ISSUE: Summer 1930
I
If a cultured person should wish to know more about any of the distinctive concepts or propositions of religion, he probably would not think of seeking his information in an article on modern science. In all his experience, he has not yet encountered an instance in which scientific discourse and religious discourse were successfully blended; nor would it be strange if he suspected that their blending is impossible.
A great many experts have been offering to explain the fact that these two systems of discourse are disjoined. Most of them insist that at is unnecessary to keep them separated. Many of them say that the inconsistency that one seems to find between the two is apparent rather than real; and that it will surely vanish as soon as both systems shall have been perfected. Between a complete and “genuine” science on the one hand, and a complete and “true” religion on the other, it is inconceivable, so they tell us, that there should be any conflict. Indeed, they say, it would be most unfortunate for both science and religion if they were thought of as being opposed to each other in any essential respect.
These arguments are now well worn. Some of us have read them so often that we now skip them whenever we find that they, are about to be presented. And yet, they are certainly debatable—so much so that we ought to examine them very carefully before we allow them to be used on us again. If we should look into them carefully, I think we shall find that there is good ground for considering quite a different view of the case, which may be outlined in this manner:
Scientific discourse and religious discourse are distinguished from each other by the fact that they are framed so as to satisfy two distinct sets of axioms. Certain axioms of one system contradict certain axioms of the other system. A “reconciliation” would therefore require that one system, the other system, or both systems should be destroyed. If one desires to preserve both systems, the only logical way of doing it is to recognize that both of them cannot be entertained at the same time.
We shall have to weigh these assertions in order; but inasmuch as the first one deals with axiomatic propositions, we might pause long enough to recall what an axiom, in general, is.
An axiom is a proposition that is accepted without debate in order to give form to a discussion. We often talk as if we refrain from debating it because its truth is self-evident —as if no proof that might be offered in its favor is more convincing than its mere statement would be. So we were told in high school, but, as Professor Eddington has remarked, this view is universally rejected nowadays. Many axioms in geometries and algebras are not convincing of themselves; neither can one prove them by any of the other standards of the system in which they are accepted. Axioms in themselves are neither true nor false; they are nothing but arbitrary assumptions. A proposition in geometry is said to be “proved,” if it is shown that it satisfies the axioms of the system; but there is no way, of proving the axioms themselves. If we should replace any of the axioms in Euclid’s geometry by a proposition that denied it, but which yet contradicted none of his other axioms, we should get, as a result, a different geometry from Euclid’s, but not a false geometry. From time to time in the past 109 years, a variety of such geometries have been so constructed, each of which is internally consistent, and each of which can be used, more or less conveniently, for describing tracts of land. Some descriptions, indeed, can be made more conveniently in other systems than in Euclid’s, but that is not the point that concerns us.
I have mentioned these facts to emphasize that axioms are not things before which we are to bow down and worship. They are in fact instruments of our own creation; being nothing but arbitrary rules for determining the form of our description or other procedure. Without some such set of rules, we should get nowhere. The selection of one set rather than another may, rest entirely on the convenience of him who makes it.
Let us turn to a more familiar example than a formal science. In any game of cards, the evaluation of the various cards is axiomatic. In the game of Bridge Whist, we arbitrarily declare that the ace in a trump suit shall win any trick on which it may legally be played. In the game known as Five Hundred, we arbitrarily say that three other cards, namely, the joker, the right bower, and the left bower, shall supersede the ace. The latter game is an orderly game, but it is not Bridge Whist; if we should employ no rule of valuation, or if we should try to employ both of these rules at once, we would find that we had no game at all.
It is precisely the same with a description of the facts of experience: unless we select some axioms, or descriptive rules, the thing will have no form; if we observe one set of rules, we have a scientific description; if we take another set, we have a religious description; if we take still another set, we may have a romantic description, and so on. Whatever sort of description is made, it is to be deemed valid or true, if it presents all the facts in such a manner as to satisfy a complete set of axioms or rules. This criterion implies, of course, that the truth of a proposition is never absolute; but is relative to the formal rules that were chosen arbitrarily.
