Hume’s Tacit Atheism (1975)
A recent paper, ‘Hume’s Immanent God’, (in (1)* by George Nathan, contains an insightful interpretation of Hume’s Dialogues Concerning Natural Religion (henceforth, briefly, D). Insight is no guarantee against error. I shall argue that Nathan’s interpretation is mistaken, and then offer my own. Nathan observes that the general tendency in scholarship on D has been to focus on its sceptical side. He proposes to ‘bring out Hume’s positive contribution’. Nathan’s thesis, briefly, is that D best supports a modestly theistic interpretation according to which God is the ultimate cause of order in the universe. The critical side of D, he maintains, is meant only to show that this ultimate cause is immanent, rather than transcendent, and that we can have no positive knowledge of the nature of this being, though we can be reasonably sure of its existence. Nathan’s evidence is as follows (a) Philo never denies that the question for discussion concerns not the being but the nature of God. (b) Philo does not deny that the universe resembles an artifact in having ‘an accurate adjustment of the parts to each other’ and ‘an adaption of means to ends’. Philo criticizes Cleanthes’ argument solely on the ground that, like an artifact which it resembles, the cause of its ordering must be external to the effect, (c) Philo does not deny that the universe exhibits rational order because he is prepared to extend the concepts of reason and rationality to allow that the cause of order need not perceive the order which it produces, (d) The inference to a transcendent designer is unwarranted according to Philo because, if rational order requires an external cause, then an external cause is required for that external cause, and so on, ad infinitum. (e) Philo’s confession of faith in part XII would be strange only if he had ever denied that there was a striking order in nature. ‘However’, Nathan says, ‘he has gone to great lengths not to deny such organisation, but only to explain it by an internal principle’. ((1), p. 417) (f) Philo wonders whether it is proper to call this cause of order ‘a mind’ or ‘intelligence’ and decides that to avoid verbal controversies, one might best describe it as ‘mind or thought’ which Nathan describes as ‘a rational organising factor which might organise either ideas or matter’, ((1), p. 419)
* Parenthesised numbers in running footnotes refer to numbered listings in the Bibliography to be found at the end of the paper.
‘… since this internal rational cause is the ultimate explanation for all order in the universe, it is, in a sense, also entitled to be called God. This God has only the remotest connection with the one traditionally conceived … By Nature we mean not the sum total of things, but rather the dynamic internal structuring principle in the universe. Since Philo’s God or “mind” is such a force, the identification of God with nature is certainly intended by Hume.’ ((1), p. 421)
As attractive and pleasing as the above interpretation may seem, I intend to show that all is far from well. The reader may think he understands what kind of God it is that Nathan is depicting Hume as acknowledging. I, for one, fear that I do not understand Nathan on this central point and I shall show how any initial flush of approval that one may feel for Nathan’s interpretation is due to a blurred impression created in the reader by some fast shuffling on Nathan’s part. It is simply not clear after granting all of Nathan’s premisses, that Hume is a theist in any appropriate sense of the word. One wants to say ‘Nathan has shown that Hume believes in something but….’ What kind of God is Hume’s ‘immanent God’? I see only a few possibilities. First, given Nathan’s interpretation, consider what one might see in the notion that there is a principle of order in nature which is inherent and original. One possible meaning of ‘internal principle’ is: since something (order) exists, something else (the internal principle) necessarily exists as its basis or ‘ground’. One recognises the latter as the general form of the cosmological argument. If correct, such an interpretation would lead one to conclude that Philo is a theist. According to Nathan, Philo, in arguing for an internal principle of order, ‘is saying that, not only is it a sufficient reason, it is a necessary one’. ((1), p. 412)
‘To prevent the regress in which the order of the universe is explained by an external cause … it is necessary that there be an internal principle of order somewhere in the series. If the internal principle is necessary, then an external principle is impossible … Therefore the universe has within it a necessary and sufficient explanation for its order.’ ((1), p. 412, 413)
One issue, as I see it, is over Necessity. Let N = ‘necessarily’, p = ‘there is order in the universe’ and q = ‘there is a principle internal to the universe which accounts for its order’. What does Nathan say at this point? It is not clear exactly what the form of the argument is that Nathan is attributing to Hume. There are three possibilities
A p → Nq
If it is A above then we are dealing with a valid argument. The conclusion of A, however, violates a central principle of Hume’s ontology, namely, every matter of fact and existence is contingent. Suppose, then that it is not A but rather
B N(p → q)
If so, we are dealing with an argument which is both invalid and whose conclusion violates Hume’s ontology. Suppose, thirdly, then, that it is
C N(p → q)
This argument is valid, but its conclusion expresses a contingent fact. It seems to me that we have no choice here other than to assume that Hume was neither in the habit of committing elementary blunders in deduction nor in the habit of forgetting central principles in his own ontology. The danger here is that, if one is content with the idea that Hume was a theist, one may unwittingly come to think that Hume at some point accepted the notion of a necessary being, in spite of the condemnation the latter receives in D and elsewhere. I believe we have no reason to suppose that Hume accepted anything stronger than the hypothetical de dicto necessity, ‘N(p → q)’, and the validity of C. Nowhere is there adequate evidence to suppose that Hume accepted the de re necessity ‘Nq’. I do not charge Nathan with explicitly asserting that Hume did accept the latter. But he unfortunately does not caution the reader against it. His language and general conclusion both encourage it. Hume’s immanent God can’t be the necessary being of the ontological argument (ens a se, causa sui.).
