In this lecture I shall try briefly to describe the actual status of the cosmogonical problem, beginning with factual aspects, and ending with the philosophical questions.

Let us start from the biological theories as outlined in the preceding lecture. We do not yet understand the causes of evolution too well, but we have very little doubt about the fact of evolution; it is even most commonly held that organic life has developed out of what we call inorganic matter in the early stages of geological history. What are the reasons for this general belief? In the last lecture I formulated them negatively: We do not know how life should have come to exist in its actual form in any other way. This formulation leaves silently aside any possible supernatural origin of life; such is the faith in science of our time which we all share. But in order to be convincing our belief in evolution must at least be founded on an analysis of possible alternatives that lie within the frame of natural law.

We believe that living beings can only descend from living beings rather similar to themselves. If, say, monkeys or oak trees did not descend from mammals or trees on our earth they might only descend from mammals or trees outside our earth. It seems that flying saucers do not yet belong to the trusted elements of the scientific faith; in any case they have not led to any scientific theory of an extraterrestrial origin of higher forms of terrestrial life. The theory, however, was put forward more than fifty years ago, that at least the most primitive forms of life might be able to resist the physical conditions of cosmic space, and that in this way life might wander from star to star and might thus be as eternal as the universe was then thought to be. Today this theory seems less attractive, for two reasons. First, no evidence of life has been found so far in meteorites or in any ingredients of our atmosphere that might have a chance of being of cosmic origin. Secondly, the prevailing view of astronomers in our times is that not only the earth but probably the universe as a whole is not eternal but has had a beginning. This view is the main topic of my present lecture.

The age of the earth can be estimated to be of the order of magnitude of four thousand million years, plus or minus one thousand million years, perhaps. How can we give such an estimate?

For the details of this problem I may perhaps refer you to my lectures on the History of Nature. Here I shall only draw the main line of the argument.

Radio-active atoms decay at a fixed rate. Of a given amount of say, polonium, one half will be transformed into lead within 137 days; the corresponding time, the so-called half-life, is a little less than 2000 years for radium and about five thousand million years for uranium. Now there are typical minerals of uranium in the crust of the earth that contain an amount of lead which is greater in proportion to the age of the geological stratum in which they are found. This serves as a kind of clock which was set when the mineral was formed, and which gives us its actual age. Even the age of terrestrial rocks in general can be estimated in a similar way, using its average content of uranium and of lead isotopes. Indeed, the earth cannot be much older than the half-life of uranium, considering that there is still much uranium present on the earth and that no way is known in which uranium might have originated on the earth once the earth was existing as a planetary body.

Of course, this line of argument presupposes that the known laws of nature have held quite accurately all through the time that is to be measured by the uranium clock. Can we be certain that the decay rate of uranium has stayed constant through geological history? Our present answer is: this is by no means a priori certain, but it seems highly probable, considering that many different methods of estimating cosmic ages converge towards the same result. Again, I leave out the details. The point of principle is that we consider laws of nature as subjected to empirical proof which in many cases—as e.g. if we are interested in the far past of the universe—cannot be given by actual experimenting, but by accumulating the probability of a system of hypotheses. The hypotheses may lead to an ever increasing number of consistent predictions and interpretations of such facts as must be considered to be documents of the far past; and this is all we can hope for.

If the earth has a finite age it must have originated somehow, and in the ideas about the way in which it originated, modern astronomy, after some vacillations, has returned to views very similar to those proposed by Kant. Again, I leave the details aside. In any case, it seems probable today that the planets have formed out of a nebula that surrounded the sun. One of the most active investigators of our days, G. Kuiper at Yerkes Observatory, thinks that some denser parts of the nebula condensed further under the influence of their own gravity; this is precisely what Kant thought. I have proposed a slightly different theory in which the planetary bodies were formed out of dust that originated by chemical condensation of the heavier elements within the nebula. I do not know myself which theory I should prefer in our actual state of knowledge. In any case you see that the possible field of controversy is about points of the detailed mechanism but not on the principle.