In saying that some particular religious doctrine is based on a set of axioms, nobody means to suggest that the author of the doctrine first drew up a formidable system of postulates, and then deduced his points of doctrine from them, after the manner of the inventor of a geometry or a card-game. All that is meant is that he states his doctrine in a form that satisfies certain propositions that may be considered as axioms; and that if he should alter his statement so that it contradicted any of these propositions, it would lose its distinctive character.
II
To establish the main point of the unpopular argument that I have cited, it is unnecessary to list all the axioms of universal science and all the axioms of a typical religion and compare the two lists. If a single axiom of one system of discourse should be found to contradict any axiom of another system, the two systems are not compatible with each other. We shall therefore limit the discussion; we shall consider two propositions that are said to be accepted as axioms in universal science and rejected as axioms of the most typical systems of Judaism and Christianity.
Scientific procedure observes a rule that is often called the axiom of determinism, or of the “uniformity of nature.” It might be stated thus: If two or more isolated physical systems have identical configurations at present, and identical histories within a limited but immediate past, their future performances will be identical. This proposition is a fact of assumption. It has never been proved, and we now know that it never will be, for there are no possible operations by which we can derive a perfectly exact description of the configuration of any physical system. Even within the uncertainties of physical experiment (which are never infinitesimal) the assumption has been justified for only a small portion of the universe. Nevertheless, the scientist insists upon working by it. Not only does he apply it to non-living physical systems; he uses it to justify the researches of the biochemist and the biophysicist. Even such an act as human volition is conceived as an activity of a system of mechanisms, each of which acts according to rigid laws of mechanics. The fact that the behavior of so complex a system has thus far resisted analysis is laid to one of two assumptions: (1) the system is too complex to permit us to describe it readily in terms of our present-day geometry of time and space; or (2) as Professor G. N. Lewis has hinted, it may turn out that the system is simple enough, only we have chosen the “wrong” geometry in which to describe it. However that may, be, it is assumed that there is nothing capricious in the system. It may prove to be too complicated a puzzle for a human investigator to decipher, but it is not too sacred for him to attempt.
Now for aught we know, the world of observable fact may not be a single orderly world. It may be that the order that we discern is an order that we imposed upon the system, by our selection of facts that fit into our predetermined forms, and by our rejection of all the rest. There may be a capricious Providence interfering with the world outside our skins, and a capricious Will—a sort of minor Providence— interfering with the world within them. But scientific rules forbid us from considering such a possibility. Why? Because that would give us a perfectly good excuse to stop our search whenever we might become baffled; and experience has shown that if we had stopped at other such times, we would have stopped too soon, We cannot reject the axiom of determinism as long as we are to continue in scientific endeavor.
How does the notion of a rigidly determined universe fit into such religions as conventional Judaism and Christianity? None too well. It is introduced into one chapter of the Westminster Confession, in which we are assured that the whole course of events is already established, or determined. To be sure, it is not necessarily, determined by its own past, according to laws that man can discover; it is determined by the deliberate, uncoerced, and prescient fore-ordination of God, who has not fully revealed his plans. Nevertheless, it is all thoroughly determined—even the actions of human will were foreknown and foreordained. But even this notion of determinism did not seem to work, for it is restricted, in other parts of that document, by the equivocal use of names. I know of no better example in any modern religion of western civilization.
Certainly we reject the notion of an intervening Providence when we adopt the axiom of uniformity; we dare not entertain it for a moment in our scientific endeavors. How would it do, for instance, in a case like this: an experimenter gets a set of results that do not satisfy the formula that everybody at the present time treats as a “law.” Instead of saying that the supposed law is not the true one, and then hunting for one to which these results, as well as earlier results, conform, he says, “Well, let us suppose that Providence has intervened in this one instance; and but for that intervention, the old formula would still hold.” If he did so, he would no longer hunt for the more generally valid law; whereas if he sought it he might possibly find it. He must allow himself no excuses for not seeking. But it is also certain that you cannot be the particular object of God’s intervening care if nature invariably observes a set of impersonal laws. It is nonsense to talk as if both of these assertions were tenable at the same time. I have examined a good many attempts to “reconcile” the axiom of determinism with axioms of divine freedom, human freedom, or material freedom; all of them proved to depend on the use of the logical fallacy called Equivocation.
