9 Design (1935)

In the Introduction to his just published edition of Hume's Dialogues concerning Natural Religion,note Professor Kemp Smith remarks (p. 38) that the Dialogues have a “unique place in philosophical literature. They are unique in two respects. As Leslie Stephen has pointed out, they were the first work in our literature to subject the argument from design to a passionless and searching criticism. And secondly, Hume's destructive criticism of the argument — allowing for the limitations…under which it was formulated — was final and complete. For a couple of generations, theologians, especially in Britain, may have continued on the old lines, as if the Dialogues had never been written. But in the altered outlook of present-day theology, these older ways of argument have in large measure ceased to be approved, and Hume's indictment of them is now but seldom challenged.” Kemp Smith admits that the argument from design, in the form in which Humenote criticised it, not only “persisted into the nineteenth century in the writings of Paley” and others, but “survives in popular and semi-popular forms to the present day” (p. 36). He appears to think, then, that these uncritical conceptions, like those of popularisers of the Jeans type, carry no great weight not merely among philosophers but even among the more serious theologians. This is a view which, while I have no great acquaintance with present-day theology, I should be very much inclined to question. For, apart from the mere ignoring of Hume, there is also the tendency to substitute for the criticised doctrine other doctrines which are open to the very same type of criticism. And both these reactions are facilitated by Hume's “scepticism”, i.e., by his failure to work out a logic.

Arguments like that of design, in fact, cannot be thoroughly gone into until attention is directed not merely to the illogicality of the proofs offered but to the illogicality of what they are supposed to prove. And, whatever suggestions Hume may make towards a thorough-going solution, he cannot arrive at it and force subsequent attention to it, because of the defects in his philosophical outlook — because, like the other “English empiricists”, he was rationalistically concerned with “ideas” (that whose nature it is to be perceived) and not with propositions (what is the case). So long as any admission of “natures” is made, dualism is

  ― 89 ―
inevitable; and even if the term “design” is abandoned, some comparable and equally confused way in which the nature may “express” itself must be retained. Hence, while Hume could point empirically to many of the difficulties of the theory of design, not only could he not prevent it from raising its head again, but his own rationalist preconceptions were an influence towards its doing so. Similarly, it may be remarked in passing, it is because of its rationalism, its retention of “principles” and the like, that the realist movement set going by Moore and Russell has failed — failed, i.e., to work out a realist philosophy. To reap the full benefits of Hume's work, then — and the same applies to the work of the later movement — it is necessary to follow up the questions raised, to cut away rationalist conceptions and so to arrive at a logical position.

That progress has been made in this direction by philosophers (whatever may be said about theologians and pseudo-philosophical scientists) is due not merely to Hume's work but to the fact that Kant, as Kemp Smith points out (p. 39), incorporated Hume's main criticisms in his “Transcendental Dialectic”. But, whatever stimulus Kant may have given to logic, he remains a dualist, and theories which do not go beyond Kant's position can be criticised in Philo's manner, even if not with his precise arguments. One of the most important points made by Kant is that the physico-theological argument, as he calls it, is not an argument a posteriori or from experience, but has as much of an a priori or ontological character as any of the others, that it depends on the conception of that which is ultimate or establishes itself. It should be emphasised here that ontological arguments are not confined to proofs of the existence of God or of something described as the Absolute, but are the means of establishing all ultimates, even in nominally pluralistic theories. It is really in the ontological fashion that the Pythagoreans set up their units or Socrates his forms or that anyone sets up that which is by its own nature. It seems indeed to be shown by the Eleatic criticism of the Pythagoreans, and similarly by Spinoza's criticism of Descartes, that there cannot be more than one ultimate, but the more important question is whether there can be one at all. And until it has been shown that there cannot, until the very conception of “ultimacy” has been rejected, until it has been demonstrated not merely that certain arguments are unsound but that their supposed conclusions are untenable, the position has not been worked out and the same types of error recur. Thus not only can we say that Kant argues ontologically in his theory of the good will, i.e., that which establishes itself by willing itself (just as in the conception of consciousness, by theorists from Descartes onwards, as that which establishes itself by thinking itself); we must also note that the very conception of the “thing-in-itself” involves an ontological argument — for what is an argument but a recognition of dependence? — and that the only solution is not a problematical recognition but the rejection of that which is in itself.