There is another danger concerning necessity. This concerns the proposition that there is an internal principle of order in the universe in the sense of ‘naturally inherent in and inseparable from’. Interpreting Philo’s argument, Nathan says:
‘(We may) suppose that God’s mind is a rational faculty and that this is the cause of order. But, again, if this were a valid explanation for God’s mind, it is also valid for the material universe. We could say that order pertained to its very nature … The use of such explanations remains a sure sign of our ignorance of ultimate causes.’ ((1), p. 410)
The Scholastics sought explanations in appealing to such entities as substantial forms, essences, entelechies, and souls of the Aristotelian variety. They believed in metaphysical de re necessities. If one is not careful, it would be easy, under Nathan’s influence, to get the impression that Hume believed that there is a principle of order which inheres naturally in the material world with something like the status of these scholastic entities. I believe that Hume instead meant to show the futility of appealing to internal principles of the (scholastic) metaphysical kind. That is not to say that Hume rejected the notion of an internal principle in every sense, but surely the author who condemned volumes of divinity and school metaphysics to the flames would not have easily taken them to be metaphysical entities to which ‘substances’ have some necessary relationship. Hume’s ‘immanent God’ cannot be a scholastic metaphysical entity such as, e.g. ‘animal soul’. It is thus clear what internal principles can not be for Hume. As Philo warns Cleanthes:
‘It was usual with the Periapatetics … when the cause of any phenomenon was demanded to have recourse to their faculties or occult qualities and to say that bread nourished by its nutritive faculty and senna purged by its purgative: But it has been discovered that this subterfuge was nothing but the disguise of ignorance.’ ((4), p. 162)
What sort of God is this immanent God then? I see only one other possibility but it is a disturbing one. The clue is the connection between ‘internal principle’ and ‘God’. We find Philo using the notion of an internal principle of order in reference to matters of ordinary experience. Animals and plants continually reproduce without intentional design on their part. In other words Philo is saying we have a concept, of which we make constant use, according to which it is the animals (plants) themselves which produce offspring without the aid of an intermediary agency. It is therefore obvious that, according to this concept, there is something which is part of the organism itself, by means of which it confers order on its offspring. Granted that Philo is not referring to anything metaphysical, it seems most reasonable to construe internal principles as unobserved causal factors in the structure of the organism. Even with regard to the orderliness of our own thought-processes, Hume says that:
‘the experimental reasoning which we possess in common with the beasts, … is nothing but a species of instinct or mechanical power that acts in us unknown to ourselves.’ ((3), p. 118)
There are many reasons why it seems highly unlikely that Hume, in defending the concept that Nature has principles of order internal to itself, was endorsing a metaphysical, theistic concept. Consider the idea that there is a metaphysical entity which is the cause of every event in Nature. The theistic consequences of Cleanthes’ Design argument bring one close to this Malebranchian view. And it is precisely Malebranche that Hume has in mind when he formulates a devastating critique of the doctrine of ‘unknown causes’ in the Enquiry Concerning Human Understanding, (pp. 75-79) The main force of this critique is; we have no grounds on which to infer the existence of unknown (supernatural) causes; it is epistemically pointless to appeal to them because there is nothing about them that explains anything. We have no better idea of how minds cause events than we have in the case of how one physical event causes another. The doctrine of supernatural causes multiplies mysteries while explaining nothing.
In light of this part of the Enquiry it would seem that Philo’s position is surely not reducible to the claim that theistic explanation is useless unless we take God to be immanent (rather than transcendent) in which case it is quite acceptable. Let us now pick up the main thread of the argument.
In view of Hume’s general ontological beliefs, it seems most implausible to suggest that Hume was content with any such concepts as necessary being, metaphysically grounded order, or that he endorsed a wholesale use of metaphysical explanations at any point. Hence, if we remain on our guard against such an interpretation, (though Nathan’s language insinuates it) what else could Hume have meant by internal principles of order? The answer must be given in steps.