While one simply cannot proceed in scientific endeavor without relying on the axiom of determinism, one cannot enact some of the most distinctive rites of Judaism or of Christianity without rejecting it. While you retain it, how can you expect a successful result in case you should pray for such things as rain, for fair weather, for a fruitful harvest, for the comfort of absent people who are in sorrow, for the recovery of a sick person who is not present, for the conversion of the heathen, for the endowment of the President and the Congress with grace and wisdom, for protection during the night, for the protection of those at sea from rocks, fire, storms, and enemies? It is rather puzzling to hear some of the reasons that the “rational religionist” gives for praying for events which he believes to be uninfluenced by, human desires.
The other axiom of universal science which these typical religions reject is the criterion of the validity of knowledge. Science rejects all propositions as allegations of fact, unless they, can either be validated by common observation, or else inferred, by the rules of conventional logic, from the results of common observation.
On the other hand, the most typical forms of Judaism and of Christianity maintain that there is a third kind of knowledge which is also valid—knowledge that does not depend on observation or on reasoning, but supersedes the knowledge that does. It is upon a claim to the possession of this third, and mystical, kind of knowledge that they base their most effective claims to authority. It may be well to recount some of the various sources of mystical knowledge that have been alleged.
One source is said to be intuition. Its results are supposed to be validated by the mere “feeling of assurance” that attends them. Such a plea is not lacking even in as nearly rationalistic a religion as Unitarianism. If the question at issue is grand enough, such a feeling of assurance may be designated as a sign of Divine Revelation to the individual, or of the Testimony of the Holy Spirit. According to the Westminster Confession, it is by such means that one is assured of the authority of the Holy, Scriptures, and of the correctness of his private interpretation of them.
Now in a religious setting, this doctrine seems to be appropriate. But suppose that while I am engaged in a scientific experiment, I should develop such an inward feeling of assurance about the way my results should come out; and suppose that my observations contradicted the assurance. How would it do for me to proclaim that the facts accorded with my intuition, instead of telling what I observed? A system that admitted such a guarantee of truth would resemble no science that we now recognize as such. I do not mean to say, that the scientist is “right,” in any absolute sense, in rejecting the axiom of intuitive knowledge; but I do insist that he could not accept it without destroying his system,
But intuition and private revelation do not exhaust the list of alleged sources of mystic knowledge. It is claimed that we may also acquire the results of intuition at second hand. For instance, there are the utterances of an infallible prophet or leader; the assertions of an infallible collection of writings; or the official decisions of a priesthood, whose origin antedates its written history, and which is accepted as an infallible fact-finding body in matters spiritual. The claim last mentioned appears to be the most convincing of all; although all of them may reduce, directly or indirectly, to special instances of intuition. I shall not debate any of these claims here; but shall repeat that if a science should accept any of them it would thereby commit suicide.
But what kind of religion would it be that did not claim the possession of any form of mystic knowledge? It would be rationalistic and, I fear, rather uninteresting. Probably, it would not be very procreative, either. A successful claim to religious authority should be put into such form that nobody can successfully question it, and few will care to try. Claims that rest on axioms which are not accepted in science have one quite valuable property: they cannot be decided by scientific tests. Their settlement therefore can never become a scientific problem.
III
I have endeavored to show that these typical forms of religion contradict scientific systems, by denying some of the axioms of the latter, By affirming an intervening Providence and a free human will, they deny the axiom of uniformity of nature. By affirming the validity of mystic knowledge, they reject the scientific standard of “truth.” Now we learn in elementary logic that there is no possible way of reconciling a contradiction. It is futile to attempt a compromise on these points. You either accept an axiom and thereby deny its contradictory, or else you reject it and thereby accept its contradictory.
