume”³¹) at the most crucial points in his long (and given Bricke's premise) desultory debate with Cleanthes. So, of course, it is Philo who must be hobbled by inconsistencies and sundry bad arguments in order to even out the score. Then the often muddle-headed but sometimes brilliant Philo can be regarded as an equal match for the mostly mediocre yet usually consistent Cleanthes.

I hardly need to spell out why, had Hume adopted the scenario for the Dialogues that Bricke's paper suggests he did, they would not be the superb intellectual and artistic achievement that they actually are. It is enough to say that simply the problem of making the intellectual characters of the two protagonists, but especially Philo, believable to the reader, would have been insurmountable. (Surely, making its characters plausible as an important part of what Hume means when he requires that a dialogue be natural.) In Bricke's account both protagonists are now psychological paradoxes, but Philo has been turned into a veritable monster—an intellectual chimera, one part nit-wit, the other genius. But an even more important consideration is this: The chemistry of the total aesthetic pleasure one derives from a great philosophical dialogue may not be easy to analyze, but the chief element in it is, to use Aristotle's famous phrase, the “proper pleasure” of the sustained intellectual activity in which such a dialogue engages the reader. A philosophical dialogue is not what we may call the work of “pure literature”; it is not “expression which satisfies us simply by existing as expression.”³² In a philosophical dialogue no clear-cut distinction is possible between expression and content, between the style and substance, as it is, for example, in the Ode on a Grecian Urn. So there is an apparent paradox in Bricke's account of Hume's Dialogues: In attempting to highlight the literary qualities of the Dialogues by a consideration of it as purely “a work of literature” (albeit “in which problems of great philosophical complexity are discussed”), he distorts our perception of those very qualities which make it a great work of literature of the genre philosophical dialogue.

III

What I will now do will amount to giving a general direction as to how the Dialogues ought to be read so that most of the apparent inconsistencies that Brick and other commentators find in them are either resolved or dissolved.

Let me begin with the obvious comment that philosophical balance is not something one can achieve in a dialogue once and for all. Rather, it is something that must be continuously exemplified in the course of the discussion, and the process in which the balance is exemplified I call the dialectic. Though the term is not used by Hume, by the dialectic I mean just that quality of the dialogue form of writing which can be described by the use of terms that are the exact opposite of the terms that the narrator uses to describe what he characterizes as “the methodical and didactic manner” of writing philosophy. The latter is described as “the direct style of composition” where the author “can immediately, without preparation, explain the point at which he aims; and thence proceed, without interruption, to deduce the proofs, on which it is established” (127). By contrast, the dialogue is indirect and non-immediate, constantly drawing the reader into the process (“preparations and transitions”) that is involved in a gradual and progressive refining and maturing of positions in dialectical confrontation with each other.

But Pamphilus' implicit warning to the contrary notwithstanding, the debate he relates has much too often been read as if it constituted a straight-forward piece of philosophical writing. Such a reading assumes that Philo and Cleanthes each have a position to defend which they state at some point early in the dialogue and thereafter what the two say to each other in defense of their respective views makes no, or at any rate a negligible, difference in their respective viewpoints. Little give and take and no dialectical interplay between their arguments is conceded. Instead of portraying the discussion as being impelled forward by that magical chemistry of debate so essential to the success of the dialogue form, as opposed to what Pamphilus calls the “methodical and didactic” manner of writing, we are treated to the spectacle of two ossified positions being obstinately defended by their protagonists by whatever means they can command. It is no wonder, then, that the received wisdom has it that the goings on in the critical final part of the Dialogues are thrust upon us unexpectedly in a most abrupt manner.

The only person in the Dialogues whose views do not undergo any significant change is Demea, and this is perfectly in tune with his “rigid orthodoxy.” But Philo and Cleanthes both begin with more extreme positions than they end up with. Like any other good debaters, each concedes some points under the impact of the criticisms from the other, and by the time they finish, their views have begun to converge.

