ON THEORIES AND OBSERVATIONS
An observation is always preceded by a particular interest, a question, or a problem - in short, by something theoretical. After all, we can put every question in the form of a hypothesis or conjecture to which we add: 'Is this so? Yes or no?' Thus we can assert that every observation is preceded by a problem, a hypothesis (or whatever we may call it); at any rate by something that interests us, by something theoretical or speculative.
Darwin knew this when he wrote: 'How odd it is that anyone should not see that all observation must be for or against some view'.
Any attempt to provide explanations presupposes a theory. The difference between so-called theory-neutral and theoretically based explanations is not really one between the presence and absence of an appeal to theory, but a difference in the sophistication and depth of the two theories involved.
EMPIRICAL THEORIES MAKE FALSIFIABLE PREDICTIONS
The value of any scientific hypothesis lies in co-ordinating known facts and in suggesting new inquiries likely to advance our knowledge of the subject under investigation.
Every scientist who claims that his theory is supported by experiment or observation should be prepared to ask himself the following question: Can I describe any possible results of observation or experiment which, if actually reached, would refute my theory?
If not, then my theory is clearly not an empirical theory. For if all conceivable observations agree with my theory, then I cannot be entitled to claim of any particular observation that it gives empirical support to my theory.
Linguistics likes to think of itself as a science in the sense that it makes testable, i.e. potentially falsifiable, statements or predictions.
A THEORY MAY ONLY BE DISCARDED IN FAVOUR OF A BETTER THEORY
All theories are guesses or conjectures, and all that we can rationally justify is a tentative preference for one or two of the competing theories. But there is all the world of difference between the justification of a preference - for the time being - for one of the competing theories, and the justification of a theory. To justify a theory is to show that it is true. But we may justify a preference, even for a false theory, if we can show that of all the competing theories it appears to come nearer to the truth than any of the others.
A theory is not refutable in the way that a law is, but only replaceable by a superior theory, a theory which may indeed maintain some if not all the laws of the earlier theory.
USING AN OLD THEORY AS THE BASIS FOR A NEW SUPERIOR THEORY IS NOT CHEATING
The progress of science consists in trials, in the elimination of errors and in further trials guided by the experience acquired in the course of previous trials and errors. No particular theory may ever be regarded as absolutely certain: every theory may become problematic, no matter how well corroborated it may seem now. No scientific theory is sacrosanct or beyond criticism. (...) We have now come to see that it is the task of the scientist to subject his theory to ever new tests and that no theory must be pronounced final. Testing proceeds by taking the theory to be tested and combining it with all possible kinds of initial conditions as well as with other theories, and then comparing the resulting predictions with reality. If this leads to disappointed expectations, to refutations, then we have to rebuild our theory.
The result is generally viewed as cheating by non-theoreticians: if your theory gives you the wrong result, you change (one tiny sub-part of) your theory. This of course is the normal means of scientific progress.
WHAT IS AN EXPLANATION AND WHAT IS AD HOC ?
In order that the explicans should not be ad hoc, it must be rich in content: it must have a variety of testable consequences, and among them, especially, testable consequences which are different from the explicandum. (...)
Only if we require that explanations shall make use of universal statements or laws of nature (supplemented by initial conditions) can we make progress towards realizing the idea of independent, or non-ad hoc, explanations. For universal laws of nature may be statements with a rich content, so that they may be independently tested everywhere, and at all times. Thus if they are used as explanations, they may not be ad hoc because they may allow us to interpret the explicandum as an instance of a reproducible effect. All this is only true, however, if we confine ourselves to universal laws which are testable, that is to say, falsifiable.
The question 'What kind of explanation may be satisfactory?' thus leads to the reply: an explanation in terms of testable and falsifiable universal laws and initial conditions. And an explanation of this kind will be the more satisfactory the more highly testable these laws are and the better they have been tested. (This applies also to the initial conditions.)
The growth of knowledge depends entirely upon disagreement.
It is not the objectivity or detachment of the individual scientist but of science itself (what may be called 'the friendly-hostile cooperation of scientists' - that is their readiness for mutual criticism) which makes for objectivity.
There is even something like a methodological justification for individual scientists to be dogmatic and biased. Since the method of science is that of critical discussion, it is of great importance that the theories discussed should be tenaciously defended. For only in this way can we learn their real power. And only if criticism meets resistance can we learn the full force of a critical argument.
Cajori, Florian: 1928, A History of Mathematical Notations, Open Court Publications, Chicago.
Popper, Karl: 1972, Objective Knowledge, Oxford University Press, Oxford, revised edition 1979.
Popper, Karl: 1994a, Knowledge and the Mind-Body Problem, Routledge, London.
Popper, Karl: 1994b, The Myth of the Framework, Routledge, London.
Smith, Neil: 1989, The Twitter Machine, Blackwell, Oxford