Are we then to conclude that one cannot be a scientist, and also a religionist? Hardly. The proper conclusion is that he cannot he both at the same time. Such an attempt would be as absurd as trying to play a game of cards according to the rules of Bridge Whist, and also according to the rules of Five Hundred, at the same time. He cannot do it, because their rules conflict. But at any time he tires of one game, he may reject its axioms, or rules, and adopt the other set. The choice between scientific and religious rules depends on exactly the same principle, even though the comparison may seem trifling. The axiomatic rules of scientific activities and of religious activities are alike arbitrary. Any proposition in science is strictly relative to the axioms of the system. It is precisely, so in religion. Reject the axioms of an intervening Providence, of a free will, of the validity of mystic knowledge, and you remove the rational basis from the most distinctive doctrines and practices of these religions. The scientist, while working at his trade, must reject these dogmas, and replace them with a more convenient set of assumptions. This change makes a radical difference in the way he talks about things: it also makes all of his assertions highly conditional. Likewise the assertions of the religionist: they are not absolutely true or false, but are made so by a choice of the axioms by which they are to be tested.
The proposal that one turn from a religious system of discussion and practice to a scientific system and back again has been derided. Philosophers have given the practice a contemptuous name. They call it “dividing the mind into thought-proof compartments,” and thereby make it sound like something that one ought to be ashamed of. But this is exactly, the procedure that we observe whenever we play Contract Bridge instead of Auction Bridge or Five Hundred; whenever we talk (after the manner of Copernicus) about the earth’s rotation toward and away from the sun, instead of talking (according to the fashion of Ptolemy) about the sun’s revolving about the earth; whenever we describe a tract of land in terms of Euclid’s geometry instead of one of the geometries of Riemann; whenever we decide to treat the expression 100 as meaning ten groups of ten members each, instead of twelve groups of twelve members each, or nine groups of nine members each (as we should have to do if we had not rejected the laws of duodecimal and nonesimal number-systems when we adopted the decimal system); whenever we “interpret” a clause in the federal constitution according to the rule of Justice Taney instead of the rule of Alexander Hamilton. In brief, we do just this whenever we decline to entertain, at the same time, any propositions or doctrines that contradict each other. In all other situations this method is considered to be demanded by logic; it hardly becomes the philosopher to taunt us with using it here. If he can invent a system of rules that will give the generality and simplicity of description that the scientist aims after, while bringing into emotions and appreciative attitudes the order that the religionist desires, perhaps, if it seemed to be otherwise all right, all of us might be willing to give it a trial. Meanwhile, he should not murmur if we choose our system according to the use that we intend to make of it at the time. That is exactly what we do, and what we shonid do, until a better way shall appear.
IV
Meanwhile, it may seem that some people are losing too much sleep over the fact that the scientist and religionist insist on talking in different languages, and that the same person varies his language as he turns from one role to the other. For my, own part, I should regret to have to describe my observations of human and animal behavior according to the rules employed by St. Thomas Aquinas, I probably could do it, somehow; but it would make me too much extra work. I should like it no better to have to phrase them in terms that satisfied the tenets of Dr. Harry Emerson Fos-dick’s admirable group, or even in terms of the “strictly rational religion” that has been proposed by Dr. Harry Elmer Barnes. For scientific description is provisional; it is relative to its axioms, and to the operations by which its workable concepts are defined. At the present time, for example, the mere adoption of Professor Bridgman’s operational rule of definition is promising to work as profound a revolution in the biological and social sciences as is now occurring in the field of physics: concepts, propositions, and doctrines that have been fashionable for generations are utterly swept away. Scientific fashions will change whenever its needs are modified; and in the recent past, these changes have been radical, sudden, and frequent. When the scientist finds them called for, he does not wish to consider anything but his own needs. If scientific fashions were a part of any religion, they would probably acquire an emotional significance for religionists; and as Mr. Walter Lippmann has noted, whenever the scientist should consider making a change of fashion, he then would have to take account of its probable effects on the emotions of non-scientists—old people whom he might not desire to offend, and others. All this would be very awkward for him. He should not desire that any religion that is important in numbers or influence should adopt his present system of doctrine as it stands. He might wish to modify or abandon it long before the religionists might be ready, and they might object.
