File Name: karl popper conjectures and refutations .zip
To browse Academia. Skip to main content.
Karl Popper: Conjectures and Refutations
The problem of induction was highlighted by David Hume , who pointed out that particular experiences tell us only about themselves: they give us no information about other experiences.
So the fact that every thing of type T1 that we have experienced has also been of type T2 does not imply that things of type T1 that we have not experienced will also be of type T2. The fact does not even imply that it is probable that things of type T1 that we have not experienced will be of type T2, because we can have no idea whether our experience of things so far has been representative of how things are in general. Consequently, we can have no justification for thinking that our general theories are either true or even probably true.
Of course, we may find our general theories plausible, or 'subjectively probable,' given our particular experiences or inclinations; but we can have no grounds for linking plausibility or subjective probability to truth or to objective probability. Induction and DemarcationPopper criticised and rejected empiricist and rationalist attempts to solve the problem of induction.
Empiricist attempts involve a vicious infinite regress if they invoke a principle of the uniformity of nature, because justification of such a principle will again assume that the future will resemble the past. Attempts to show that successful scientific theories have a high objective probability fail for the additional reason that they would not even solve the problem of induction, since highly probable things may fail to be the case, so a theory with very high objective probability might still be refuted the next time we test it.
Any proposed solution is irrelevant if it divorces justification from truth. Rationalist attempts to solve the problem by maintaining that some of our scientific theories can be proven a priori are either dogmatic or circular or involve a vicious infinite regress. For the question arises concerning how the premises of the proof can be known.
If they must in turn be proven, we set off on an infinite regress unless the premises come from the theory to be proved, in which case we argue in a circle. If the premises are accepted without proof they are simply dogmas. Empiricists and rationalists offered positive solutions to the problem of induction by attempting to show that scientific theories can be verified or supported either empirically or a priori.
But Popper's solution is negative. It does not matter that scientific theories cannot be verified or supported. What matters is that we can learn from experience by testing our theories against experience and then rejecting or revising those that are refuted. This solution to the problem of induction also provides a solution to the problem of demarcation.
What distinguishes scientific theories from metaphysical ones is that they may clash with experience: the mark of a scientific theory is its empirical refutability or falsifiability.
The problem of demarcation was highlighted by Immanuel Kant Kant , who tried to identify how empirical science differed from speculative metaphysics. Empiricists affirm that scientific theories are distinguished by being derived from or verified by experience; but that claim founders on the problem of induction.
Kant offered a rationalist solution according to which the general principles of the empirical sciences can be proved a priori; but Kant's 'proofs' are not only invalid or unsound, they are also ultimately dogmatic. Falsifiable TheoriesWhen we say that a theory clashes with experience we mean that the theory is inconsistent with a statement reporting an observation.
Popper defines a 'basic statement' as one that ascribes an observational property to a thing in a specific region of space and time. An observational property is one that we could conceivably observe to be instantiated, with our actual powers of observation.
So, the following are basic statements: Statement 3 is, we assume, false; but it is a basic statement because it is conceivable that there are pink elephants and, if there were one in Trafalgar Square on 4 March , we could have observed it. In contrast, an accepted basic statement is one that we agree describes something that we observe. Statements 1 and 2 may be accepted basic statements. A theory is falsifiable if and only if it is inconsistent with a basic statement.
A theory is falsified or empirically refuted if and only if it is inconsistent with an accepted basic statement that describes a reproducible situation. A reproducible situation is one that either occurs regularly in nature or that we can bring about regularly by means of experiments. So, 'there are no pink elephants' is falsifiable but not falsified; and 'all swans are white' is falsified because there are numerous basic statements of the form 'that is a black swan' that have been agreed in different contexts.
The observational terms in basic statements are theory-laden. For example, 'this is a tree' uttered in a particular context is a basic statement to which observers may agree. But if the thing thus described suddenly shed its leaves and retracted its branches, or if it waddled off, or if it screamed when carvings were made on its bark, the previously accepted basic statement, 'this is a tree,' would be normally be rejected because the accepted basic statements describing the unexpected behaviour are inconsistent with our understanding of what a tree is.