But in our days such a theory is only a minor step in the ladder of cosmogonical questions. Whence did the nebula come? Part of which greater system is our planetary system? These questions to which Kant could only offer highly speculative answers are better understood today than the particular model of the origin of planets.

Let us first glance at the distribution of objects in space as we know or suppose them to exist now: the sun is quite an ordinary star, similar in its physical characteristics to the fixed stars which we see at night. One hundred thousand million (10¹¹) stars of this type, roughly speaking, form the system known as the galaxy, the bulk of which is not seen as individual stars but forms the cloudlike ring of the milky way. This system is something like a disk, not unlike a huge simile of what Kant considered to be the original shape of the planetary nebula. We are inside it, as we are inside the planetary system; thus the milky way, just like the zodiac, is a great circle on the sky. The similarity of shape implies a similarity of origin. The galaxy still contains large amounts of interstellar matter, in the state of gas or of dust, and it is natural to suppose that there was a time in which it consisted of an extended gas throughout. Then parts of it will have condensed into stars. An intermediate step of this condensation may have been the formation of the smaller, separate nebulae, one of which gave rise to our planetary system. We do not know positively whether other stars have planetary systems, too. But our actual knowledge contains no reason why at least some of them should not have planets. Thus astronomy has nothing to say against the speculations on life and on intelligent beings on other stars which are so popular today. Only the other planets of our own system seem less apt to carry higher forms of organic life than the earth, and thus the probability that intelligent beings from other stars will visit us may not be overwhelmingly great. But I should not dare to make a definite prediction.

The part of space that can be penetrated by our existing telescopes contains about a hundred million systems similar to our galaxy. In general they are termed spiral nebulae because many of them, our galaxy included, have a spiral structure. I shall not go into any detailed description of these systems. But I must spend a little time with the question of what lies beyond them.

It is not the unknown answer to this question but its methodological structure that should concern us here. The question is of a puzzling character. Of course we do not know what is beyond the limits of our knowledge; this is the definition of the word limit. Still we are convinced that there is something beyond these limits. How can we be convinced; do we have knowledge beyond our knowledge? In any case, we should say, there must be space outside the part of space we know. But how do we know that? Do we have an a priori knowledge of cosmic space, five thousand million light years from here? In our century Einstein has surprised the scientific world by assuming that perhaps our a priori assumptions on space are wrong. He offered the hypothesis that space itself has a mathematical property which can be imprecisely visualized by calling space itself curved. In a curved space a body flying along a straight line may return to its starting point. I think Einstein’s particular assumption on cosmic space is neither proved nor disproved up to this day; in any case we know no reason why we should not consider it to be a theory as plausible as any that has been proposed. Now, if it is true, the question: what is beyond the limits of our knowledge? may sooner or later turn out to have a surprising answer. Beyond a certain limit there may neither be other galaxies nor will there be empty space but there might be the very same galaxies as on this side of the limit. If I may slightly simplify the historical facts, the history of geography may serve as an example. After Marco Polo’s travels to China and after the discovery of America the Pacific was the limit of our geographical knowledge in the East and in the West, and when Magellan crossed it he found the same continents on both sides of the border.

The philosophical view about the nature of space that lies behind Einstein’s hypothesis, is again a topic for my second lecture series. I had to mention it here for two reasons. Materially seen, it belongs to my intermediate subject matter, the distribution of objects in cosmic space; a subject matter I am leaving at this point where it ends, as usual, in an unproved hypothesis. Formally seen, Einstein’s proposal shows what unexpected answers may be given to age-old questions. This we should keep in our minds when we now turn to our main topic which is not cosmic space but cosmic time.

I gave an estimate of the age of the earth. How old is the sun? How old are stars in general? How old is our galaxy? How old are galaxies in general? Is there anything like a well-defined age of the universe?

There has been a very marked convergence of empirical facts towards a universal statement which I shall express in two forms, one cautious and one less cautious. Being cautious we are entitled to say: There is good evidence for the assumption that there is a common time-scale of the evolution of most cosmic objects with a characteristic time-constant of roughly five thousand million (5.10⁹) years.1 Less cautiously we may yield to the temptation of saying: The universe has a finite age of x.10⁹ years.