This is connected with the inseparability of cosmological and ontological arguments. Kant shows that the cosmological argument depends on the ontological, that that on which all other things depend can only be

  ― 90 ―
that which establishes itself. But, equally, that which establishes itself has to be taken as that on which other things depend, because in its very conception there is the distinction between its character of establishing and its character of being established; and, while the latter now appears, like a created cosmos, as having “dependent existence”, the same problem as before breaks out in regard to the “self-subsistence” or self-supporting character of the former, the same distinction has to be made between its supporting and its being supported, and so on indefinitely. There is likewise no logical division between cosmological and physico-theological or teleological arguments, because in each case we have the dualism of ways of being, that which has its being in supporting and that which has its being in being supported. The only way to escape from the vicious circle, in which dualism collapses into monism and monism explodes into dualism, is to adopt a pluralist position in which variously characterised and related things are recognised as existing in the same way (spatio-temporally) — a single logic of existence replacing conceptions of “self-subsistence”, “relative existence” and any other flights of rationalistic fancy.

It is worth noting that all theories of higher and lower realities are stated in terms of the common reality we all know — and, indeed, can be stated in no other way. Thus, as Berkeley points out in criticism of Locke, we are all acquainted with the way in which one thing supports another, we know the empirical relation of supporting, but we are not acquainted with the way in which “matter”, in the Lockian theory, supports accidents. Locke had to use an ordinary term, at the same time suggesting that it was to be understood in an extraordinary sense; but he could not even begin to tell us what that sense is, how the “supporting” of accidents by matters differs from the supporting that we know. Thus the common relation gives us no help in the understanding of the metaphysical one, but the use of the common word tends to confuse our minds and makes us imagine that we have understood something. The force of Berkeley's criticism here is not weakened by the fact that his position is open to the same objections, that he utilised our acquaintance with dependence as an actual relation between actual and distinct things, in order to make it apparently intelligible to us that certain things have “dependent existence”. In so doing, in separating the active from the passive, the effective from the effected, he takes up a position logically indistinguishable from that of Cleanthes (and, of course, closely resembling it in detail; Berkeley, we may say, was a notable participant in that movement in eighteenth century thought which Cleanthes represents in the Dialogues), but in his criticism of Locke he did some of the work of Philo.

What Philo does, in fact, is to lay bare just such empirical material as the rationalists use in their arguments, and to show that it will not bear the metaphysical doctrines erected on it. All that his criticism lacks is development in the direction of a thorough-going empiricist logic, which would show that his declension to scepticism is uncalled for. For the sharpening of the logical issues, then, we may take as our point of

  ― 91 ―
departure the famous passage (Part II) in which Cleanthes sets forth the main theme of the discussion. “Look round the world: Contemplate the whole and every part of it: You will find it to be nothing but one great machine, subdivided into an infinite number of lesser machines, which again admit of subdivisions, to a degree beyond what human senses and faculties can trace and explain. All these various machines, and even their most minute parts, are adjusted to each other with an accuracy, which ravishes into admiration all men, who have ever contemplated them. The curious adapting of means to ends, throughout all nature, resembles exactly, though it much exceeds, the productions of human contrivance; of human design, thought, wisdom, and intelligence. Since therefore the effects resemble each other, we are led to infer, by all the rules of analogy, that the causes also resemble; and that the Author of nature is somewhat similar to the mind of man; though possessed of much larger faculties, proportioned to the grandeur of the work, which he has executed. By this argument a posteriori, and by this argument alone, do we prove at once the existence of a Deity, and his similarity to human mind and intelligence.”

