Karl Popper (1902–1994) is a philosopher and scientist of the contemporary era, essential in the reflection on science.
In The Unfinished Quest, Karl Popper writes about science:
1. The role of the scientist is to criticize
“I came to the conclusion in this way, towards the end of 1919, that the scientific attitude was the critical attitude. It was not looking for verification, but for crucial experiments. These experiments might well disprove the theory under examination; they could never establish it.”
The Unfinished Quest, Karl Popper, 1981
2. Theory of impregnation
Karl Popper presents the theory of impregnation of Konrad Lorenz:
“This theory implies that young animals are endowed with an innate mechanism which enables them to arrive immediately at unshakable conclusions.”
3. The notion of falsifiability
“When a little later I was introduced, on an experimental basis, the notion of falsifiability (or testability or refutability) of a theory as a criterion of demarcation, I quickly realized that any theory can be ‘immunized’ against criticism (…). If we let this immunization occur, then no theory can be falsified anymore. We must therefore proscribe, at least, some immunizations.”
The Unfinished Quest, Karl Popper, 1981
Thus Popper then describes four examples:
(a) Metaphysical theories, which are neither falsifiable nor testable.
b) psychoanalytical theories or the (sufficiently vague) knowledge of astrology, which are also neither falsifiable nor testable.
c) “unsophisticated” theories like “all swans are white”. These theories are falsifiable, although, of course, falsifications can be avoided: immunization is always possible. But protective maneuvers would generally be dishonest, for example, by saying that a black swan is not a swan.
d) the case of Karl Marxism, for which“one can consider that Karl Marx’s theory has been refuted by events”, and“the reinterpretation of Karl Marx’s theory of revolution, intended to escape this falsification (…) transforming it into a vulgar (or socio-analytical) Karl Marxism”
e) more abstract theories like Newton’s or Einstein’s gravitation, which are falsifiable.
4. Natural sciences are not inductive sciences
“I understood why the erroneous theory of science that had dominated since Bacon – that the natural sciences were inductive sciences and that induction was the process of establishing or justifying theories by repeated observations or experiments – was so deeply entrenched.”
The Unfinished Quest, Karl Popper, 1981
5. All unfalsified scientific theories are hypotheses
“Scientific theories, if not falsified, always remain hypotheses or conjectures.”
The Unfinished Quest, Karl Popper, 1981
The book you are referring to is called The Unfinished Quest. THANKS
The Popperian approach has made it possible to elucidate important points concerning the problem of the growth of scientific knowledge. The relationship that Popper established between empiricality (the possibility of a relationship with experience) and refutability is a formal relationship (see notably The Logic of Scientific Discovery). Beyond the problem of empiricality, this implies that we must admit that epistemology is a formal science, and not, as many authors thought, a historical, or sociological, or psychological discipline, etc. (see my personal website: http://www.dblogos.net/er/index.php). This formal character of epistemology itself has the consequence that it admits of no derogation: any theory which claims, not to be true (Popper shows that one cannot prove the truth of a theory, even if one has a true theory), but at least have a relationship with reality, must be refutable by experience. The examples of non-refutable systems of thought quoted here are not exhaustive: one could also in economics mention Milton Friedman, saying that it is a false idea, and which has caused great damage, to want to test the postulates. Another example of non-respect of the principle of falsifiability (or falsifiability) can be found in economics, moreover at the origin of Mandelbrot’s work in economics and finance, in the practice of selecting data (for example stock indices) in order to obtain Gaussian distributions, excluding extreme values (I use this example in the last article on my site). The Popperian approach, which is explicitly in line with certain epistemological texts of Einstein (notably Geometry and experience), has been well integrated for a long time in physics, biology, etc. But it is precisely where this approach is not taken into account that the awareness of its fruitfulness can be the most profitable.