First Philo. His opening position is an extreme pyrrhonism but it is not long before, smarting under Cleanthes' taunts, he reveals himself to be a mitigated skeptic and agrees to “argue with Cleanthes in his own way” (145). Cleanthes' way is, of course, to apply the ordinary inductive reasoning of common life and experience to theology via the design argument. Philo begins his counter-attack in Part II whose conclusion is that we cannot know anything about the existence of the cause of the design in the world (without resorting to faith), on grounds that we have no experience of such a cause. In the crucial Part III Cleanthes' rebuttal is said to confound Philo and, whether it really does so or not (I think it does³³), Philo thenceforth falls back on his second line of defense, viz., that we can know nothing about the nature of such a cause. We cannot prove its perfection, infinity or unity; nor even that it is transcendent and personal. Philo does, however, show some partiality for the view “which ascribes an eternal, inherent principle of order to the world” (174). And he is not backward in acknowledging that the four principles he canvasses as possible sources of order in the universe, viz., “reason, instinct, generation, vegetation … are similar to each other, and are the causes of similar effects” (178). This brings us up to Part VII. In part VIII Philo presents his remarkable development of the “Epicurean hypothesis” “justly esteemed the most absurd system, that has yet been proposed” (182). I will pass up Part IX in which Demea's a priori argument for God is rejected by both Cleanthes and Philo. Philo's celebrated discussion of “the problem of evil” in Parts X and XI is meant to show how crucial he thinks this issue is to the possibility of supporting “natural religion” by common sense inductive methods of proof and how disastrous for such a religion are the consequences of Cleanthes' inability to give a rational account of God's moral nature.

It is even less well understood that there is a progressive development in Cleanthes' viewpoint as well. Cleanthes begins very ambitiously and there is little doubt that he thinks he can prove the truth of traditional theism by his argument from design. His opening statement of the design argument is in the form of what may be called a “direct inductive argument.” Without mincing his words, he claims that the world is “nothing but one great machine” and that “the curious adapting of means to ends, throughout all nature, resemble exactly, though it much exceeds, the production of human contrivance” (143; italics added). It is only a bit later, however, that he is forced to concede that the resemblance he claims is far from being “exact,” and speaking strictly, it is at this point that his argument becomes an argument from analogy. He continues to argue, nonetheless, that the resemblance is still strong enough for the conclusion of the design argument to appeal to a reasonable person (144-145). Cleanthes' theism is gradually emaciated in the course of his discussion with Philo but in the process he has exacted a price for this from Philo. Philo has not succeeded in doing without the notion of an ultimate determining source of order in the world. Cleanthes has some right to gloat over the fact that (as he says addressing Philo):

by the utmost indulgence of your imagination, you never get rid of the hypothesis of design in the universe; but are obligated at every turn, to have recourse to it. To this concession I adhere steadily; and this I regard as a sufficient foundation for religion.

(169)

But whether this belief in a source of order with which both Philo and Cleanthes go into Part XII, can be a sufficient foundation for religion, is the topic of debate in this final dialogue. Now there is no sense in denying the obvious ambiguities in Philo's “summing up” in the final dialogue. But, as I have already indicated, I believe that Kemp Smith's expert detective work succeeds in showing that Philo speaks for Hume and that Kemp Smith is right also in his estimate of Philo/Hume's final position:

If … belief in God be taken as consisting in the assertion that something not altogether different from human intelligence is the ultimate determining source of order in the universe, Hume is not questioning this belief, but is merely insisting that it is a more ambiguous, less definite assertion than is commonly supposed.³⁴