The weakness of analogy, the question-begging character of the reference to the Author of nature, are obvious points of criticism here. But a more important line of attack is that which proceeds from the fatal admission of Cleanthes that there is nothing which does not bear the same marks as are borne by the productions of human contrivance, i.e., that what he regards as marks of contrivance are conditions of existence. If we could distinguish the contrived from the uncontrived, then we might be able to discover certain marks of the former; but if everything bears certain marks, then, naturally, human contrivances will do so, but this will be no argument in support of the view that they are marks of contrivance. As far as human operations are concerned, the materials worked upon, the things not humanly contrived, will bear the marks in question just as much as the things produced or humanly contrived. Accordingly, the alleged a posteriori argument from the special mode of operation of human contrivance, conceived as the introduction of marks of design, disappears, and we are left with an a priori argument, of the cosmological type, to the effect that everything in nature, whether it is worked upon by human minds or not, is necessarily dependent on some creator or source or has a subordinate form of existence. Such a position is involved, indeed, in this very use of the term “nature” (generally, in our day, with a capital). Now, as we have seen, even an a priori position has to make use of empirical material; we could never entertain the conception of dependence as a peculiar form of existence (i.e., as a qualification of the copula), unless we were acquainted with it as a character of actual situations — unless we had had experience of the occasioning of one thing by another, whether the latter is a mind or not. But there is nothing in this experience to warrant the conception of dependent existence; on the contrary, we are concerned with whether or not the one thing is actually dependent on the other — and not with whether it is

  ― 92 ―
dependently dependent, and so on indefinitely, as we should have to assert if we took it to have this peculiar mode of being.

Logically, then, even if there are marks of design, this is only a matter of a particular relation between different sorts of particular things, which are all alike “actual” or, as I should prefer to put it, spatio-temporal. But Cleanthes is quite correct in taking the adjustment he refers to as characteristic of things in general, i.e., anything we like to take has characteristic ways of working, one phase of which leads on to another, and acts differently in different situations. Accordingly, since adjustment is found alike in the things we ordinarily call machines and in those we do not, there is no ground for tracing the adjustment of the former to the fact of their being machines and thence concluding that the latter also are machines produced by some non-human contriver; just as there is no ground for Berkeley's argument that those of our “ideas” which we do not control must be controlled by some other mind, but a mere assumption, in spite of appearances to the contrary, that control or efficacy is always mental, so that working on mind must be working by mind. There is then, no special “working” character of mind and no special “wrought” character of machines or other things. According to the admissions of Cleanthes — and if he does not make them, his argument cannot proceed — human contrivance consists not in making things work together, since they already do so, but in making things work together in a certain way, i.e., in making certain “workings” or activities, which simply means making certain things.

Human contrivance also involves the working of things on minds, but that is a matter to which I shall return. Meanwhile, it may be of some interest to consider another incidental point. If human contrivance consists of making things which worked together in a certain way work together in a different way, then we have human interference with God's arrangements; in other words, Cleanthes, in starting from human contrivers, has no argument for a single contriver but must admit a multiplicity of competing contrivers. The alternative is to take human beings also as contrivances of the original contriver, and not as contrivers at all. A similar dilemma faces Berkeley in his theory of “the conduct of life” and of our knowledge of other finite minds. Unless the conduct of life consists merely in imagining what we like, without the slightest effect on the sensations that are forced on our minds, Berkeley must admit that we can operate on God's ideas, i.e., that we can give him new ideas just as he can give us. And, again, unless other minds can alter God's ideas, we cannot have the slightest reason for taking any particular sensation as indicative of the operation of another mind, but only as coming direct from God. Finally, as in the case of Cleanthes, Berkeley either has to admit that we are simply some of God's ideas and have no “agency” whatever, or has to recognise a thoroughgoing interaction and abandon his doctrine of the “active” over against the “passive”.