The paradigms of ‘springs and principles’ constantly referred to in D are Reason, Instinct, Generation, and Vegetation. Generation and Vegetation are described as the causes of animals and vegetables, Demea asks:
‘But what is this generation and vegetation of which you talk? Can you explain their operations and anatomize that fine internal structure on which they depend? ((4), pp. 177-8)
‘At least as much as Cleanthes can explain the operations of reason, or anatomize that internal structure on which it depends … These words, generation, reason, mark only certain powers or energies in nature whose effects are known but whose essence is incomprehensible … ((4), p. 178)
In this little corner of the world alone there are four principles … which are similar to each other and are the causes of similar effects. What a number of other principles could we naturally suppose in the immense extent and variety of the universe … Any one of these … may afford us a theory …’ ((4), p. 178)
Reason and instinct are the principles by which minds operate. Generation and vegetation are the principles by which living bodies operate. Each are ‘powers’, ‘energies’, or ’causes’ of these natural phenomena. To explain them would be to ‘anatomize’ the ‘internal structure or machinery’ on which they depend. On my reading of Hume, to explain generation and vegetation would be to find the presently unobserved internal structures of organisms which are responsible for reproduction, nutrition, growth, etc., describe them and identify their causal properties. To explain reason and instinct would be to discover the structures in conscious organisms whose causal interrelations account for (e.g.) love, hatred, nest-building, thinking. My interpretation at this point is that, in one sense, for Hume, a principle is any configuration of lawfully related internal structures of empirically distinct kinds of natural beings, which structures, for the science of Hume’s day, were not observationally known. By an empirically distinct kind of natural being, I mean any system in nature which characteristically exhibits m (or m-n) kinds of behavior (or change) in k kinds of environmental circumstances. Thus, human beings think (out loud), feel, take nourishment, reproduce and grow. Amoebae exhibit fewer kinds of behavior (changes). Rosebushes even fewer. The idea is roughly expressed but will suffice for our purposes. Principles are what might be called relatively theoretical entities: ‘relatively’ in the sense that their nature was still a matter of conjecture for the science of the time. Recently, for example, the nature of genes has presumably been identified as DNA molecules in the chromosomes. When the nature of once relatively theoretical entities has been described, there is, for Hume, a sense in which we are left with as deep a mystery as before. We uncover a set of uniformities. These explain phenomena at a level previously lacking an explanation. But what is the explanation of these new uniformities? We may seek the answer to this question by postulating a still ‘finer’ theoretical network, and perhaps later identify the nature of still another set of theoretical entities. The same questions would seem bound to rise again. Can there be an end to this pattern? Hume’s answer would appear to be that as long as we are dealing with contingent empirical uniformities, there cannot be a ‘metaphysically satisfying’ end to it. In the sense just outlined, a principle in a natural system inheres in that system as genes are in organisms. But in still another sense, there must be ‘principles’ inherent in the world for which there are no further explanations. Philo says:
‘And were I to defend any particular system of this nature, I esteem none more plausible than that which ascribes an eternal inherent principle of order to the world … This at once solves all difficulties; And if the solution… is not entirely complete and satisfactory, it is, at least, a theory that we must sooner or later embrace. How could things have been as they are, were there not an original inherent principle somewhere in thought or matter?’ ((4), p. 174)
Explanation must come to an end somewhere. One claim that Hume is making here is that, no matter how fine our theoretical explanatory network, we are dealing with relations of cause and effect, and that relation is mysterious at bottom. Here is a point at which it seems to me that Nathan’s thesis is radically mistaken. He says that Philo neither denies that there is order in nature nor that there are ultimate, original principles of order. All that he denies, according to Nathan, is that these principles are external to the material world. Nathan concludes that, according to Philo, these principles of order are immanent and, as the explanation of all order, the sum of them considered as a cause is, ‘in a sense’ entitled to be called God.
One can stipulate senses for any expression he likes. This one misses the whole point of what Hume is driving at. The real force of Philo’s argument is that theistic explanation of any kind is bound to be worthless. The level at which explanation comes to an end may well be a level at which we still have complex physical relationships. And if, as causal phenomena, they are in a sense mysterious, this is as it was bound to be. The mystery would not be any less if the level of explanation’s end were a system of divine ideas. To go anywhere beyond material entities is useless. If we hypothesise anything nonmaterial as a first cause, it does not matter whether one supposes this nonmaterial entity to be immanent or transcendent. If it can be distinguished from matter and its qualities, then it is ontologically distinct on Hume’s atomistic criteria, and, if nonmaterial, might just as well be transcendental. Eventually, we must be left with brute facts, no matter what hypothesis we choose. Theism proliferates entities without necessity.
