We have been speaking of the origin of heaven and earth, or, to put it in modern words, of the sun and its planets. But remember what I said about the Old Testament. Heaven and earth are only the stage. Grass and fruit are needed for the scenery, fish and fowl, beasts and cattle for ‘extras’, and the drama to be played is the history of man. God made every one of them after his kind. Science that had set out to explain the origin of earth and stars according to the laws of nature had to face the question how life on earth has come about.
I have never studied the history of biology in any detail. Hence I shall concentrate on the problems of principle which come to the forefront whenever the questions in these lectures are discussed with intelligent biologists of our time.
The problem of the evolution of life can be divided into three or four subquestions:
- Has there been an evolution at all?
- Can we give a causal explanation of evolution?
- Do plants and animals form a separate realm of nature or does evolution bridge the gaps
- between inorganic matter and organic life?
- between brute animals and man?
These questions form a series of increasing importance for our problem of the relation between religious and scientific thought. Speaking of earlier times, I have tried to speak their language up to the point where I may have become obscure to a modern mind; but I did not see how we could hope to understand the questions inherited from those times without understanding their frame of mind. This is an excuse, addressed to the modern scientist. Now I have to make a similar excuse to the religious person. Speaking of modern science we shall have to face its ways of thought as clearly as possible. If you do not see what the scientific mind can achieve you will not understand our century. I have tried to speak like an ancient Jew and like a Platonic philosopher; I shall now speak like an ultra-scientist. And, strangely enough, in all three cases I do not need to say many things which I do not believe.
The first question: has there been an evolution at all? was settled in the 19th century by a bluntly positive answer. The steps of the ladder leading to this insight are simple enough. The first step is classification. As I said in the third lecture, already Genesis 1 shows a marked classificatory interest; in this respect the first chapter of the Bible is early science rather than myth. Greek biology, of which, for example, the wonderfully detailed studies of Aristotle are preserved, did a great deal of careful work. How, then, did these many species arise? Have they been in existence since infinite time, as Aristotle thought? Did God make them all in the first week, each after his kind? In the 18th century documents on earlier forms of life on earth became well known, and the idea that the earth has slowly formed entered into the scientific mind, progressing from astronomy to geology and biology. Early in the 19th century Cuvier still thought that there were sequences of geological catastrophes, each of which was followed by a direct creation of new types of living beings. This view is a biological analogy to Newton’s and Bentley’s view of the planetary system. It could not be upheld against the increasing amount of evidence for a more or less continuous transformation of organic forms. Most of the credit for this idea is probably to be given to Lamarck; Darwin’s and Huxley’s spectacular public success was in this question only indicating that the general public had become aware of the problem. Today, evolution as such is no longer questioned by serious biologists. And religious thought has found it easy to acquiesce along the lines I indicated when speaking of Kant: why should God have refrained from using the natural laws of growth and transformation in creating living beings?
But once the fact of evolution is admitted, the question of its causal explanation cannot be evaded. We are accustomed to plants and animals descending from parents similar to themselves. We are accustomed—that means that we cannot explain this fact either. But the question of the causality of events becomes more outstanding when events become known with which we are not acquainted; the well-known is the thing whose mysterious nature we realize last. From the modern point of view, however, we have to combine both questions: how is ordinary heredity to be explained and how the apparently even higher achievement of life, its upward evolution?
Here I first want to discuss a question of principle. Earlier views on organic life which have not completely disappeared even today would have led to a negative answer to the very attempt to ask such a question. One would have said: Life cannot be explained mechanically. Life is an entity of its own nature. It is not to be understood by concepts of causality but by concepts of finality. You will never be able to explain how an eagle’s eye, a dog’s nose, a lion’s tooth have come about. But you can easily see what they are for: the eye is there for seeing, the nose for scenting, the tooth for biting. Explanation in the realm of life can only mean to tell us the what for, not the whence. This view seems to be favoured by the great authority of Aristotle, and it very naturally leaves the gap for religion: living beings are as they are because God made them such.
This view, I think, is wrong. But we shall learn a good deal by trying to say precisely in what respect it is wrong. For the flatly opposite view held by the popular philosophy of many modern biologists is, as I am afraid, no less wrong.
What is wrong in this view is not that it uses final concepts at all but that it opposes them to causal explanation. In this it is only in a very limited way warranted by Aristotle’s own views which, as usual, are closer to common sense than anything believed by later thinkers. Let me make a few remarks on his famous four causes. This excursion may seem to lose time, but in fact it may gain time for the modern problems by clarifying some concepts in advance.