Consequently, an inconsistency between a general theory and an accepted basic statement does not show that the theory is false, since it might be that the basic statement is false. A falsified theory might not be a false theory. Popper often equates falsified with false in his more popular expositions; but that is not his considered view. Better ExplanationsIn our quest for knowledge we attempt to get solutions to theoretical or practical problems.
Our aim, says Popper, is explanations. But when we are seeking explanations we already have some explanations which are provided by our inherited theories, the latter being supplied in part biologically and in part culturally with consequent differences between cultures.
Indeed, the problems we are trying to resolve derive in some way or another from the inadequacy of the explanations we currently have. So our aim is more justly described as being to get better explanations than those that we currently have. If we do achieve better explanations, then they become the ones we currently have, and they can be expected to generate further problems in their turn; so we then look for new explanations that are better still.
Popper suggests a wide range of properties in terms of which we can grade explanations as better or worse, including the following. One explanation is better than another, other things being equal, if: i it is falsifiable while the other is not; ii it explains what the other explains and some other things besides; iii it provides a unified solution to problems that were explained in different ways by the other; iv it generates surprising falsifiable predictions that are not derivable from the other and those predictions survive testing; v it corrects the other, that is, it shows that some falsifiable predictions of the other that were thought to be successful were actually not successful and it explains why they seemed successful.
In the seventeenth century, successful scientific theories of motion had been propounded by Galileo Galilei , with regard to terrestrial bodies, and by Johannes Kepler , regarding celestial bodies. Galileo's 'law of freefall' stated that freely falling bodies have constant acceleration. Kepler's 'first law' said that the planets have elliptical orbits around the sun.
Isaac Newton then provided a theory that explained both terrestrial and celestial motions, including things unexplained by the conjunction of Galileo's and Kepler's theories, such as the motions of the moon and of the tides.
It also had many surprising predictions that survived testing, including some that contradicted the previously successful predictions of Galileo's and Kepler's theories. In particular, it predicted that acceleration of freely falling bodies is not constant, while explaining why, for bodies relatively close to the earth, the assumption of constant acceleration gives a close approximation; and it predicted that the orbits of the planets are only approximately elliptical because of the effects of interplanetary gravitation.
Thus, Newton's theory had properties ii , iii , iv and v in comparison with the conjunction of Galileo's and Kepler's theories; and, like that conjunction, it was falsifiable.
It was a better explanation. Popper emphasises that one theory may be better than another with regard to some of the properties of a better explanation but worse with regard to some others. As a consequence there are times when it is not possible to place rival theories in a hierarchical order; and even when we can do that, it is a matter of critical assessment rather than of algorithmic computation.
The Duhem ProblemsPopper says that to try to obtain better explanations we should propound conjectures and then criticise them, using empirical criticism where possible, discarding poorer explanations in favour of better ones.
There are, however, two problems which had been identified by Pierre Duhem The first was mentioned above: accepted basic statements may be false, so a refutation poses the question of whether we should discard the explanatory theory or the theory-laden basic statement. The second problem is that, in the developed sciences, major theories are abstract and highly general. They are, in fact, not directly falsifiable: no basic statement is inconsistent with them. They are falsifiable only indirectly, that is, in conjunction with some other accepted theories.
Willard Quine Quine labelled this problem 'holism' but it is in fact always a limited, though sometimes quite large, number of theories that are used in deducing a falsifiable prediction from a theory, so 'largism' coined by John Worrall is a more appropriate label.
Since a refutation often involves an inconsistent conjunction of a basic statement and several theories, then, even if we decide not to reject the basic statement, the refutation poses the question of which of the several theories to discard. Popper says that, when we obtain a refutation, we may reject any proposition in the inconsistent conjunction provided that we replace it with a proposition the conjunction of which with the remaining propositions in the set produces a better theory than we had before, particularly in ways ii - v , in section 4 above.