Philo envisions an even more radical possibility. After ‘inviolable laws’ on p. 174 of (4), he suddenly interjects a straightforward and out-of-place assertion.
‘And were the inmost essence of things laid open to us, we should discover a scene, of which at present, we can have no idea. 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.’ (My italics)
In this remarkable passage, Philo has just envisioned what amounts to a variation on the ‘brute fact’ possibility just described. Given a humean sense of ‘absolute impossibility’, what he has conjectured is that the ultimate principles in nature may turn out to be both concrete and a priori. Physical laws which are logical truths! This is nothing less than the Cartesian dream of ideal science; but it is something more as well. Descartes accepts a cosmological argument for the existence of the world and the origin of motion. The present hypothesis is to be seen in the light of (a) Philo’s earlier reference to the possibility of an ‘eternal inherent principle of order (in) the world, though attended with great and continual revolutions and alterations’ ((4), p. 174) and (b) the fact that Hume rejects the cosmological argument.
It may be said that my interpretation leads back to the conclusion that Hume did accept the notion of a necessary being. The substance of the charge does not follow. Cleanthes in effect says: if the notion of a necessary-being made sense, then the material universe might just as well be the necessary being as God. But he and Philo both reject the notion of necessary existence as meaningless. So the hypothesis according to which the ultimate principles might constitute an ‘algebra of nature’ (Hume’s phrase) would have it that nature is described by a hierarchy of laws, some providing explanations of others, and such that the order they describe is ultimately based on a priori principles. Natural organisation or ‘design’ would turn out to be something derivatively necessary in the sense that it is explainable in terms of a set of laws which ultimately flow deductively from the a priori principles which describe the ‘inmost essence’ of matter. Was not Hume haunted by this dream when he makes Philo say
‘It is observed by arithmeticians that the products of 9 are always either 9 or some lesser product of 9; if you add together all the characters of which any of the former products is composed. To a superficial observer, so wonderful a regularity may be admired either as the effect of chance or design; but a skillful algebraist immediately concludes it to be the work of necessity. Is it not probable, I ask, that the whole economy of the universe is conducted by a like necessity, though no human algebra can furnish a key which solves the difficulty? And, instead of admiring the order of natural beings, may it not happen that could we penetrate into the intimate nature of bodies, we should clearly see why it was absolutely impossible that they could never admit of any other disposition.’ ((4), p. 191)
Here we have another sense of ‘internal or inherent’ principles of order. Natural objects may turn out to ultimately consist of entities whose properties are such that principles truly relating them are a priori truths. Metaphysical necessities replaced by Humean necessities. Note: Philo says ‘the intimate nature of bodies‘. Not something co-existing in the physical realm. That is why Philo says:
‘So dangerous is it to introduce the idea of necessity into the present question! And so naturally does it affort an inference directly opposite to the religious hypothesis.’ ((4), p. 191)
Matter may contain the explanation of its own organisation, i.e., it may turn out to consist of entities that ground Humean a priori truths. But even if things should turn out so, even if nature as a totality is an eternal system based upon (physical) a priori principles, Hume can still consistently deny that it (nature) is a necessary being; he can without flying in reason’s face assert it to be a cosmic brute fact. Whether nature’s ultimate principles are contingent or a priori, theism has no case.
I have been taking it for granted that Kemp-Smith (in (4)) and Hendel (in (2)) are right in their arguments that Philo speaks for Hume. I have also been arguing that, given that this is Philo’s role, the overall direction of D supports an atheistic interpretation much better than a theistic one, even the diluted sort of theism that Nathan claims to see in Hume’s words.
When all is said and done, just what is Nathan attributing or not attributing to Hume? Does Philo come down on the side of theism or not? My answer is No, but that his atheism is tacit, subtle, and ironic. Nathan’s answer appears to be Yes, but the sort of deity he describes Philo as worshipping Nathan qualifies (as to its ‘divinity’; and has to so qualify in light of text) to a point at which we may justifiably wonder, what is the difference between his sort of ‘god’ and just no God at all? One is often solemnly assured by undergraduates that ‘Everybody believes in a God of some sort’. They may be right. But, surely, it must at least be logically possible to state the idea of atheism. There must be some position which would count as atheism, even if in fact there are no atheists—otherwise the very proposition, ‘There are no atheists’, would be meaningless as ‘There are no Blibignags’. The general position argued for by Philo is, I think, undeniably atheistic.