But whether we accept Kemp Smith's estimate or not,³⁵ one thing must be insisted upon. Whatever the final view we attribute to Hume, it must emerge from a reading of the last dialogue and it must not be “disconnected” with what has preceded it. Nothing less would suffice if we think Hume was a great dialogist, which he indeed was. But when we turn to the various commentaries what do we find? Let me just confine myself to Philo, since the most seemingly intractable disagreements center around his view. Typically, commentators, lighting upon some thesis stated by Philo in one of the earlier parts of the Dialogues and having made up their minds that that is Philo's considered view, try to read the rest of what he says in the light of that view. Thus, for example, one says that he is a pyrrhonist (Noxon³⁶), another that he is a moderate skeptic who is a total agnostic about God's existence and nature (Wollheim³⁷ and numerous textbooks), and still another argues that he believes in some kind of immanent God (Nathan³⁸). I have even read and heard it said that Philo accepts the possibility of the world-order having arisen by chance out of original chaos (that is, the “Epicurean hypothesis”). The result is all those ad hoc attempts which Bricke speaks of, in such commentaries to explain away Philo's “inconsistencies.” But Bricke goes one better than the commentators he criticizes. According to him all these “views,” which when placed side by side and accepted all at once would result in quite glaring inconsistencies, are to be attributed to Philo!

I suggest that the most straightforward way of resolving the problem of interpretation that the protean nature of Hume's Dialogues presents, is simply to emphasize the obvious, namely, that they are a dialectical work in which the view of each of the main protagonists is unfolded in stages under the impact of the criticism levelled against him by the other. When Hume decided in his mature years, “To deliver a system in conversation,” pitting two fully developed philosophical outlooks against each other, each defended to the death by its proponent, was the last thing he had in mind. On the contrary, he used all his considerable skills as a philosopher and writer to produce a work in which the natural rhythms of a dialogue are allowed to dictate its contents in a way unequalled in any philosophical dialogue written since antiquity. The constantly shifting intellectual scene, the ebb and flow of thought, the frequent use of rhetorical or argumentative devices that we find in the Dialogues, are the very stuff of what its narrator calls “preparations and transitions” by means of which we are gradually led to the denouement in the final dialogue. To view the Dialogues in this light is to see at once the folly of imposing a false ideal of consistency upon them. Rather, we must now speak of earlier positions being abandoned for newer ones, and of views stated and explored but not necessarily espoused. Even so, some inconsistencies and ambiguities will remain, but most of these will be interesting ones either because they are endemic to the position being explored or because Hume was himself not able to resolve them in his own mind.

Notes

1. I wish to thank the Rutgers' Research Council for a grant that assisted me in completing this article.

2. David Hume, A Treatise of Human Nature, ed. L. A. Selby-Bigge (Oxford: Clarendon, 1888), 271.

3. Ernest Campbell Mossner, “Philosophy and Biography: The Case of David Hume” in Hume: A Collection of Critical Essays, ed. V. C. Chappell (New York: Doubleday Anchor, 1966), 6-34.

4. David Hume, My Own Life, in Dialogues Concerning Natural Religion, ed. Norman Kemp Smith (Indianapolis: Bobbs-Merrill, 1947). All future references to the Dialogues will be to this edition and will be incorporated in the text.

5. Mossner, 7.

6. Terence Penelhum, Hume (New York: St. Martin's Press, 1975), 180.

7. John Bricke, “On The Interpretation of Hume's Dialogues,” Religious Studies 2 (1976): 3.

8. Bricke, 16.

9. That there may be (and I believe there is) an element of deliberate irony here—Hume is a shade overstating a point in order to escape the consequence of being too closely associated with one of the developing positions in his Dialogues—in no wise detracts from what I am now saying. The irony is squarely implanted in the midst of a truth. One might say of the relationship between irony and truth what Mr. Nigel Dennis has said of satire and religion: “they have lived in adjoining cells ever since civilization began: the ravens have not brought to the [one] a single loaf or fish of which the [other] has not eaten half” (Nigel Dennis, Two Plays and a Preface (New York: The Vanguard Press, 1958)).