To return to the main argument, human contrivance means that human beings, by “thinking” and other operations, produce certain things, and it

  ― 93 ―
may well be that only human beings produce things of certain sorts. In that case, if we see such a thing, we can infer that a human being made it; but the inference depends not on our finding in the thing a peculiarly designed or contrived character, but, as Philo points out, on our knowledge of the fact that things of that kind are made by men. We have all seen men going through a series of operations culminating in the existence of a house, and we have never seen a house coming into existence in any other way. It is on this account that, when we come across a house which we did not see being built, we conclude that somebody built it, and not because we see a contrived character in it. But, if Cleanthes were right in finding contrivance in everything, then we should be just as ready to conclude that a house or a ship which we did not see constructed had been made by the contriver of things in general and that no man had a hand in its construction, as to draw the conclusion that we do. Moreover, if Cleanthes were right, he would have no need of any analogy; he could argue straight to a designer from the designed character of anything he liked to consider, without having to refer to human performances. The fact is that there is no designed or contrived character, that contrivance is a relation between different things and not a character of either by itself. Knowing such relations (e.g., all houses are made by men) we can draw inferences — we can infer the human contrivance of some things, the mouse or bird contrivance of other things, and so on — but never contrivance by the contriver, and always on the basis of experience of things of the sort A contriving things of the sort B.

The fact that Philo does not enforce the logical issues, does not insist, e.g., that “the adjustment of means to ends” signifies merely that when something happens to a thing it does something else, accounts for his being able at the conclusion of the argument to put forward the modified, and, as he admits, useless, assertion “that the cause or causes of the order in the universe probably bear some remote analogy to human intelligence”. Nevertheless, his suggestion that what is really required is revelation may quite well not be ironical, but, taken in conjunction with his earlier argument, may rather indicate that Hume had some conception of the weakness of relativism, the impossibility of characterising a thing by its relations or of taking the character of a thing in itself as testifying to the character of a related thing. This, indeed, is the very point that Hume made in regard to causality in the Treatise; but only the abandonment of his rationalistic theories of “ideas”, “relations of ideas”, and, still more important, spatial and temporal units, would have enabled him to bring these questions to a decisive issue.

Cleanthes, at any rate, does not and cannot explain why what he takes to be provable is not also observable, just as, when we see a house and infer that men built it, we also believe that someone saw them build it. He wishes to argue to the existence of something with which we are not

  ― 94 ―
acquainted and whose modes of operation we therefore do not know, on the ground that these (unknown) modes of operation are the only thing that will account for the existence of something with which we are acquainted; and a good deal of Philo's “scepticism” is just the bringing out of the sceptical character of this position. The particular contention that a greater contrivance implies a greater contriver is one which, as Philo suggests, we are not in any way bound to accept; for if by a greater contriver we merely mean one who makes a greater contrivance, we have not indicated any characteristic of his which we can call his greatness and which can distinguish him from lesser contrivers — and if we try to specify such a characteristic, if, e.g., we make it a matter of physical size, then, while it is still a question of experienced connections, of finding a bigger thing making a bigger thing, we must admit many contrary cases of a physically smaller cause having a physically more extensive effect. Only its vagueness can protect such a contention, which is still an attempt to find in the effect a positive characterisation of the cause.

Equal difficulty is found in the attempt to characterise the cause negatively from the effect. Cleanthes argues — all machines are designed by minds (which we have seen to be a mere assumption, if all that is meant by a machine is something which has characteristic modes of operation under various conditions); the universe is a machine not designed by a human mind; therefore, the universe is designed by a non-human mind (and, as above, since the universe is greater than any human contrivance, this non-human mind must be greater than any human mind). Now, leaving aside for the present the questions arising in connection with the conception of “the universe”, we may object to the above argument that, unless we had independent evidence of the existence of a non-human mind, or if human minds were the only minds we knew, the acceptance of the second premise would lead us to cast doubt on the first. In other words, we should require to have some acquaintance with non-human minds before this argument would produce conviction, and consequently we cannot take it as a proof of the existence of a non-human mind or a means whereby we could be led for the first time to suppose a non-human mind to exist.