A few problems remain for my interpretation. Why does Philo describe the ultimate cause of order in nature in apparently traditional theistic terms? If my interpretation is correct, part of the answer is easy. To the careful reader, Hume has by now made his position subtly but bountifully evident. If one may make a brief ontological comparison with the philosophy of mind, the most extreme materialist, once he has put all his cards on the table, can refer, with equanimity, to ‘the mind’. One who understands what his position is will not be misled by the reference. Just as we know that by ‘the mind’ the materialist is talking about a complicated system of nervous tissues, giving over to it the role once played by the spiritual entity of dualistic metaphysics, so it is clear that a known atheist who appears to seriously make reference to ‘the supreme being’ is ironically referring to the totality of nature in the aspect that completed physico-cosmology would describe to us. The idea of a new player stepping into an old role is clear in both cases. If Nathan takes Hume’s ‘immanent God’ to be something like my materialist’s ‘mind’, I cannot but feel deep dissatisfaction with his interpretation. Surely it should not be called theist in concept.
Two other factors with a clear basis in text appear to confirm a Nathan-like interpretation, (1) Philo characterises the ultimate cause of order in nature as intelligent. (2) Philo settles on ‘mind or thought’ as the most accurate names for the original cause(s) of order. Nathan makes a great deal of these. I must show that there is a reasonable and satisfying way of interpreting these parts of text as something other than signs of a genuine commitment to theism. There are two obvious ways in which ‘intelligent’ and ‘rational’ are used in D. The first is that use in which one describes beings essentially like persons as intelligent. Call this the agential sense of ‘intelligent’. What distinguishes agentially intelligent beings, I shall say, is the fact that the behaviors of these beings which we term actions are expressions of intentions on their part to so act. The second is that sense of the word in which one describes the products of an agentially intelligent being as ‘intelligent’. Thus: ‘European rail transit systems are intelligent’. Call this the artifactual sense of ‘intelligent’.
Clearly: (1) Nothing is artifactually intelligent unless something is agentially intelligent. Given the latter, it might be thought that, (2) The disjunction ‘agential or artifactual’ exhausts the domain of intelligent beings. The conjunction of (1) and (2) is the strength of Nathan’s interpretation. But I believe that (2) is false, that there is a third, crucially important sense of ‘intelligent’ which, if accepted as operative in DXII, removes a major difficulty for my interpretation. The sense of the word I have in mind is considerably different from the first two, but will be seen to have important logical relations to them. This third sense of the word would make it roughly synonymous with ‘rational’ as when one says, e.g.
Crystal structures are rational.
The behavior of freely moving bodies is rational.
The genetic distribution of inherited characteristics is rational.
It may be said that this use of the term is purely metaphorical. But, if so, then the same charge should be levelled at the artifactual use, something that Hume surely would not do. I will say that something is intelligent or rational in this third sense by describing it as systematic or a system.
What is the logical relation of intelligence as systemicity to the concepts of agentially intelligent being and artifactually intelligent being? The most important questions are: (i) if x is a system, must x be an artifact? and (ii) if x is a system, must x be an agent, have intentions? The answer to (ii) is clearly in the negative. Logic (relations of ideas) does not force us to assent to (i). Yet, Hume argues, our nature does not allow us to believe everything that pure logic allows to be possible. At this point I take Hume to be saying: Experience inclines us irresistibly to believe that similar effects have similar causes. In the present context, one cannot help believing that the original cause (s) or order in the universe resembles human intelligence. If we define ‘artifact’ in its most general sense as ‘system whose cause of existence is analogous to an agentially intelligent being’, then it is psychologically impossible to deny (i), even though strictly speaking, the argument from design suffers from all the flaws brought out by Philo in connection with the standards for good arguments, demonstrative and experimental. Such is the essence of Philo’s ‘lead-in’ argument in XII as I interpret it. Let us recall it now:
‘That the works of nature bear a great analogy to the productions of art is evident; and according to all the rules of good reasoning, … we ought to infer that their causes have a proportional analogy … Here then the existence of a Deity is plainly ascertained by reason …’ ((4), p. 217)
But it is to be remembered that, in DXII, the only content for ‘a Deity exists’ that can be provided by reason is ‘There is a cause or causes of order in nature and it (they) are somehow analogous to human reason’. The evident problem (to Hume) with theistic interpretations at this point is that, if x is the cause of an artifact, then, whatever else may be true of x, x itself must at least be (what I am calling) a system. The latter fact, together with (i) (put in statement form) is how the threat of infinite regress arises in the first place.