Aristotle applies four concepts to every existing thing which he himself calls its four archai, its four beginnings or origins. In Latin they are called the four causes: causa materialis, formalis, efficiens, and finalis. As examples, look at this glass and at this apple-seed. Both have a causa materialis, the matter they are made from or consist of: the glass tumbler consists of the particular mixture of silicate we call glass, the apple-seed consists of a complicated mixture of organic compounds. Both have a formal cause, a form. Here remember what idea or form meant in Plato’s philosophy; it is the answer to the question: what is this thing? Thus the form can to some extent be identified with the species; the word species in fact translates the Greek eidos or idea into Latin. The form of the glass tumbler is that it is a glass tumbler; if you want a definition I say it is a cup-like thing in which liquid can be contained for drinking. The form of the apple-seed is that it is an apple-seed; you will permit me not to bother you with a verbal definition of an apple-seed.
If you think that these two archai or causes are highly trivial you are perhaps right. Aristotle wanted to formulate the simplest statements which can be made saying what a given thing is, and simple statements tend to sound trivial. Philosophy on the extremely high level of methodical self-control on which Aristotle is thinking begins by trying to be aware of what we really say in our apparently trivial everyday speech. Thus everything admits of at least two questions: what is it?—the answer is its form, its species, and: what does it consist of or what is it made of?—the answer is its matter. Matter in the Aristotelian sense thus is not an ultimate substance, but a concept of relation: the matter of the glass tumbler, the chemical substance glass, has a form and a matter of its own: to the chemist, for example, its form would be indicated by its chemical formula and its matter might be the atoms of silicon, oxygen and other elements. The atoms in their turn have a form described by atomic physics and a matter called elementary particles. What about the elementary particles? Let us wait for the second series of lectures.
Now most things admit of two further questions: how has this thing been made or come into being? and what was it made for? This glass tumbler was perhaps made by a glass-blower; in this case the glass-blower is its causa efficiens. The apple-seed was not made by human design. It grew on an apple-tree. The apple-tree is its causa efficiens. This glass tumbler was made for your Gifford lecturer to quench his thirst. This is its causa finalis. This apple-seed grew in order that there should be a chance of another apple-tree growing out of it. This is its causa finalis.
Here we reach the critical point. The modern biologist is tempted to say: now Aristotle in the end has turned out to be the metaphysician we always suspected him to be. The apple-seed has not grown for any purpose; it just grew according to the laws of nature. But this criticism is only half correct. I admit that even Aristotle could not quite avoid metaphysics. He could not explain the things we see quite without assumptions about entities which we do not see. Just as the modern scientist believes in such things as laws of nature which nobody has ever seen and which cannot even ever be proved strictly, Aristotle believes in some ultimate finality in the world. But as we can offer many examples in which the visible events follow very closely the laws which we have stated hypothetically, Aristotle can offer many examples in which things behave very precisely according to his hypothetical ultimate finality. Our simple examples are in the fields of physics and astronomy, his are in the field of biology. If there were no apple-seeds, apple-trees would soon disappear from the world; if eagles had no eyes, dogs no noses, lions no teeth, they would be unable to survive. If you use the final concepts only as a description of the actual functioning of all these organs you stay within a simple phenomenology of organic life. Modern biologists avoid the word finality because they do not want to imply the metaphysical hypothesis of conscious design, and they are certainly right in that. But in words like “function” or “use” of an organ they describe exactly the phenomena to which Aristotle’s concept of causa finalis pointed.
Thus, if we could stay within the framework of pure phenomenology there would be no conflict between causal and final explanation. According to Aristotle we do not have to choose between causal and final explanations. On the contrary, every thing that has a final cause, an end for which it is made, also has a causa efficiens, for it cannot have sprung out of nothing. Just because lecturers want to drink water, a glass-blower was needed to make the glass. Since the eagle needs an eye there arises the task for the physiologist to understand how this necessary organ could grow in the embryo in the egg according to the laws of chemistry. This is very mysterious but it must have happened somehow. The idea that the laws of physics and chemistry should have been violated in this process of growth is, I think, not of the slightest use for a better understanding of the usefulness of the eye to the eagle. Leibniz pointed out that a clock fulfils its maker’s design not by disobeying the laws of mechanics but on the contrary only because these laws are followed strictly by its springs and wheels.