10. The Letters of David Hume, ed. J. Y. T. Greig (Oxford: Oxford University Press, 1932), I: 154.

11. The fact that in Hume's Dialogues it is Philo who is bucking the tide of received opinion would explain in part at least why he has a disproportionately long part when compared to that given his adversary. Hume can, and does, rightly assume that his readers will all be a good deal more familiar with the position Cleanthes advocates, than the attack that Philo launches against it. Whatever else one may say of the confrontation between Cleanthes and Philo, there is no denying the fact that most of what is original in these Dialogues is put in the mouth of Philo. (Cleanthes himself admits as much when he pays his rival the left-handed compliment of saying that his “greatest errors proceed not from barrenness of thought and invention, but from too luxuriant a fertility,” Dialogues 155.) For some interesting speculations as to which philosophers Cleanthes and Demea might have been modelled after, see E. C. Mossner, “The Enigma of Hume,” Mind n.s. 45 (1936): 334-349; R. H. Hurlbutt, “Hume and Scientific Theism,” Journal of the History of Ideas 17 (1956); R. H. Hurlbutt, Hume, Newton, and the Design Argument (Lincoln, University of Nebraska Press, 1956); Andres Jeffners, Butler and Hume on Religion, tr. K. Bradfield (Stockholm: Diakonistyrelsens, Bokforlag, 1966).

12. See A. G. N. Flew's article on “Hume,” in A Critical History of Western Philosophy, ed. D. J. O’Connor (Glencoe: The Free Press, 1964), 266.

13. P. S. Wadia, “Philo Confounded,” in McGill Hume Studies: Papers Read at the Bicentennial Conference, eds. Norton, Robinson, and Capaldi (forthcoming).

14. All quotations in this paragraph are from Kemp Smith's “Introduction,” 61-65.

15. Greig, Letters.

16. Kemp Smith's “Introduction,” 30. The explanation for Kemp Smith's curiously exaggerated denigration of Cleanthes' intellectual abilities is simple enough. Kemp Smith's chief contention in his commentary is that it is Philo who is Hume's principle spokesman and that he holds throughout the Dialogues a consistent and well-argued position against the possibility of a full-blown natural religion. Kemp Smith evidently felt that in arguing for this contention he was going against the current of the received opinion of the time and, in the circumstances, it is not surprising that he tended to overplay his hand. One result is that Kemp Smith can see no merit at all in anything that Cleanthes has to say in defense of the design argument, thus in effect removing from the field the only other plausible contender for the role of Hume's primary spokesman among the participants in the Dialogues. Kemp Smith acknowledges, of course, that Cleanthes displays remarkable force and clarity in demolishing Demea's a priori argument for God's existence at Part IX. But (although not a single example is cited of a doctrinal inconsistency committed by Cleanthes) all this proves to Kemp Smith is that Cleanthes is both “psychologically and intellectually” self-inconsistent! (Cleanthes, in other words, is a psychological paradox: a giant intellect at one place, a mere mite at others.)

17. Bricke, 1.

18. Bricke, 1-2.

19. Bricke, 1-2.

20. Bricke, 14.

21. Bricke, 14.

22. Bricke, 16.

23. Bricke, 16.

24. Bricke, 16.

25. Bricke, 16.

26. Bricke, 16.

27. Bricke, 12-13.

28. Bricke, 4-5.

29. Bricke, 5-8.

30. I don’t include a mention of the respects in which Philo is said to be unHumean for want of space. But the so called unHumean theses that Bricke attributes to Philo—for example, that we can know something exists without quite knowing what kind of thing it is—are such only in the sense that they don’t belong to the traditional textbook picture of Hume, but which are to be found in his work. There is no reason why we should expect Philo to be more “Humean” than Hume himself.

31. Bricke, 17.

32. Lascelles Abercrombie, “Principles of Literary Criticism” in An Outline of Modern Knowledge, ed. William Rose (London: Gollancz, 1931), 865.

33. Wadia, “Philo Confounded.”

34. Kemp Smith's “Introduction,” 36-37.

35. The chief defect in Kemp Smith's account of Philo's final views is that he sometimes writes as if he thought that these views are thrust upon us unexpectedly in the final dialogue (see particularly Kemp Smith's Introduction, 68-69).

36. James Noxon, “Hume's Agnosticism,” The Philosophical Review.

37. Richard Wollheim, ed., Hume on Religion, (New York: World Publishing Co., 1963). “Introduction.”

38. George J. Nathan, “Hume's Immanent God,” in Hume, ed. Chappell, 496-523.