‘If reason … be not alike mute with regard to all questions concerning cause and effect; this sentence at least will it venture to pronounce; that a mental world or universe of ideas requires a cause as much as does a material world or universe of objects, and, if similar in its arrangement, must require a similar cause.’ ((4), p. 160)
The cause of a system must itself be a system. To that extent alone, the cause of any system must be something analogous to an intelligent agent. But we already know that a system need not be an agent. As I interpret, Hume’s way out of the problem is through the notion of analogy. Merely being a system is sufficient to classify an entity as analogous to an intelligent agent. But we already know that a system need not be an agent. The reason for this is: systemicity is the notion which every meaning of ‘intelligent’ presupposes. Thus, though we must agree that the cause or causes of order in the universe is (are) somehow analogous to an agent, we are not thereby obliged to concede that this cause or these causes is (are) agents. Yet it will be describable as intelligent, though not necessarily intelligent in any other sense of the world than that sense in which crystal structures are ‘rational’, i.e. systematised. So, understood, Philo’s belief in the intelligence of the ultimate cause is nothing more than a belief in its intelligibility. It implies not pure faith, but a faith in reason and science. For there is no reason to believe that it must be an agent, though system it must indeed be, and will therefore obviously be analogous to an agent in sharing one of the latter’s essential properties. We have already seen Philo hinting that the ultimate causes of order may be found in the inmost essence of matter and that these may be intelligible to pure reason, in the Humean conception of that faculty. We are here brought round once again to the same point.
In the light of these suggestions, consider again the passage from part XII.
‘… if we make it a question whether, on account of these analogies, we can properly call him a mind or intelligence … what is this but a mere verbal controversy? … if we are not contented with calling the first and supreme cause a God or Deity, but Desire to vary the expression, what can we call him but MIND or THOUGHT, to which he is justly supposed to bear a considerable resemblance?’ ((4), p. 217)
I believe that this passage can be given a rather natural interpretation which agrees well with my explication of Hume’s meaning in referring to the ultimate cause as ‘intelligent’. The important clue here is that in this passage Hume has Philo twice show reluctance to accept the most well-entrenched aspect of the traditional concept of deity. In my terms, this aspect is that of being an agent. The idea is conveyed that it is ‘a question’ whether we can properly call him a ‘mind or intelligence’. Philo then directly implies that he is not ‘content’ with calling the first and supreme cause a ‘God or deity’. By studying the grammar of Philo’s statements at this point we can see the difference between the sort of entity Philo is ready to acknowledge and the sort of entity over which he has qualms. If we refer to a so-and-so we evidently refer to a particular of a given kind. Philo’s final term for the supreme cause very significantly contains no indefinite article. It is a completely general term. Instead of a mind it is MIND. Instead of a set of (divine) thoughts it is THOUGHT. I believe that Hume has excellent reasons for this qualification and that they have already been clearly articulated by Philo. Aristotelian language is best suited to make the relevant point: If ’cause’ is taken to mean efficient cause, something temporarily antecedent to its effect, we have already been given powerful arguments by Philo that a search for such an ultimate cause inevitably leads to an infinite regress. An attempt to stop the regress at some point by saying, ‘here we have an agent; therefore the regress stops’ does not work. We cannot restrain ourselves from an enquiry concerning efficient causes no matter what entity we have arrived at, even a divine mind with its system of ideas. So long as our enquiry concerns particulars
and efficient causality; we cannot stop. An ultimate cause cannot be found, unless we are using ’cause’ in a sense other than that of ‘efficient cause’. Philo announces belief in an ultimate cause. I conclude that he does not mean ‘efficient cause’. I submit that his ‘Mind or Thought’ refers to a kind of formal cause. Given Hume’s general position on causality, it is only of a formal cause that one could say that it necessarily had no prior cause. Because of Philo’s arguments over the infinite regress problem, and the absence of an indefinite article from his most judiciously chosen term for the ultimate cause, it is difficult to imagine what else could be meant. These facts fit well with my previous hypothesis about the meanings of ‘intelligent’. In every sense of ‘intelligent’ (crystal, clock, man, God), if x is intelligent, x is a system. What we are talking about is the cause of all systems whatsoever, albeit their formal cause. In naming this MIND or THOUGHT I believe that Hume was trying to express what I have been trying to capture with the term ‘systemicity’. If I am right in this suggestion, then Philo’s apparently profound mysteriousness at this juncture is really a masquerade. What appears to be profundity is triviality in disguise. The triviality emerges when we realise that the proposition—The cause of order in the universe is intelligent—translates in my terms into the unexciting proposition that the (formal) cause of everything systematic is systemicity. That is the hard cash for ‘The first cause is intelligent or rather analogous to an intelligent agent’. Hume’s irony is disguised and subtle.
