According to Freeman Dyson, there are two types of scientific revolutions. Beside the well-known Kuhnian, concept-driven revolutions (such as the Copernican revolution or Einstein’s new physics), there are tool-driven ones, too. What is more, they dominate our age, although “the concept- driven revolutions are the ones that attract the most attention… but… they are comparatively rare.”
Since the 15th century says Dyson, only a half dozen concept-driven revolutions happened: Copernicus, Newton, Darwin, Maxwell, Freud and Einstein, but about two dozen tool-driven revolutions from Galileo’s telescope to the Watson-Crick DNA model based on X-ray diffraction [Imagined Worlds, p. 50].
It is not an accident that the idea of tool-driven revolution appeared in the last century. It was the age of the “big science” when industrial nations’ mega projects from Hubble Telescope to Large Hadron Collider and to HUGO became to dominate experimental sciences, and it was evident to plan the research programs to fit to the billion dollar equipment, and not vice versa.
Parallel with Dyson, scientific historian Peter Galison published a book about the “visual culture of microphysics” arguing that traditionally the scientific activities are divided into theoretical; instrument building and experiments [Image and logic, 1997, p. 799] and theory was regarded to be the most important part.
Historically, an experiment can served for three purposes. In the Middle Ages its function was to demonstrate the known truths (similarly to the universities’ public debates which aim was to demonstrate debate skills. A student had to defend, another had to attack an Aristotelian statement, but proving that Aristotle was “wrong” showed only the winner’s persuasive argumentation, not the ancient philosopher’s fault [John Henry: A Short History of Scientific Thought 2012, p. 49].)
Today an experiment can be used either for “verify” a theory or as a starting point of a new one and it can lead to a tool-driven revolution. In short, a theory-driven revolution is a new answer for old questions; tool-driven revolution is a new answer for new questions. A philosopher can says that a Kuhnian revolution is a posteriori, while a tool-driven one is a priori.
Obviously there is no a way to sharply divide theory from practice. The classical mechanics of 19th Century relied heavily on models where the forces, effects etc. were visualized by the imagined combinations of pulleys, levers etc. And Galison pointed out that Einstein’s position in the patent office was not isolation, but ideal to review the technical/mechanical innovations – including the problem of the synchronization of electric clocks [Jon Agar: Science in the Twentieth Century and Beyond, 2012, p. 29].
Similar to experiments which have different usage, there are some different methods to use mathematics in natural sciences.
- Newton developed a new kind of mathematics to solve certain physical problems. Opposite to it,
- Einstein choose an existing mathematical solution to answer a known physical question
- A third opportunity is to apply mathematical concepts for natural sciences directly. In mathematics not only potential, but actual infinity exists, so a physicist who simply use mathematics as a tool, probably accept the existence of actual infinity in physics without any reflection (see black holes and other exotic entities), although its existence in reality is not proven.
It is self-evident that mathematics is a “tool” form our approach, and we can generalize this approach arguing that there are even more general “thinking tools” than mathematics. I.e. the anthropic principle as a tool determines our world view.
These mathematics-like thinking tools became more and more important, since there is no opportunity for a real experiment in connection with the multiverses, for example.
“Philosophy has not kept up with modern developments in science,” thus it “is dead” writes Hawking [The Grand Design 2011, p. 13]. But Roger Scruton underlines that philosophy’s aim is not to find real solution for real problems (i.e. the path of an asteroid), but to answer “abstract” and ultimate” questions [Modern Philosophy 1994, p. 3].
My prediction is that the traditional, experiment dominated science will be split up to tool-driven big science and a field dominated by thinking-tools. So we have two choices. If we want to believe that science is competent even there is no a chance for an experiment then we have to redefine the meaning of science. Or, we have to recognize that philosophy annexes some parts of modern natural sciences.
