by Karl Popper.
I.
The theory of knowledge is at the very heart of philosophy; not only of the philosophy of science, but also of ethics, political philosophy, and even the philosophy of art. This, no doubt, is at least partly due to the fact that whenever we try to justify any of our assertions, we begin to wonder whether our justifications are valid. In this way we are led to think about the justification of our assertions, our beliefs, our theories, and our hypotheses. And the resulting theories of justification form what is called the theory of knowledge (or epistemology).
Broadly speaking, there have been since antiquity two main schools in the theory of knowledge: one is the school of the pessimists, the sceptics (or the agnostics) who deny the possibility of justification, and with it, of any established knowledge; the other is the school of those who claim to know, to be able to attain knowledge. I shall call this latter school simply the school of the optimists.
It seems fairly obvious that the sceptics must be mistaken in denying the possibility of knowledge; for they seem to be refuted by the staggering success of science.
This argument against the sceptics was first used 200 years ago by Immanuel Kant who referred to the success of Newton's theory; and I think it is decisive. It could not have been used before Newton because Newton's was the first really impressive success in science - a success capable of shaking even a sceptic
During the last 200 years, Kant's argument has become even more compelling; and I think the fact that we are forced to make use of Kant's argument from the existence of scientific knowledge goes far towards transforming the theory of knowledge into the theory of scientific knowledge.
By saying we are forced to make use of Kant's argument I mean that but for Kant's argument which shows that the sceptics or pessimists must be wrong the sceptics, clearly, would have won the debate on all counts.
In our own age it is difficult to umpire the age-old debate between the optimists and the pessimists in the theory of knowledge, simply because today - after Newton, after Maxwell, after Einstein - we know that the pessimists must be in the wrong. Yet those of us who are able to discount this knowledge must, I think, admit that the arguments of the sceptics were always superior to those of the various schools of optimists. Moreover, while the sceptics stood their ground well, the optimists developed a vast number of tricky arguments and theories of knowledge both unconvincing in themselves and mutually incompatible. Some were naive, some were extremely subtle and ingenious, but not one of them, I fear, was in the least convincing to anybody except - perhaps - its author.
The picture I have drawn of the situation is of course oversimplified, but it explains why at the end of the last century a new philosophy of since arose in the United States which tried to break the deadlock between pessimists and optimists. I mean of course the philosophy of pragmatism.
Seen in this particular context, the position of the pragmatists may be stated as follows: scientific theories are extremely successful instruments - instruments, both for prediction and for technology. Yet they're nothing but instruments.
It is fairly clear that to instruments, the problem of justification simply does not apply. You simply cannot produce arguments to justify a shovel, or a bicycle, or a fountain pen - although there are of course better and worse shovels, or bicycles, or fountain pens: their goodness or badness is no a thing which can be justified in any argument; it is simply a question of success - of practical success.
Now one might say this instrumentalist philosophy tried to adopt a third position differing from those of both sceptics and optimists, yet giving both their due.
To the optimists it conceded that Kant's argument is valid - there is science, and its success counts.
To the pessimists it is conceded that there is no pure or theoretical knowledge, and that purely theoretical justification does not exist. For it is conceded that for genuinely theoretical knowledge - knowledge whose value does not merely lie in its practical or instrumental success, but knowledge whose value lies in satisfying a purely intellectual curiosity (if such a thing exists) - some theoretical or argumentative justification would be essential - essential for this kind of pure knowledge.
Thus the pragmatist agrees with the sceptic on the impossibility of pure knowledge. Yet he can also accept the Kantian argument from the existence of science because this argument relies on the success of science, which is never more impressive and indubitable than when it becomes practical, technological, and instrumental.
A slightly different way of looking at pragmatism would be this: pragmatism might be said, is a barely disguised form of scepticism. For the original quarrel between the two parties, the sceptics and the optimists, was over pure knowledge or theoretical knowledge. Nobody ever quarreled about the fact that instruments may be highly successful or unsuccessful. Thus pragmatism is merely a kind of reformed skepticism - reformed to allow for the rise of modern science, first used by Kant as a stick to beat the sceptics; we may thus view pragmatism as a form of scepticism; as a philosophy which admits that there is science, but denies that science is knowledge.
If we look at pragmatism in this way, then we find that it is quite an ancient doctrine. It was used by Bishop Berkeley in order to show that the new secular science - which was then, in the main, Newton's theory - was not knowledge, but simply an instrument - an instrument for calculation and prediction. And it was earlier used by Cardinal Bellarmine against Galileo, for precisely the same purpose. In other words, it was originally hostile to the rise of science: it was an attempt, in fact, to belittle science. Its tendency was to show that science is merely a kind of glorified gadget-making, or glorified pumping; that it cannot have any message for the human spirit, though it may help to provide comfort for the human body.
Now I happen to believe that this is all wrong. I admire Newton for his intellectual achievement - for giving us hope, for the first time in the history of humankind, that we may really grow to understand something about the world we live in. And I admire Einstein for similar reasons. Einstein himself, incidentally, was quite unimpressed by the technology which he helped create; he even said of the bomb that as an intellectual achievement, it was a trifle.
Those who like myself wish to dissent from pragmatism or instrumentalism admit, as a matter of course, that most scientific theories are instruments. What we deny is that they are nothing but instruments - more precisely, that they cannot have value as pure theoretical knowledge. And we shall assert the old-fashioned thesis that the pure scientists searches for truth, and not merely for usefulness.