Again, referring you to more explicit presentations elsewhere, I shall mention the empirical evidence very briefly.

The sun is radiating away energy at a high rate. Today we are convinced that we know the sources of this radiation; it is the consequence of nuclear processes in the deep interior of the sun, processes not quite unlike those that take place in the hydrogen bomb. In fact, astrophysical theories on nuclear processes in the sun may have contributed to the idea of making a hydrogen bomb. Now the amount of hydrogen present in the sun is finite, and hence the sun cannot uphold its radiation indefinitely. Since the sun is still radiating it must have begun to radiate a certain time ago. This beginning, according to our present knowledge, may well have been x.10⁹ years before the present era. The time span might also have been a little longer, but not much longer.

Similar arguments apply to all stars which we can see today. Recently it has become fairly certain that stars are being formed out of interstellar gas even in our time; but most stars seem to have an age not too far different from that of the sun, and there is no positive evidence for any age greater than the highest estimate indicated by the “x”.

The inner dynamics of a galaxy—which is a rotating bulk of stars and interstellar matter—is not yet too well understood. Still, it seems quite plausible that galaxies themselves should develop within a similar time-scale. The strongest argument for this view is what I might call the outer dynamics of galaxies, that is what we know about their relative motions. The famous red shift of their spectral lines is most naturally explained by the assumption that they are all receding from each other, not unlike the pieces of an exploding bomb-shell. If Einstein is right in considering cosmic space to be curved this implies that its curvature is a function of time, the total volume of space increasing with time; in this sense the term “expanding universe” is often used. But we do not need to express the natural interpretation of empirical facts in the terms of Einstein’s hypothesis. If we consider space to be Euclidean we may just speak of an expanding motion of all known matter within space, and we will have described the same observations. It must be admitted, however, that the motion itself is not seen. What we see is a red shift of spectral lines. Physics, as we know it today, does not offer any other natural explanation of this red shift than the assumed expanding motion; but whoever does not like the theory of expansion seems, in the present state of our knowledge, free to invent new laws of physics that would explain the red shift in a manner he likes better.

There is, however, one additional argument for considering the expanding motion as real, and in my view this argument is very strong. If there is a real motion, it defines a time-scale. Assume the comparison with an exploding bomb-shell as correct: then, if you can measure the distances and the velocities of the fragments in a given moment, you can calculate at which moment of time the explosion took place. Now the distances of galaxies are roughly known, and the red shift, if interpreted as indicating a velocity, gives you the numerical value of this velocity: hence we can calculate the time of the first explosion. It turns out to be roughly x.10⁹ years ago. Thus, in a completely independent way we have come back to the same age as in our earlier estimates. This result would be a very surprising accident if the red shift were not caused by a real motion. I think this argument has convinced most astrophysicists, and it is the real reason for the common use of the term “age of the universe”.

Of course, admitting this time-scale as meaningful, we are still very far from having a consistent theory of what happened x.10⁹ years ago. How broad a scale of possible interpretations is still left I may indicate by mentioning two extreme views, both of which have found defenders. The metaphysically most conservative hypothesis, if I may say so, is that of the so-called theory of continuous creation. It says: Our own galaxy may be x.10⁹ years old, and in any case this age indicates a characteristic time-scale of cosmic events; but in the universe taken as a whole nothing of any importance at all happened x.10⁹ years ago or at any other time; the universe has always looked exactly the same way it looks today. The apparently most radical hypothesis says on the contrary: Not only the whole universe began to exist x.10⁹ years ago; even time itself did not exist earlier, for before the world there was no time. I want to discuss these two views in more detail.

