Will Bouwman's Philosophy Now article, "Thomas Kuhn (1922-1996)" is a wonderful illustration of what is wrong with philosophers attempting to identify what science is and how it ought to be done. For example:
"In the middle of the twentieth century the philosophy of science was almost exclusively focussed on defining the scientific method. The assumption was that science is an objective ideal method independent of human foibles, and if we could just describe its characteristics then everyone would have a template for doing proper science."
Just as politicians, who have never produced a product or performed a medical procedure, believe they should determine how business is done and what good medical practices are, philosophers, who have never made a single scientific discover or produced a technological improvement are certain they know how those things should be done. While scientists are actually accomplishing scientific discoveries and engineers are using those scientific principles to produce technological wonders (which happen to be the proof of the science--see list) philosophers are twiddling their mental thumbs inventing some kind of abstruse template of how science is to be done.
Here's a list of just a few of science and technology's twentieth century accomplishments without a philosopher's template:
1900 Quantum theory proposed / Planck
1901 Discovery of human blood groups / Landsteiner
1905 Wave-particle duality of light / Einstein
1905 Special theory of relativity / Einstein
1906 Existence of vitamins / Hopkins
1911 Discovery of the atomic nucleus / Rutherford
1911 Superconductivity discovered / Onnes
1912 Discovery of cosmic rays / Hess
1915 General theory of relativity / Einstein
1921 Isolation of insulin / Banting & Best
1923 Nature of galaxies discovered / Hubble
1928 Discovery of penicillin / Fleming
1930s Theory of chemical bonds developed / Pauling
1932 Discovery of the neutron / Chadwick
1932 Discovery of the positron, first antimatter particle / Anderson
1939 Discovery of nuclear fission / Meitner & Frisch
1944 Evidence in bacteria that DNA / Avery, MacLeod, and McCarty
1946 Initial elucidation of the reactions in photosynthesis / Calvin
1947 Invention of the transistor / Shockley, Bardeen, and Brattain
1956 Discovery of the neutrino / Cowan & Reines
If philosophers truly want to know how science ought to be done, perhaps they should examine how those who are actually successfully doing science do it. The one thing they would discover is that there is not one basic template, one plan, one, "paradigm," for performing scientific research. The "right method" is determined by the physical phenomena, entities, events, qualities, or relationships being studied.
After a brief description of logical positivists versus Karl Popper, Bouwman writes:
"One major problem—which in fairness the logical positivists were well aware of—is that no amount of empirical evidence (or logic) can prove a scientific claim. The classic example is that a million white swans do not prove that every swan is white. Popper's innovation was to point out that it only takes one black swan to prove that the proposition 'all swans are white' is false. So the evidence could show you either what was only likely to be true, or what was definitely false. Therefore, as an endeavour seeking certainty, science should commit itself to trying to prove its own theories wrong. This is Popper's principle of falsification."
Whenever I read something like, "no amount of empirical evidence (or logic) can prove a scientific claim," I have to wonder if the writer knows anything at all about science. Does he believe the circulation of the blood is still in doubt, that anesthesia might not work, that wireless communication is not certainly possible, that geostationary communication satellites are only a hypothesis, or that science is still waiting for more evidence that lased light is possible?
I'm not sure what prompts philosophers to embrace extreme skepticism about science, unless it is their experience with their own discipline, about which I am extremely skeptical myself. I do know all the arguments for such doubt are spurious.
Popper's absurd argument, "... that it only takes one black swan to prove that the proposition 'all swans are white' is false," is wrong ontologically and epistemologically. Ontologically there are only individual entities. There is no ontological "swanness." There are no ontological essences.
"Swan," is only a word which represents a concept. A concept is an identification of an existent (if it is a particular concept) or a class or category of existents (if it is a universal concept). What existents a concept identifies is specified by the concept's definition. The definition of the concept, swan, might be, "large white water birds with graceful necks."
If that is the definition, the proposition, 'all swans are white,' is true. If another bird is discovered with all the same attributes as swans, except for color, it is not a black swan, because swans by definition are white. The proposition, 'all swans are white,' is still true.
If the black birds are similar to swans in all other ways, one can either change the definition of swans to include black ones, or create an new concept for the black birds.
There is no epistemological or ontological principle that dictates how words must be defined. Epistemologically, the definition that best identifies existents and their relationships will be the most useful, but there is no authority determining that.
There is another wrong assumption implied by Popper's mistake which is that science proceeds by means of "induction." It does not. Nothing is established on the basis of many similar observations are made. A frequent observation might be the basis for further research, but scientific principles are established by the identification of the nature of existents, their attributes (qualities), behavior, and relationships. It would required the observation of only one swan to form the concept that identifies swans, which would be defined in terms of the attributes of that single swan. The concept would be valid if it were the only swan ever observed, or if it were only one of an indefinite number of additional observed swans. They would all be swans because they would have the same attributes, and therefore the same definition.
There is one more even worse issue, "falsification." There is hardly any idea that could be more ignorant or destructive of science than the one that says, "science should commit itself to trying to prove its own theories wrong." If that were true, scientists should be busy proving blood does not circulate, that anesthesia does not work, that wireless communication is impossible, that geostationary satellites have all fallen down and there is no such thing as lasers. The absurdity is obvious.
There is one case where the idea of falsification is useful and that pertains to hypotheses. No hypothesis is a legitimate hypothesis if there is no way to prove it false, if it is false. If there is no way to prove a hypothesis false, if it is, just anything can be put forth as a hypothesis, from miracles to little green men.
What most people do not realize is that falsification is actually a way of proving a hypothesis is true. Since there must be a way to test a hypothesis to determine if it is false, if it is, when the test is performed, if it fails to prove the hypothesis is false, it proves the hypothesis is true, because if it were false, the test would succeed in proving it. Most scientist will understand that. Apparently most philosophers do not.