This criticism has a certain similarity to Alexander's criticism of the view that we argue to the existence of minds other than our own by analogy, i.e., that, since we know that certain bodily processes are associated with certain of our own mental processes, then, when we find such bodily processes not associated with our own mental processes, we assume that they are associated with some other mental processes and thus recognise the existence of other minds. Alexander's contention is that such bodily behaviour could not give us the conception of “another mind”, and that, if we were not independently acquainted with other minds, we should simply take it that the bodily behaviour in question was not associated with any mind. Apart, however, from the special difficulties in which Alexander is involved by his conceptions of our “assurance” of other minds and “enjoyment” of our own, it may be pointed out that if we are acquainted with our own minds, then we may, as Moore puts it, reasonably suppose a similar mind to exist in similar observed conditions. So that, even if inference is not the only or the regular way in which we

  ― 95 ―
learn of the existence of other minds, it may be (and is) one of the ways. But it cannot be the way in which we learn of the existence of a special type of mind, with which we are not otherwise acquainted. Or, if the last statement is one which must be received with caution, at least we do not so learn of a type of mind (or of anything else) with which we cannot be otherwise acquainted.

The caution referred to is enjoined by the following considerations: we may believe that things of the sort A come into existence by the contrivance of minds or mental processes of the sort a, B by b, C by c, and so forth. Then if we come across a thing of the sort ABC, we are thereby led to suppose the existence of a mental process of the sort abc, even if we have no direct acquaintance with such a process. There would still, of course, be the possibility of our being mistaken about some of our premises, and we might retain a lingering doubt about the existence of the mentality in question until we had observed it; but at least we were led to conceive it by that evidence. Even so, we think of its contrivance of ABC as a particular contrivance going on at a particular time, just as we found the contrivances of A, of B and of C to be in the first place. It is another matter entirely when it is a question of Contrivance as a whole; in that case there is not even the remotest analogy with human contrivance, there is no reason for calling the assumed “higher” operation contrivance at all.

As we saw, if we take positive account of human contrivance, we find that it is a relation between human minds and other things, and that both exist in the same way, not that one “exists contrivingly” and the other “exists contrivedly”. The question is then just to find what things human beings do contrive, and this raises no question of any contrivance of other things by other beings. Equally, there is no question of “marks of contrivance”. But if, like Cleanthes, we take as marks of contrivance working in certain regular ways, having a certain harmony of parts, and so on, i.e., having a certain character or “constitution”, then we must take the things we call contrivers as also contrived; the workings of the mind are of the same “contrived” character as the workings of any other thing. But to say in this way that all contrivers are contrived is not to argue for the existence of one great contriver who is not contrived; it is to argue against it. On the other hand, if we say that contrivers are not contrived, then we cannot take regular working, etc., as marks of contrivance, because they also have these marks, and so we have no ground for calling the ordinary “works of Nature” contrivances, and the argument of Cleanthes has not even a beginning.

There are, of course, additional difficulties in the way of the conception of an original contriver of the whole of “creation”, but in the first place there is the same difficulty, viz., that if we are to say anything about this contriver at all, we must regard him as having characteristic ways of working, as being of the same logical order as his alleged “creation”, as being a set of interacting situations — and, apart from the question of this set being itself situated or environed, we are already committed in those

  ― 96 ―
admissions to the rejection of an origin of things and to the treatment of the emergence of further situations in the same way as we treat developing situations now, viz., as exhibiting certain forms of action and interaction (though indeed this emergence and the implied pre-existence already constitute an environment). It seems of minor importance then to point out that there can be no contrivance of a “universe” or totality of things, because the contriver would have to be included in the totality of things; and if things constituted “one great machine”, he would be part of the machine and not its maker. Alternatively, things would not constitute one great machine, but the least we could think about would be the contriver contriving contrivances — in which case, of course, there would be no argument to a contriver. Even that, however, does not provide an escape; but (as in the case of Berkeley's similar minimum of a mind having ideas) the different elements in this complex situation must be taken as all existing in the same way and none as having “higher” being than another. So that, as in the case of human contrivance, we have a particular relation between particular things — or else we have nothing at all.