The theory of continuous creation has acquired a good deal of fame, mainly due to the gifts of imagination and of popular presentation of my honoured colleague Fred Hoyle. It satisfies the desire which, I think, must be termed metaphysically conservative in the scientific situation of our century, to reconcile the new empirical data with the traditional view of modern science that the universe has no beginning and no end in time. This is achieved by assuming that at any time and in any place some matter is originating (or “created”) out of nothing. This matter will at times condense into galaxies. Looked at from the matter originating in other parts of the universe, matter is created with an initial velocity such that in the course of time it will drift into the infinity of space, thereby vacating the place where it originated for new matter to be created ex nihilo. The so-called expansion of the universe is nothing but this continuous drifting-away of matter continuously replaced, and the idea of an infinite space, which is absolutely necessary in this view, serves to secure sufficient room for the newcomers. This process is considered to have taken place since infinite time and to go on for infinite time, and thus on the average the universe will never change.

This is a very clever theory indeed. I know of no particular empirical facts that would favour it more than its competitors, and personally I admit that I hesitate to admit it as true or even as very probable. But my reasons are reasons of methodological taste on which a discussion cannot be decided. My own conservatism lies in the field of method rather than of metaphysics. Thus I would not be much shocked by having to admit a finite duration of time while I like to trust empirically established laws of nature as long as possible, not because I think them to be better than other possibilities but because I do not easily trust our scientific imagination to hit upon the right idea on merely speculative arguments. Thus the law of conservation of matter which in modern physics is identical with the law of conservation of energy is, in my mind, closely connected with the general structure of our existing physics, and we do not know how much we shall have to change if we sacrifice it to the idea of a continuous creation ex nihilo. There are moments in the history of science in which such a sacrifice of empirically so far well-established traditional views is needed in order to take a great step forward towards a new and more comprehensive harmony between theories and facts. I want to be forced towards such a step, however, and I see no empirical reason which would force me here; for the idea of an infinite duration of time is certainly not an empirical one.

The other extreme does not now admit of any empirical proof either; it is only intended to show the possibilities for which we ought to be prepared. If we stay methodologically conservative and assume no new production of matter out of nothing, the matter now contained in the visible part of the universe must have existed x.10⁹ years ago, but in a far smaller space. This state of affairs cannot be extrapolated into the past beyond the time at which our simple model of linear expansion would lead to the absurd consequence that all matter was united in one mathematical point. Several suggestions of what was in the beginning have been made. Some of them assume an infinite time before the expansion which we see now started, e.g. a nearly stationary condensed Einstein universe. Others give the world a more or less abrupt start. I think there are good reasons for thinking that there are no strictly periodic solutions of the cosmological problem. These reasons are connected with the second law of thermodynamics which I have discussed at length in my History of Nature to which I may refer those who are interested in the problem. Yet besides this kind of very general statement I think nothing can be said positively about these events milliards of years ago, in any case on the basis of our existing knowledge. The thing I wanted to point out by speaking of another extreme is that modern theoretical physicists certainly do not need to be less critical of their own concepts than St. Augustine was when he felt that the concept of a time that elapsed before there was a world might be a completely meaningless concept. A modern positivist would not argue differently.

To sum up: I do not know whether the world has had a beginning, but it may well have had one. If, however, it did have a beginning, then time may have had a beginning as well. How this beginning would have to be imagined I do not know; and here it may be good to remember Einstein’s theory of space, only as an example of unforeseen possibilities.

The rest of this lecture is to be devoted to the repercussions of this scientific situation on our general topic of human views on cosmogony and creation. I should like to begin by telling you a very simple story which I experienced myself and which, if I remember rightly, first made me think of the questions on which I have now been speaking so long. In 1938 when I was a young theoretical physicist in Berlin, I gave a paper in the “Physikalische Kolloquium” of the university of Berlin on the transmutation of elements in the sun. I had then just devised a particular nuclear reaction chain that could serve as source of energy for the sun. It was the so-called carbon cycle which was found independently and worked out in detail far more precisely by Bethe in the same year; from our present point of view it is the right type of reaction but not the one that actually plays the main rôle in the sun. In any case I was quite proud of my discovery, and in order to show its plausibility I stressed the point that it gave the sun a possible age which would fit in very well with the age of the universe defined by the red shift which at that time was a rather recent idea. On this point, however, I met the violent opposition of the famous physico-chemist Walther Nernst who belonged to an older generation and who then held the chair of physics in the university. He said, the view that there might be an age of the universe was not science. At first I did not understand him. He explained that the infinite duration of time was a basic element of all scientific thought, and to deny this would mean to betray the very foundations of science. I was quite surprised by this idea and I ventured the objection that it was scientific to form hypotheses according to the hints given by experience, and that the idea of an age of the universe was such a hypothesis. He retorted that we could not form a scientific hypothesis which contradicted the very foundations of science. He was just angry, and thus the discussion, which was continued in his private study, could not lead to any result; Professor F. Debye in whose institute I was then working and who had accompanied us to Nernst’s room finished it by the Salomonic remark: “Look, Herr Geheimrat. Dr. von Weizsäcker is interested in the particular problem of the energy sources of the sun, and you are interested in the problem of the universe as a whole; thus there is no contradiction between your views. Give him time. He is a young man, and if you are right, he will come to share your views in the end.” Thus we still arrived home at the right time for dinner.