Yet, by tacitly introducing the modern concept of strict laws of nature instead of Aristotle’s causa efficiens, I have disturbed the balance between causality and finality. Aristotle’s causa efficiens is guided by the potentiality of the end for which the thing is produced. The glass-blower knows that tumblers are needed and makes them for that purpose. The apple-tree, not having a rational soul, subjectively knows nothing of its objective ends, but Aristotle needed the concept of the dynamis, the potentia, i.e. what a thing by its own nature is destined to become, if he wanted to describe the facts of organic life. But if every event in this world is determined unambiguously from the preceding state by strict laws of nature, no space is left for additional determinations by design or by any other principle. This is a difference not always realized by philosophers who try to harmonize finalism with physics: civil laws determine only part of the actions of men, thus leaving space for other agencies; laws of physics, as modern times have come to understand them, are sufficient to determine the events, given the initial state. Civil laws can be supplemented; laws of physics, if not leading by themselves to the result we see, must have been broken. Whether quantum-mechanical indeterminacy will change much on this point I am inclined to doubt. Thus modern science seems to put before us this alternative: either biological finality is nothing but another expression of what follows from physical law; or the laws of physics do not hold in living organisms, at least not without exceptions.
Here we encounter a situation exactly similar to that left to Kant by Newton. As long as physics and chemistry offered no established hope of ever being able to explain the wonderful teleology of living organisms, there was a gap in science that clearly seemed to point to direct divine creation. Now the degree of complication in organisms is far higher than in the planetary systems. While the system of the planets is certainly beautiful and fairly well-ordered, it contains no element that calls for comparisons to works of human instrumental design or to that design itself like the organs and instincts of living beings. It was Darwin’s ingenious idea that chance and natural selection in the struggle for survival might explain the apparent finality in life. He introduced a truly historical way of thought into the science that is concerned with the history of nature. We should not be surprised, he says, to find only forms of life that are fit for survival; if they had not been fit they would not have survived.
Of course Darwin could not strictly prove his theory. Such a proof does not exist up to this day, and it may be doubted how it could be given except negatively, by excluding all other possibilities. First of all, it is always difficult to prove theories about past events that cannot be repeated experimentally. Even Kant’s theory of the planets cannot today, after two hundred years of astronomical discoveries, be considered as finally settled, in any case in its details (see following lecture). Even a convincing argument showing that Darwin’s theory might possibly be correct is not quite easily given, if we consider the quantitative aspect. Can chance supply a sufficient number of steps towards higher organisation within the available age of the earth which we now consider to be about four or five thousand million years? There are still biologists who flatly deny that this can be possible, while others—I think the majority—feel it might very well have happened in the way proposed by Darwin, taking account of the more modern concept of mutations.
I am not going to discuss this calculus of probabilities. I only want to point out what kind of arguments may enter the discussion by giving one example that always strikes me for its existential meaning to our human pursuit of happiness. Most organs and properties of living beings serve to keep the individual alive, like everything connected with eating, self-defence, etc. There are others that preserve the existence of the species, like the organs and instincts of sexuality and those by which the offspring are protected. But I think that there are also properties which serve to promote evolution. A species which—by chance, as Darwin thinks—may have acquired a property which makes this species develop faster than others towards different forms, will thereby have an advantage in the struggle for survival; thus properties of this kind will tend to survive themselves and perhaps become common. Now I feel that one of these properties is mortality, or, to speak more strictly, shortlivedness. The more generations a species produces per unit time the more different mutations or combinations of mutations will have a chance of being tried out. There is a strong selection pressure—to use this technical term of selection theory—favouring a good protection of the lives of young individuals; thus the love of parents for their children is common among higher animals. There is, however, no selection pressure favouring the survival of old individuals once they have generated and protected a sufficient number of children. On the contrary, they now become useless eaters. Thus love of adult children for their parents is very rarely seen in animals; the far wider and deeper vision of man seems to be needed in order to understand that caring for the old is a meaningful task. In plant and animal life, a short remaining lifespan of the old ones is favourable for the species. Perhaps the natural process of ageing would never have developed without this selection pressure, for I see no biochemical reason why individuals should not be possible that would stay alive indefinitely if not killed by force, just as a species can stay in existence for a million or even a hundred million years.
For our further consideration I shall treat Darwin’s theory as if it were true—not because of a positive proof but because I have never heard a positive argument for its impossibility that looked convincing to me. I thus use Occam’s razor: why look for additional causes of evolution as long as we do not know that those given by Darwin are not sufficient? As I pointed out when speaking of Copernicus and the Greeks, Occam’s razor can be a very misleading instrument. We should never use it dogmatically. Here I only use it to simplify the language when I try to explain what could be the consequences of taking modern scientific biology as seriously as it deserves. If a theory different from Darwin’s, but still lying within the conceptual framework of modern science, should be found, probably not much would be changed in my further arguments.