The great pragmatist William James more or less foresaw this last remark, for he tried to counter it by proposing that truth (and especially the truth of a scientific theory) is usefulness, its practical success. But not even all pragmatists accepted the Jamesian equation 'truth equal usefulness'; and I think that the I think that the theory of truth developed by Alfred Tarski has dispelled many sceptical misgivings about the idea of truth. What Tarski did was to show that there was no reason to give up the commonsense view according to which a statement or a theory is true if it corresponds to the facts. In addition he gave an analysis of what may be meant if one says of a statement that it corresponds to the facts.
Tarski's theory allows me to say, without fear of using meaningless words, that pure science aims, or should aim, at the truth, rather than at power, or at technological success.
But if I thus dissent from pragmatism and instrumentalism, I have a serious problem to solve. For if we reject the pragmatist solution to the problem of knowledge, how are we to avoid slipping back into the old controversy between the sceptics and the optimists?
I agree with the pragmatists that we have to seek a third position between optimism and pessimism. But it is clear that our new 'third position' will have to be, in contradistinction to the 'third position' of the pragmatists, much further from pessimism than from optimism.
For I shall now have to argue that purely theoretical knowledge, knowledge for which aims at truth, exists. I shall even have to argue that this purely theoretical knowledge is progressive. In addition, I shall even argue that it is rational knowledge; a claim shows how different my position is from that of the sceptics and the pragmatists.
On the other hand, it is very different from that of the optimists. For although I do hold that we are aiming at true theories, I do not assert that we have been, or that we shall be, successful. It is to say the least, very doubtful whether Newton was successful in finding a true theory, though it is arguable that he came nearer to the truth than Kepler. Einstein did not even claim that general relativity was true; but he did claim that it was an improvement on Newton's theory in being a better approximation to the truth.
I shall now try to formulate the position I am taking up towards scepticism and optimism.
This, in brief, is the position with regard to the theory. of knowledge which I adopt. It involves believing that knowledge - scientific knowledge - is unstable, growing, critical, and that it is always tentative, hypothetical. It searches for the truth, and tries to get nearer to the truth, without ever claiming, or trying to justify the claim, that it has reached it.
II.
I now come to an entirely different problem. Does this non-justificationist theory of knowledge affect the analysis of scientific method? And if so, what difference does it make?
It makes a very great difference. To show this I shall first briefly describe the main attitude toward scientific method taken up by the optimists, the sceptics, and the pragmatists.
The optimists - or at least their empiricist wing - believed that the method of science consisted in collecting observational and experimental evidence, and in proceeding from this to theories by way of generalization or induction by repetition.
The sceptics asserted that generalization or induction was invalid: according to Hume, no number of observations of white swans could prove, or make it probable, that all swans are white.
But Hume believed that animals and men tend to form their habits by repetition, that is to say, by induction. Since induction was invalid this was, he emphasized, a completely irrational procedure. But he pointed out that this irrational procedure of habit formation by induction seemed to work better than any rational procedure. With this argument he became the grandfather of irrationalism.
The modern pragmatists - or at least some of them - try to have it both ways.
They agree with Hume that induction is invalid, but they say that the word 'invalid' here means only 'not valid in the sense in which the inferences of deductive logic are valid': and this, they say, is trite.
They also agree with Hume in believing that induction - or as they prefer to call it, 'inductive procedures' - are the best and most successful methods we have. And they now argue as follows: if inductive procedures are successful, then this means that they may be called 'valid' in some sense of the word. Admittedly, not in the sense in which deduction is called 'valid'. But there is no other criterion for validity - they say - than success.
You may remember that William James suggested that we should identify truth - the truth of a theory - with success or usefulness. His modern descendants identify validity - the validity of an argument - with usefulness or success. The similarity is striking.
But I should really have said that the modern descendants of James identify validity with alleged success or alleged usefulness. For both the use of 'inductive procedures' and their success in science is I believe, a sheer myth. Scientists, non-scientists, and even animals, proceed otherwise.
The most important criticism of this view of induction is that the alleged analogy between deductive and inductive validity does not hold: it assumes the mistaken theory that deductive validity is either based on convention or on its practical success or on some irreducible intuition. But in fact deductive inference is formally valid if, and only if, no counter example exists. By 'counter example' is here meant an inference of the same logical form as the inference in question, but with true premises and a false conclusion. The non-existence of a counter example ensures therefore the transmission of truth from the premises to the conclusion (provided the premises are all true) and the retransmission of falsity from the conclusion to at least one of the premises (provided the conclusion is false).
This is the sole reason for the theoretical and practical significance of a valid deductive inference: it does not lie in a (more or less) arbitrary convention to regard certain inferences as valid, or in the practical success of the deductive inference; nor in its intuitive persuasiveness. As a consequence, the alleged analogy between the validity of deduction of inference and the supposed validity of inductive inference cannot be upheld.
III
The position between optimism and pessimism which I am trying to establish may be briefly described as follows.
I agree with the pessimists that there is no justification for the claim of any particular theory or assertion to be true. Thus there is no justification of any claim to know, including the claims of scientific knowledge. But this merely means that all knowledge, including scientific knowledge, is hypothetical or conjectural: it is uncertain, fallible. This certainly does not mean that every assertion is as good as any other, competing, assertion. For we can discuss our various competing assertions, our conjectures, critically; and the result of the critical discussion is that we find out why some among. the competing conjectures are better than others.
Accordingly, I agree with optimists that our knowledge can grow, and can progress; for we can sometimes justify the verdict of our critical discussions when it ranks certain conjectures higher than others.
A verdict of this kind aways appraises our conjectures or theories from the point of view of their approach to truth: although we cannot justify any claim that a theory is true, we can sometimes give good reasons for asserting that one theory is better than another, or even than all its competitors. In this way our knowledge can grow, and science can progress.
Popper, K., Turner, P. and Shearmur, J., 2008. After The Open Society. New York: Taylor and Francis, pp. 3-10.