We see, then, that Cleanthes, in speaking of the contriver of the universe, is not seriously thinking of the latter as a universe or totality. But it is no more possible, even if we could separate things active from things passive, to think of a totality of created things (or “Creation”, as it is called) than to think of a totality of things in general. Indeed it is a mere phrase, without any experience or serious argument to justify it, because if we do think of something we call Creation or The World, we can think of it only as certain things acting in certain ways — just as, if we thought of a creator, we could only think of him as acting in certain ways, and so could not make the logical distinction which the theory requires. But, speaking simply of the things Cleanthes calls machines, things to which we do not attribute any power of contriving, we do not find that they make up one great machine; we do not find any total situation or any way of working which is that of the whole. There is no object of which we can say, “That is the world with all its parts and characters adjusted to each other”; there is no observable situation of all things working together — for good or for anything else. On the contrary, in any observation, while we find adjustment or ways of working, they are always ways of working of particular things; and, even so, these are not all the ways of working of the things, apart from the influence of the other things with which they may come in contact. Thus there is no formula or law which will cover every action of any group of things, however large or small, that we like to take, and there is no question of one great machine or one total way of working.

The facts of complexity and interaction have a further bearing on the notion of a machine. As we have noted, contrivance is a relation between distinct and independent things — independent in the sense that, though a house, e.g., would not have existed but for the operations of men (and in that sense of dependence human beings are also dependent on inanimate things), it now does exist in the same way as they do, we can

  ― 97 ―
know it even though we know nothing about them, and we can interact with it in the same way as we interact with them. It is important to observe, then, that, apart from the fact that we can only operate on given materials, there is nothing which is a mere contrivance; anything that we contrive always has (without reference to mistakes we may make about it) characters which we did not anticipate. In the same way, of course, as against voluntarism, there are always characters of our voluntary actions which we were not aware of when we willed them, so that they are not simply “our decisions”. To say, then, that there is something more than we “contrived” about the things we have made is to recognise that they are independent things and to do away with any logical division between contrivers and contrived, and so with any conception of a totality of the contrived — or, for that matter, of a totality of any kind, as contrasted with the interaction of independent things. And to say that, while this may be the case with human contrivances, there is nothing about God's contrivances, or God's creation, of which he was not aware when he created it, is to destroy the analogy between God's contrivance and our contrivance, and to make the mere assumption that there could be a contriver who knew everything about his contrivances. There certainly could be nothing about his contrivances to prove that he did — though, if he did, he would know something, i.e., there would be two distinct terms to the relation of knowledge, and so there would still be independent existence of the contrived. But a more important point here is that, if God knew “all about” his contrivances, they would not be things at all but “natures”, and the question of the contriving of things would be quite untouched.

Connected with this question is the further point raised by Philo that the order and contrivance put into the world by God implies a pre-existing order and contrivance of God's ideas. It is not a sound objection to Philo's argument here to say that the theory of ideas is false; because, if we admit that certain beings contrive a certain order or arrangement of things, we have to admit a certain order or arrangement in them, whatever it is an arrangement of. Thus, in order to intend to create an ordered world, God would have to have a certain arrangement of intentions or gestures towards what he was going to create; and, if these expressions are criticised as merely statements of relations, there is still implied a certain order of the related things, i.e., of the things which intend as well as of the things which are intended. Consequently, rejecting the theory of ideas (or simply saying that the only positive sense that can be given to “our ideas” is our demands), we can still say that Philo's line of argument is sound, that complexity and order in the contrived imply complexity and order in the contriver, and so, if order requires a contriver, the contriver must have a contriver. And it is not open to Cleanthes to say that, just as he need not, in knowing a causal relation, inquire into the cause of the cause, so he need not concern himself with how God came to be; because it is only by having a general contriver that we can have a general contrivance or “Creation” — though, as we have seen, we cannot have it even then. The point also arises here as before, from the

  ― 98 ―
complexity of God's contriving, that any of his acts of contriving takes place among others, i.e., in an environment, and, even if we could limit that environment, that limited environment would be “the universe”, so that the universe would not be created by any act of contriving. We note, further, that, while Philo has not succeeded in working out a logical position, he has raised some of the most serious logical questions.