It is small wonder that, at this point, Philo begins to characterise the whole issue as a verbal dispute. It may appear that he intends to show that the purely verbal nature of the controversy is to the advantage of theism, but I believe the appearance to be deceptive. Dwelling on the structure of the text at this point, ((4), p. 218) one should, I think, conclude the following: Hume is really telling us that the theist is safe in saying ‘There is a great and immeasurable, because incomprehensible, difference between the human and the divine mind … the difference is of a nature which cannot be too much magnified’. But for Hume this conclusion is the most that natural religion can establish about the human vis-à-vis the Divine from the very nature of the case. On the other hand, for Hume all that the evidence of natural religion forces the atheist to admit is that ‘there is a certain degree of analogy among all the operations of nature … [that] the rotting of a turnip, the generation of an animal, and the structure of human thought [are] energies that probably bear some remote analogy to each other’. The reason why Hume thinks the atheist must agree with (natural) theism is that the only conclusion warranted by natural religion is so broad and general that it is actually consistent with atheism. But, it is not only this; the reason why it is consistent with atheism is that the conclusion is a truism: Each thing (of whatever kind) is in some remote way analogous to every other thing. Natural religion establishes no more than what an atheist can safely admit. If the controversy is merely verbal, then to Hume it is the theist who is bound to suffer. His case is necessarily the weaker.
Now, as to Philo’s final recommendations to Pamphilus. It is by far the easier interpretation to take it for what it appears to be: the making of a connection between scepticism and religious fideism. Many have been influenced by the fact that Hume is undoubtedly some sort of sceptic in such matters as causality, but who nevertheless defends the value of natural belief in such matters as cause and effect relations in spite of pure reason’s inability to demonstrate them. This interpretation is thus very natural. But one must place it beside what Nathan has indeed shown about D., viz., Philo professes belief in an ultimate cause of order, but not on the basis of faith alone. Experimental reasoning as employed by the natural theological (Cleanthes) does establish as probable the existence of an ultimate cause of order. But besides this conclusion, with which even an atheist can safely agree (especially if it is a formal cause), the only conclusion about this ultimate cause that can be given any degree of probability is that it is in some remote way analogous to human intelligence. Here the atheist can say, as Philo does, that this is true even of the rotting of a turnip. If natural religion can provide nothing more than this severely limited conclusion, then to that extent, Natural Religion eliminates all basis for debate between atheist and theist. But it would seem that eliminating the basis in the way just described is rather like eliminating the basis for debate with one’s opponent by committing suicide.
Earlier, Cleanthes defends himself against Demea’s charge of anthropomorphism, saying in effect that it is really mystics who represent the greatest danger to theism. He in effect says if we take mysticism to its logical conclusion and maintain that any comparison or analogy between Man and God is absolutely impossible, the result will eventually be that ‘God’ will become a meaningless noise or set of ink-smears. Philo appears to agree, but finally in XII shows us that the only human quality that we can plausibly say the ultimate cause of order bears any analogy to is intelligence.
‘… the analogy, imperfect as it is, can be carried no farther than to the human intelligence and cannot be transferred with any appearance of probability to the other qualities of the mind:’ ((4), p. 227)
If all that natural religion can provide us with is the severely limited conclusion that there is probably an ultimate cause of order and it probably bears some remote analogy to human intelligence (though it is impossible to know whether the analogy is any more interesting than that between intelligence and a rotting turnip), we have just seen how inadequate this is as a basis for theistic belief. If we become mystics, however, then according to Cleanthes, the content of all that we assert or deny, will become indistinguishable from the contents of all of the atheist’s assertions or denials. Is there any basis left for theistic belief? Yes, Philo says: philosophical scepticism. But what is the real import of his assertion here? Is he sincere? Counselling us to use scepticism as Pascal advised us to use the sacraments for hammering theistic belief into ourselves as by operant conditioning? I believe that Hume’s thesis here is that, for a ‘man of letters’, i.e. a person with wide knowledge and strong reasoning powers, becoming a sceptic is the only way to become a theist. The interesting thing here, though, is that this sort of fideistic theist would appear, from a Humean perspective, to be very similar to one who uses libations, incantations, and many other (symbolic) techniques for strengthening his purely superstitious beliefs. Now Hume wrote extensively on the subject of superstition. I believe that it is reasonably evident that Hume had little sympathy with ‘zealots and enthusiasts’. Reason, demonstrative or experimental, provides little or no basis for theism. Only by means of sceptical meditation can a man of letters acquire a firm belief that besides something analogous to intelligence, the ultimate cause has such other qualities as will, love, intention, wisdom, etc. It seems to me that Hume’s concealed message is: only a person dazed by skepticism to a point at which he is capable of believing anything can be sure of having an imperturbable theistic faith. (Compare Demea’s opening remarks.) By contrast, according to Hume, our natural belief in causality requires no such scepticism, and is, in fact, all the stronger for never having suffered it.