In addition, if we reject a previously accepted basic statement, the new conjunction must explain why the basic statement was previously accepted. The Problem of Theory-Laden Basic StatementsIn the late-seventeenth century, basic statements accepted by the Astronomer Royal contained statements of the positions of the moon that were inconsistent with the predictions derived from Newton's theory.
Newton rejected those basic statements. Basic statements about the position of the moon inevitably assume a theory of atmospheric refraction: either the atmosphere refracts light or it does not; and if it does, there is an infinity of ways in which it might do so. Newton replaced the theory of atmospheric refraction presupposed by the Astronomer Royal with an alternative hypothesis about refraction, which required each of the previously accepted basic statements about the moon's positions to be replaced with new basic statements which were consistent with Newton's theory.
Newton thereby explained why the moon was where his theory said it was and also why it appeared to be where the Astronomer Royal thought he saw it. Newton's new hypothesis of refraction applied not only to light reflected from the moon but to light in general that travels through the earth's atmosphere; so it made surprising falsifiable predictions concerning the observed positions of other objects; and those predictions survived testing.
The Problem of LargismIn the mid-nineteenth century, accepted basic statements of the positions of Uranus were inconsistent with some of the predictions entailed by the conjunction of Newton's theory with accepted theories about the solar system.
Rather than rejecting Newton's theory, Urbain Le Verrier chose to revise the accepted theory that there are seven planets. He proposed a new hypothesis that there is an eighth planet with a size and an orbit that would generate a gravitational pull on Uranus sufficient, according to Newton's theory, to account for the accepted basic statements about Uranus's positions.
The new hypothesis, in conjunction with Newton's theory, implied surprising predictions concerning when and where, on a clear night, the hypothesised planet would be seen, if a telescope were pointed in the right direction. Those predictions were falsifiable and they survived testing: when telescopes were pointed as directed, the new planet was discovered; it was named 'Neptune.
Newton and Le Verrier each rescued Newton's theory from falsification; but each did so in ways that gave us better explanations that yielded surprising falsifiable predictions that survived testing. Each thereby contributed to the growth of scientific knowledge. But there are other ways of saving a theory from falsification that are pseudo-scientific. For example, we may state a theory vaguely so that it is consistent with all basic statements and is thus unfalsifiable.
Popper says both that Marxism is falsified and thus falsifiable and that it is unfalsifiable. Such seemingly self-contradictory statements result from his not distinguishing clearly between two criteria of demarcation.
Falsifiability is what demarcates scientific from metaphysical statements. But what distinguishes scientific from pseudo-scientific procedures is that the former, but not the latter, encourage the development of theories that are better explanations. The pseudo-scientific stratagems of Marxists enable them to retain their theory either in the face of falsifications, which are left unexplained, or by explaining away falsifications in a way that explains nothing new.
Either way they fail to learn from their mistakes. Scientific progress depends upon the abandonment of pseudoscientific procedures.
Faster previews. Personalized experience. Get started with a FREE account. Load more similar PDF files. PDF Drive investigated dozens of problems and listed the biggest global issues facing the world today. Let's Change The World Together. Pdfdrive:hope Give books away.
Please choose whether or not you want other users to be able to see on your profile that this library is a favorite of yours. Finding libraries that hold this item Professor Popper puts forward his views with a refreshing self-confidence. In summarizing his views in this way, I have done scant justice to the subtlety and importance of his argument. His own presentation of his case is luminously clear. You may have already requested this item.
Karl Popper is generally regarded as one of the greatest philosophers of science of the 20th century. Karl Raimund Popper was born on 28 July in Vienna, which at that time could make some claim to be the cultural epicentre of the western world. His father was a lawyer by profession, but he also took a keen interest in the classics and in philosophy, and communicated to his son an interest in social and political issues which he was to never lose. His mother inculcated in him such a passion for music that for a time he seriously contemplated taking it up as a career, and indeed he initially chose the history of music as a second subject for his Ph.
This volume is a result of that effort. The concept of falsification, the problem of demarcation, the ban on induction, or the role of the empirical basis, along with the provocative parallels between historicism, holism and totalitarianism, have always caused controversies. The aim of this volume is not to smooth them but show them as a challenge.