I come to the third question which is twofold in itself: the question about the possible limitations of evolutionary theory at the lower and upper end of plant and animal life. Has life developed out of inorganic matter? Has man developed out of higher animals, say apes?
I think the answer to both questions as given by modern biologists is: with high probability yes. The natural production of amino-acids under conditions similar to those that may have prevailed on the earth long ago, has been proved in the laboratory. How proteins may have formed out of these amino-acids, and how living beings may have formed out of or together with the proteins is not well understood, but who can deny the possibility? And how else should life have begun after the earth had been formed from gas and dust?
Similarly we speak about man. Geological evidence shows that man is a newcomer on the earth. Discussion may be possible whether he is one or ten million years old; if the complete age of the earth were compressed into the twenty-four hours of a day this means discussing whether man arrived half a minute or five minutes before midnight. Whence should he have sprung if not from the higher animals? Perhaps the ape may still be doubtful as an ancestor. Not long ago I read in a newspaper that new results had made it probable that man did not descend from the ape; looking more closely into the matter the new theory turned out to maintain that he descended from the half-ape.
Perhaps more remarkable than these views of modern science is the way of arguing which I could use in order to make them plausible. Where else than in inorganic matter should life have its origin; where else than in animals should man find his ancestors? We see no other possible origin. But neither did earlier generations see another origin. Their conclusion was, however, different from ours; they looked for an origin in the sphere of things invisible. It is not by its conclusions but by its methodical starting point that modern science excludes direct creation. Our methodology would not be honest if it denied this fact. We do not possess positive proof of the inorganic origin of life or of the primate ancestry of man, perhaps not even for evolution itself if we want to be pedantic. But I confess that I would consider it quite useless to deduce any scepticism with respect to evolutionary views from this argument. Science has been successful in this kind of generalisation so many times that I would be afraid of repeating Bellarmine’s and Bentley’s mistakes if I were to doubt this most natural hypothesis on the one evolutionary track leading from the atom to man. I am to ask quite a different question: do we know what we mean by such a hypothesis? Strangely enough it is easier to formulate and even to prove scientific views than to say in clear concepts what we mean by the words we use in such a formulation. The second series of these lectures will be full of this kind of analysis of meaning a posteriori.
If life grew out of inorganic matter, and if natural selection suffices as an explanation of evolution, it is a very natural additional assumption that the known laws of physics and chemistry hold within organic bodies without any exception. This assumption is by no means necessary. For example, there might be laws of forces producing only extremely small effects in single atoms but producing decisive effects in large bodies that are organized like living cells and organisms. Under the impression of the difficulty of explaining the phenomena of life by physics and chemistry some of the leading physicists of our time, like Bohr, Heisenberg, Pauli, have explicitly rejected this explanation. Yet I do not want to follow them, again in order to simplify the argument in a still undecided case; for the time being I shall cut out their very interesting views from my trend of thought by means of Occam’s razor.
Thus I seem to assume that any living being, including man, is nothing but a piece of physico-chemical machinery. In fact I want to withstand the impact of this view not by escaping into any more or less metaphysical hypothesis but by sheer analysis of its meaning.
Give me credit for accepting the view. If you do so, I conclude: We, living conscious human beings, assembled in this lecture hall in Glasgow University, have come to the conclusion that we are nothing but physical engines. What is it I have said in saying this? Let me take the statement word by word.
“Nothing but” is a beautiful phrase. I am speaking to you. What is speech? Well, nothing but sound waves. But these sound waves carry a meaning—at least I hope they do. What do we mean by “meaning”? This is easily understood and difficult to define. It is not for the sound waves you have come here but for their meaning. But does this imply that the laws of acoustics would have to break down somewhere in order that the sound should be able to carry a meaning? I should say, on the contrary.
Go on and ask what it is in the sound waves that carries the meaning. It is certainly somehow inherent in their particular structure. Theories of information try to measure their content of information in an objective manner. I shall not go into its details; in any case it must be an audible feature of the sound waves, some element of form in them, that carries the meaning.
But probably you will say that the meaning is present only in conscious minds. There must be somebody to listen; I would not preach to the bare walls. How, then, does the meaning-bearing structure of the sound waves enter into your minds? We can follow this process by three or four steps. The sound waves make the tympanum vibrate. This mechanical vibration excites the nerve of the ear to carry a chemico-electrical signal towards the brain. There the signal is somehow brought into contact with many other signals in a way not at all well known to science. Finally your body shows some reaction: concentrated looks at the lecturer, writing down of notes, perhaps sleeping. Where is the mind in this description?