A religion which is essentially like superstition is one thing. For Philo, there is something else that he terms ‘religion of the philosophical and rational kind’. Philo expresses no disagreement with Cleanthes when the latter says
‘The proper office of religion is to regulate the heart of men, humanize their conduct, infuse the spirit of temperance, order, and obedience; and as its operation is silent and only enforces the motives of morality and justice, it is in danger of being overlooked, and confounded with these other motives. When it distinguishes itself and acts as a separate principle over men, it has departed from its proper sphere, and has become only a cover to faction and ambition.’ ((4), p. 220)
Philo condemns beliefs about the eternal, the infinite, the beyond, the hereafter, etc. as not only of no importance for human life, but as actually antithetical to the well-being of humanity. As opposed to such ‘superstition‘, he says:
‘It is certain from experience that the smallest grain of natural honesty and benevolence has more effect on men’s conduct than the most pompous views suggested by theological theories and systems. A man’s natural inclination works incessantly upon him … Whereas religious motives … operate only by starts and bounds.
… where the interests of religion are concerned, no morality can be forcible enough to bind the enthusiastic zealot … The steady attention alone to so important an interest as that of eternal salvation is apt to extinguish the benevolent affections, and beget a narrow contracted selfishness. And when such a temper is encouraged, it easily eludes all the general precepts of charity and benevolence.’ ((4), p. 221)
I believe that Philo’s ‘Religion of the philosophical and rational kind’ has two aspects. The first is a belief in the natural qualities of man as basically good. People have a moral sense as naturally as they make inductive inferences. The second is captured in the following by Philo
‘… no one has a deeper sense of religion impressed on his mind, or pays more profound adoration to the divine being, as he discovers himself to reason, in the inexplicable contrivance and artifice of Nature.’
One need not look at this passage for long before he will see the equation
And Nathan sees it. He thinks that it shows Hume to be a theist. I consider it drastically misleading to so characterise Hume. It would be far more accurate to say that, for Hume, Nature takes the place of God. Moreover, as Philo says
‘To know God, says Seneca, is to worship him. All other worship is indeed absurd, superstitious and even impious. It degrades him to the low condition of mankind …’ ((4), p. 226)
Philo is close to Spinoza’s definition of religion as Amor intellectualis Dei. It is thus reasonable to think that Hume, the great disciple of Newton, came to believe that Natural philosophy (science) was the other side of the coin called ‘Religion’. Philo makes another remark about scepticism, but in a different vein
‘… were that divine being (so) disposed … the only persons entitled to his compassion and indulgence would be the philosophical sceptics, a sect almost equally rare, who, from a natural diffidence of their own capacity, suspend, or endeavour to suspend, all judgement with regard to such sublime and such extraordinary subjects.’ ((4), p. 227)
Is Hume then praising the sort of superstition that scepticism can be used to breed? I think not. Rather, I think that we must here recall that Hume distinguishes between Pyrrhonism and mitigated scepticism, and realise that it is the latter which is referred to immediately above. The ‘scepticism’ which Philo recommends to him who would be a ‘sound believing Christian’ thus cuts two ways depending on which of two equally permissible readings one gives to these scare-quoted terms. If like Demea, one thinks of sound Christianity as an arational faith, the intent of Philo’s advice is clear. A Pyrrhonistic onslaught on reason is the best insurance that such a faith will be undisturbed. Of course, the application of this kind of scepticism will lay waste to the subject of Natural Religion as a side-effect. But if, like Philo, one thinks of sound Christianity as being that part of Christianity that can be detached from metaphysics and theology, i.e., morality, then again, it is (mitigated) scepticism which is the best weapon for blasting away the metaphysical excrescence on this sound core of religion. It is in this sense that I believe that, while Hume rejected Natural Theology, he did come to accept what might be called a religion of Man-in-Nature.
(1) Chappell, Vere C., Hume. Doubleday, New York, 1966.
(2) Hendel, Charles, Studies in the Philosophy of David Hume. Indianapolis, 1963.
(3) Hume, David, An Enquiry Concerning Human Understanding. Open Court, LaSalle, Illinois, 1963.
(4) Kemp-Smith, Norman, Dialogues Concerning Natural Religion, by David Hume. Bobbs-Merrill, New York, 1946.
Copyright ©1975 Cambridge University Press. All rights reserved.