It is in view of some of these difficulties that Cleanthes goes back on his previous position and says, since it is on the analogy of men that he is arguing, that he is prepared to accept a theory of many gods operating as various men do, i.e., to treat “God” as finite, imperfect, many, just as men are. He would be quite satisfied, he says, so long as it was still admitted that there is design everywhere. But actually, on that view, he would be dropping his theory of the world as a total machine and substituting for it a theory of various single contrivances, which then would interact in a perfectly natural manner, without any question of all these interactions being contrived — though, as has been indicated, this would be the case even on the hypothesis of a single contriver. Thus there would not be design everywhere; and, incidentally, on his “pluralistic” theory, Cleanthes could give no reason for rejecting the view that anything we have not seen being made by men was nevertheless made by some man or body of men in the past (“Once a warrior, very angry”, etc.); in other words, his position becomes manifest as one of pure guesswork or mythology. He has no escape, then, from the truly pluralistic position that even a designed thing is an independent thing, something which has its own existence and ways of acting — the alternative being that the designed has not its own ways of acting, since designing is the only way of acting, and hence that the designed does not exist, and consequently designing does not either.

The fact that we cannot think of a totality of things is brought out again when Demea, as the more consistent logician, refuses to make the admission that Cleanthes had made; for we find that Demea can give no account whatever of his all-causing deity, that his position is really a sceptical one, because he has to say that we can know only of the existence of God, or that we must postulate his existence, without knowing anything of his nature, which is in fact incomprehensible to us. This is so, of course, if he has a “nature”. But that is only to say that doctrines of “natures” or of the incomprehensible are quite otiose, that they give us no assistance whatever in knowing things — indeed they are hindrances. The point is that, if we know nothing of God's character, then we do not know what it is that is said to exist, and consequently Demea's position is a perfectly empty one. But so is the main position of Cleanthes, and it is no accident that, in the discussion of evil, he falls back on “incomprehensibility”. It is indeed quite a common type of argument to try to make out that God's goodness is of a higher order than ours and that we cannot fully comprehend his purpose or we should recognise that all men reckon evil has a place in the grand design, and meanwhile we have to take it as a matter of faith. This simply means that the postulation of a

  ― 99 ―
total and perfect design is to induce us to reject the knowledge that we actually possess. Descartes argues in a similar manner in comparing the perfection of deity with the imperfection of men and the errors they fall into on account of their limited intelligence. Now one definite objection to all such arguments concerning our limited intelligence is that it is our limited intelligence that is putting forward these arguments, and if there is any dubiety about our ability to understand these matters, then there is equal dubiety about the arguments themselves. But there is no reason whatever for scepticism or “suspense of judgment” so long as we have beliefs, i.e., so long as we find things having definite characters and acting in definite ways; and we always do.