What impressed me about Nernst was not his arguments, in which I am afraid I still think there was no substance; what impressed me was his anger. Why was he angry? What vital interest of the man Walther Nernst who was born in the late 19th century and who was certain that he would die in the 20th, what vital interest of this man could possibly be violated by the possibility that the universe might not have existed since infinite time but that it might have come into existence five thousand million years ago? I realized that he was even more angry when I added that according to the second law of thermodynamics every finite part of the world had a finite number of possible states through which it could go just once and that, as our world was still changing today, it was very probable that it had not existed since an infinite time. He flatly denied this consequence of the second law. I have to admit that the application of the second law to the universe as a whole contains some intricate problems and cannot be done in a naive manner. But I easily saw that it was not this scientific difficulty which moved him but some kind of horror in face of the thought that this world might come to an end. The universe just had to be a thing to which concepts like age should not be applicable.

Reflecting on this attitude I wondered how a Platonist or a medieval Christian would react to the same scientific theories. The answer seemed clear. Neither the Platonist, believing in the immortality of the soul, nor the Christian, believing in a resurrection on a new earth, under a new heaven, would be troubled by the discovery that this material world even for immanent reasons could only have a finite duration. I think I was not mistaken taken in supposing that Nernst, as was usual with scientists of his generation, was not positively religious, and the conclusion seemed—and still seems—natural to me that in his frame of mind the everlasting universe had taken the place both of the eternal God and of the immortal soul. I did not feel tempted to take the attitude of an apologetic theologian here. On the question of personal immortality my own views were very definitely agnostic. I had realized the basic difference between the philosophical idea of an immortal soul and the New Testament’s belief in a resurrection of the body. Their merging in many trends of Christian theology I had learnt to ascribe to the historical fact that much of Christian theology is an interpretation of the gospel by the concepts of Greek philosophy. This understanding of the historical origin of some of our most common religious beliefs had left me in an even more sceptical state of mind with respect to the traditional interpretation of religion than natural science alone might have done.

But what struck me about Nernst’s view was this: even if immortality was desirable, what would it help a man who did not believe in the immortality of the rational soul or in the resurrection of the feeling person, to believe in the unending processes of nature instead? Would stars and atoms feel and think in his stead after the irrevocable death of the ego? I think, a deeply irrational trait of scientism was revealed in his view: the world had taken the place of God, and it was blasphemy to deny it God’s attributes. Here was the place where I first realized that scientism contained an element which I now would call the secularization of Christian religion. This will be the theme of my next lecture.

Here I only add a few remarks on the particular importance of the question about the age of the universe for the dogmatics of scientism and of Christianity. I think, it may easily be seen rationally that this importance is overestimated on both sides, the main point being the irrational one of which I have been speaking.