I described what I could see or infer from my scientific knowledge. But to every person in this room there is one person who experiences the process quite differently: this is the person himself or herself. And I should add, others do so, too, in fact. You do not just look at my physical behaviour; you would not be here for this purpose. Listening to my words you know a bit about what is in my mind. Looking at you I know a bit—a bit less probably—of what is in your minds. Mind is known to mind, however this may happen.
I gave you a description; I shall not transform it into a metaphysical hypothesis. The problem of matter and mind, like many other things I have mentioned so far, belongs to the second lecture series. Here I just stay with my original question, asking whether this problem would seem to become in any way more understandable by the assumption that the laws of physics did not hold strictly in our brains? I see no reason why this should be so. Thus to say that we are nothing but physical engines still seems to be a sentence devoid of a clear meaning; but that under the aspects of physics we can be described as physical engines is a statement that might make sense.
Then we take the next step in our analysis. What do we mean by “physical engine”. Take a motor-car as an example. Are we such things as motor-cars?
One difference is evident. Machines are built by man according to a plan in order to fulfil a limited purpose. Man is not built by man. Life—as Bohr sometimes says—is not an experiment of ours. Man has a history of his own. Machines belong to the history of man.
But how relevant is this difference? Does it mean that the really important traits of living beings will never be imitated by man-built machines? We here enter the field of research called cybernetics.
A motor-car is not to be compared to a living being but to one of its organs, say to its feet. Human beings are born of human beings, cars are not born of cars; but neither are feet born of feet. The car imitates one particular activity of animals: locomotion. Radar imitates another one: vision. Calculating machines probably offer the most relevant comparison; they imitate certain activities of the mind. Man is not in practice interested in a machine that imitates all of his actions at once, but in machines which do better than man himself in a very limited field.
Thus, take the most relevant example, the imitation of the mind by electronic calculators. As I said in the first lecture, their degree of complexity so far is comparable to the nervous system of an earth worm; hence we will not expect them to imitate man very well. Still they have beaten him definitely in numerical calculations and I have studied myself the first game of draughts played by an electronic calculator. The machine was still inferior to its inventor in this game. But probably machines will improve faster than the human brain. Where are the limits?
A machine cannot make another machine, you will say. This is not so certain. Certainly a very simple machine cannot build another one. At a higher level of complexity this may change. J. von Neumann has outlined a machine that would be able to compile one exactly equal to itself out of some simple “bricks”. One may think of a machine which would have to be even more complicated and which would be able to build another one superior to itself. Where are the limits?
A machine does not act spontaneously, you will say. This is not so certain. I am told by a man who knows the facts that when the draught-playing machine was about to lose its first game, it cheated. Whenever I tell this story to laymen in the presence of calculator-specialists, the specialist is eager to explain: “but that, of course, was only a mistake”. Of course it was a mistake, it was so by definition. Whenever the machine does what we have not planned it to do we call it a mistake. But even if we rule out such mistakes which are contrary to the programme, we have to admit that we do not know in advance what will come out when the machine obeys the programme; else we would not have cared to build the machine. For the simple programmes of existing machines at least some general expectations about the results can be given. But in any case a machine is a real thing and what it can do once it is built need not be identical with what we think it can do. Where are the limits?
A machine does not have consciousness, you say. How do you know that? I admit the statement, but I admit it since comparison with animals and with the human brain makes it appear improbable that the precise activities of existing machines have any close similarity to activities of nervous systems which are connected with what we call consciousness. But as long as you have not solved the problem of the mutual relationship of matter and consciousness, how can you know that machines cannot have consciousness?
It is an ancient human dream to make a human being. I do not see that our present knowledge proves that this is impossible. Probably, if we could build a man, it would be a horrible thing really to do it. It might be the final sacrilege and its consequences might be disastrous. Perhaps we are rightly afraid of it, and perhaps our fear takes the form of the belief that it is impossible. I think many of our beliefs are disguised fears. In fact, again, I think it will not be possible. But the reason might just be that what you need to make a man is history; perhaps it cannot be done in less than four thousand million years.
You should rightly understand this repeated question: where are the limits? I want to say how little we know about these questions, and how impossible it is even to make negative statements. In order to understand the problem better we would first have to go on with our analysis of meaning. I said that under the aspects of physics we might be likened to physical engines. What are the aspects of physics? Do we know what we mean by the laws of physics that are supposed to hold in our body? As to this question, however, I once again refer to the second series of lectures. It will be their guiding question.