But would it not be just as bad for religion if it should adopt any of the dogmas of science? Religion seems to thrive best on permanent fashions. It is hard to introduce any formal innovations without creating new heresies and schisms. All religious doctrines and practices tend to assume a rigid form in time: what would be the status of a religion which depended, or which seemed to depend, on scientific dogmas that science itself had cast aside? I think Mr. Lippmann is correct in saying that nothing could be deader than those scientifico-religious cults that science has deserted. This consideration seems to have escaped some of the clergy, who have appeared flattered by the approach of a group of distinguished men of science, bearing olive branches in their hands and offers of “reconciliation” upon their lips. Their intentions are noble enough to meet the most exacting standards; but what they have to propose is virtually that the religionists should reject the axioms that distinguish their discourse from other discourse. I do not contend that this would necessarily be a bad thing, but I doubt if many people would recognize these doctrines after they have been so recast.
If this presentation of the case is correct, then religious discussion and scientific discussion cannot blend. I should like to comment on the misgivings that some religionists have developed with respect to that idea. As a scientist, I haven’t the slightest objection against a preacher presenting his case from the standpoint of a romanticist, a poet, or a mystic. If he should then fail to present it effectively, the reason will be that he uses bad poetry, and inferior mysticism. Such a mode of presentation, if properly used, is far more likely to set up a habit of appreciation, of sympathy, and perhaps of acquiescence in certain human relationships than an appeal to reason would be. This fact seems to be rather clearly apprehended by the Catholic group; and much less clearly by the Protestants. Perhaps it is the former group that have the greatest assurance of the validity of mystical knowledge. I suspect that the Catholic leaders, both Roman and Anglican, are now pretty well convinced that as long as the aesthetic and emotional habits of a group are firmly bound to religious practice, their loyalty to the church will not be greatly affected by the formation of new intellectual habits, and particularly, by the pursuit of science. Thus they can afford to be placid. Perhaps this is why Catholic temper now seems to be more cordial toward scientific endeavor than Protestant temper.
If religious discourse and scientific discourse are based on two different sets of axioms, they have in general two different tests of truth. It is therefore as absurd to submit a religious dogma to test by the axioms of science as to proceed conversely. The religionist who invites such a test resorts to poor tactics. He joins, let us say, in the prayer of St. Chrysostom: “Almighty God, who . . . dost promise that when two or three are gathered together in thy Name, thou wilt grant their requests: . . .” What is his evidence—scientific or mystic? Scientific evidence would consist of a census of a world of events, classified (a) according as they were prayed for or not; and (b) according as they occurred or not. When was such a census taken? Was it adequate? Did it yield a correlation between praying and occurrence, non-praying and non-occurrence? If so, how probable is the fulfillment of requests of any given class: one in 100,000; 99,999 in 100,000, or what? Of course, there is no answer to such questions that would satisfy, a scientific standard of proof. No such evidence has been gathered. A religious leader who proposed to gather it would be disparaged. So would a scientist—I, for one, would think that if he were really capable, he could find a better way of employing his time; for even if the experiment were to be properly made, its results would probably have no effect on human practices. If a doctrine or a practice is not actually based upon an appeal to reason, it will not be much affected by the answer to that appeal.
Finally, the religionist may consider the following prophecy in making his plans: scientists generally will continue to work, and to describe the fruits of their efforts, in accordance with the rules that suit them best, and will alter those rules whenever it appears that a change will simplify their work. However they may put the matter verbally, they will continue to employ, the axiom of determinism, and to reject the axiom of mystic knowledge. The religionist can do nothing to stop these practices. Will he allow them to injure his cause?
Whether he does or not will depend, in my opinion, on his ability to see that science and religion cannot blend and cannot logically interact. If he can perceive this clearly, ’ and relate his doctrine to its own proper basis, it will probably persist as long as it supplies a widespread aesthetic, appreciative, and emotional need.