Take scientism first. Western scientism does not seem to stick any more to dogmas about infinity. This was different in the 19th century. But western scientists have thoroughly learned the lesson of empiricism, and their dogmatism today is one of scientific method rather than of particular positive statements. The situation is different in the communist ideology. This ideology has not avoided the temptation of dogmatizing particular positive views. This fact became best known in the episode of the officially enforced non-Mendelian genetics which, I hope, now belongs to the past. But as far as I am informed the infinite duration of the world is still considered to be an irremovable part of dialectical materialism. Rationally seen, there does not seem to be an inherent necessity for this view in Marxism. Neither the deep insights of Marx on the importance of the material—i.e. of the economic—element in human history, nor the criticism of that philosophical attitude which is termed idealism in the Marxist language, seem to make this view logically necessary. I think the important elements of Marxist philosophy would be far more convincing to a modern mind if they were clearly separated from scientific dogmas that were modern in the lifetime of Marx and Engels and are out of date now.

This decision may well be left to those who have the right to decide what is true in Marxism. Yet I have felt tempted to ponder upon the reasons that make this modernisation of Marxism such a difficult task, and I have found three closely connected possible reasons. First, Marxism does have to play the rôle of religious dogma to its own believers, and religions always find it difficult to admit that they are not infallible; the fact that Marxism considers itself to be science does not seem to change this psychological problem. Second, I think there is a positive attraction of the traditional view to the Marxist mind, on lines very similar to those I described when speaking of Nernst; infinity is something like a symbol of the naturalist creed. Third, communism, being a fighting community and carrying on the doctrine of class conflict, considers it a political necessity to take a definite part in ideological controversies; and seeing that Christian churches consider the creation of the world in time to be a religious truth, communists are afraid of the detrimental political consequences of admitting that the Church may not just have been a bulwark of superstition in this respect.

If, however, we now turn to Christian theology, I do not see a rational necessity for Christians to believe that the world did have a beginning in time. As long as the text of the Bible was considered to be the ultimate authority even on scientific questions the situation was different. But this battle was decided in the time of Galileo, and today there are only very few Christian groups left that would read the relevant passages of the Bible like a scientific textbook. Christians today find no difficulty in admitting that the world is more than the mere six thousand years that a chronological evaluation of the Bible text would allow. So I do not see why it should be more Biblical to assume that the world is five thousand million years old than to assume that it has existed since an infinite time. Once you have reached the theological level of St. Augustine, saying that divine creation is not a process in time but an act by which even time is constituted, there seems to be no necessity that God should create a finite time rather than an infinite one. Neither do I share the view that science, considering the possibility that the world may have had a beginning in time, thereby contributes to the classical proof of the existence of God as the ultimate cause of the world. If I am right in thinking that there is no philosophical reason for extending the concept of time beyond the existing world, it will be consistent also to limit the concept of a cause to application within the world. If, admitting a beginning of the world, we see no meaning in the question what happened in the time before the world existed, we may see equally little meaning in asking what pre-existent thing caused the world to come into existence. Science fits in with religious agnosticism at least as well as with the dogma of creation.

Meditating about the reasons that make the idea of a beginning of the world in time so attractive to confessing Christians, I find the strict parallel to the reasons I suspected for the Marxist predilection for the opposite view. First, Christians, too, like to find their religion infallible in some points, especially after having suffered so many setbacks in the conflict with modern science. Second, the finite duration of the universe serves as something like a symbol of the Christian understanding of human history; and since I think this understanding is to be taken most seriously I want to return to this point in the next lecture. The third reason, being a reason of practical politics, may be less important with the Protestants of our time who have submitted to the liberal idea that religion and politics ought to be kept apart; in Roman Catholic circles another tradition is still alive to which such considerations are not altogether foreign.

If I am right in seeing this complete parallelism I think I am right in refusing to take sides in the material question of infinite time, at least so long as no strong scientific arguments can be produced. I think there is a slight advantage for the idea of a finite time, but probably the truth is different from everything we imagine today. The point that ought to concern us then is no longer what we believe on this question but what human attitude is expressed in a propensity for this or that belief. I have already touched on this question, and it will be one of the subjects of the next lecture.

• 1.

  Since these lectures were given, recent results seem to have shown that our quantitative estimate of the time-scale has been inaccurate. Ten or fifteen thousand million years are probably a better guess. I express this uncertainty by speaking of x.10 years in the printed text. I think nothing in principle is changed by this readjustment of a scale.