It is curious that Philo should propose suspense of judgment (end of Part VIII) immediately after he has come nearest to presenting a position which would completely dispose of that of Cleanthes. “In all instances which we have ever seen, thought has no influence upon matter, except where that matter is so conjoined with it, as to have an equal reciprocal influence upon it. No animal can move immediately any thing but the members of its own body; and indeed, the equality of action and reaction seems to be an universal law of nature.” Here, in spite of the rationalistic theory of equality (and the theory of ideas appears in the same passage), we have an approach to the recognition of interaction as a condition of existence, so that even a contriver is seen to be influenced by his material — just as Socrates, in his attempt to show that the mind “rules” the body, cannot get over the fact that, in order to do so, it must act in certain ways on the occasion of certain bodily conditions. Earlier in the same Part, Philo has recognised the fact “that matter is, and always has been in continual agitation, as far as human experience or tradition reaches”. But, in going on to consider the hypothesis of a material cosmogony, he considers it possible that matter may exist for a time in a “disorderly” state, after which it may or may not achieve “order” — as if there could be any existence at any time except that of sorts of things, as if “chaos” itself, if it had any meaning, could mean anything but certain things going on in certain ways, i.e., a certain “order”. The position is, as indeed we otherwise know, that Hume has not freed himself from dualism, from the doctrine of kinds of existence; yet it is remarkable how little further pressing of the argument of Philo would enable him to do so.

Philo's scepticism, however, his deficiency in logic, comes out most clearly in the well-known passage in Part VII, where he says: “In this little corner of the world alone, there are four principles, reason, instinct, generation, vegetation, which are similar to each other, and are the causes of similar effects. What a number of other principles may we naturally suppose in the immense extent and variety of the universe, could we travel from planet to planet and from system to system, in order to examine each part of this mighty fabric? Any one of these four principles above mentioned (and a hundred others which lie open to our conjecture) may afford us a theory, by which to judge of the origin of the world; and it is a palpable and egregious partiality, to confine our view entirely to that

  ― 100 ―
principle, by which our own minds operate.” The hypothesis of Part VIII, then, is the conjecture of a fifth principle of explanation, that of motion or agitation. But, in the light of the above passage, it is clear why Hume did not see how near he then was to the mark; for, although the question is plainly one of the conditions of existence in general (somewhat confused, no doubt, by the reference to an origin), he does not see that these conditions will govern alike the various forms of operation he mentions, that they are not “principles” but particular proceedings of particular things. Philo's question reminds one of the supposition that in certain parts of the “universe” two and two may not be four; indeed, it is even worse, since it is a logical question that is at issue, since we cannot travel away from logic, however distant a system we go to, but the very supposition of such a system is a supposition of complex and interacting things.

Certainly, he quite correctly says (Part VIII) that: “Every event, before experience, is equally difficult and incomprehensible; and every event, after experience, is equally easy and intelligible.” Certainly, again, Hume, recognising that the effect always differs from the cause, could have no logical difficulty in supposing, e.g., that minds arise from the non-mental. We have to remember also that the theory of Cleanthes is the theme of the Dialogues, and that Philo is more concerned with weakening the position of Cleanthes than with presenting a position of his own. Nevertheless, design has not been thoroughly dealt with until such a position has been worked out; but that would require the removal of the defects of Hume's theory of causality and more particularly recognition of spatio-temporal continuity and rejection of the theory of spatial and temporal units.

Once that is done, the removal of the dualism of active and passive presents no difficulty. For although, “before experience”, we could suppose these classes to be exclusive or again to intersect, so that, in addition to things which are active and passive, there are active things which are not passive and passive things which are not active, in actual experience we find only interaction, things which act and are acted on. Apart from a rationalistic theory of “natures” (as presented, e.g., by Berkeley) we find no basis for the conception of separate classes of agents and patients, arrangers of phenomena and phenomena to be arranged, designers and the designed. We do not require to go beyond the facts themselves, and of course we cannot logically do so, to obtain an answer to the question how things came to be arranged as they are. The answer to the question how any particular arrangement arose is that it issued from a certain other arrangement, and there is no question of any total arrangement demanding (and being unable to receive) explanation by something more than total, viz., the arrangement of its arranger and itself. Anticipation, as far as it goes, is only one particular relation of the general type in question, and the anticipator is not pure “force” but also “matter”, i.e., also acted upon, even in his anticipating, just as the anticipated also acts. This pluralistic conclusion is the upshot of Philo's argument or, at least, it is that alone which would make his position consistent; and it is a positive, not a sceptical conclusion.