Is there Creative Anarchy in Science?
In my own experience scientists are actually fascinated and thrive by their own lack of knowledge. But what about their methods? Is there such a thing as the scientific method or do scientists make great discoveries in the absence of logic and method, driven by intuition, anarchy and (mostly) uncommon sense? Is there any value in the philosophy of knowledge and the study of the scientific method?
Paul Feyerabend, an Austrian philosopher born in 1924, would likely suggest that philosophical inquiry of the nature of knowledge and the scientific process has been useless. Initially trained as an opera singer and as physicist, Feyerabend studied philosophy under Karl Popper, became a critic of Popper’s theories and is now recognized as one of the twentieth century’s most iconoclastic philosophers of science. In his book Against Method, Feyerabend rejected the existence of a scientific method and advanced his theory of epistemological anarchy, the idea that there are no useful rules governing the growth of knowledge and scientific discovery. The idea that science should operate according to universal methods is unrealistic and detrimental to science.
Feyerabend objected to any prescriptive scientific method on the grounds that any such method would limit scientific creativity and restrict scientific progress. Science would benefit from a dose of theoretical anarchism. Feyerabend showed that some historic events in science, such as the Copernican revolution, did not conform to scientific methods described and prescribed by philosophers. He claimed that the use of such methods would have actually prevented the discoveries.
Feyerabend is much closer to the scientific than the philosophical establishment. His criticism is aimed primarily at science philosophers who have been struggling for centuries to prescribe an effective scientific method, while brilliant scientists, although philosophically uninformed, continued to make advances with extraordinary new discoveries, proving that successful scientific research does not conform to any models prescribed by philosophers. The scientific process is complex and philosophy cannot devise methods of differentiating science from pseudo-science or myth.
We will borrow the following excerpt from the Stanford Encyclopedia of Philosophy reporting an interesting incident during a Feyerabend lecture: “His listeners were enthralled, and he held his huge audiences until, too ill and too exhausted to continue, he simply began repeating himself. But not before he had brought the house down by writing “Aristotle” in three-foot high letters on the blackboard and then writing “Popper” in tiny, virtually illegible letters beneath it!”
Feyerabend’s work changed the focus of the philosophy of science away from the verificationism of the logical positivists and from Popper’s falsifiability criteria. This shift would take another turn toward a historical and evolutionary perspective with the work of Thomas Kuhn, an American physicist, historian and philosopher.
Kuhn was born in Cincinatti in 1922 and studied physics at Harvard, where he developed a strong interest in the philosophy of science. Kuhn is probably the most influential philosopher of science in the twentieth century. He taught history and philosophy of science at Harvard, Berkeley, Princeton and MIT. In 1957 he published his first book, The Copernican Revolution, where he studied the development of the heliocentric theory of the solar system during the Renaissance. Five years later he published his landmark work, The Structure of Scientific Revolutions, which offered a brand new perspective of the evolution of science.
Kuhn was the first author to articulate an alternative to the traditional view of scientific progress. In Kuhn’s terms, “normal science” is the regular experimental work scientists conduct within a given paradigm, a given framework of theories, rules and assumptions. Normal science is based on the assumption that the scientific community knows what the world is like. Normal science works within the existing scientific paradigm. Working within the current framework, scientists generally devote themselves to solving scientific puzzles. Their solutions reinforce and extend the scope of the framework without much interest in challenging it. When experimental results fail to conform to the existing paradigm, anomalies are created, which are often ignored until they accumulate, develop into a crisis, leading to a scientific revolution which establishes a new paradigm with new rules and theories. This last phase is revolutionary science and the transition from the normal to the revolutionary phase creates a paradigm shift, an expression that was not coined by Kuhn but was popularized by Kuhn and is in wide use today in many disciplines, including business. Its use has actually become a cliché.
One of the most important concepts in Kuhn’s theory is the idea of incommensurability. Well, this is a long fancy word but it has a simple meaning: different paradigms have no common standards of comparison. The languages of the two theories lack sufficiently overlapping meanings and their conceptual frameworks are so different that scientists are unable to use empirical evidence to compare one theory with the other. There are accounts of reality in the new paradigm that cannot be reconciled with certain aspects of the old paradigm.
The idea of incommensurability was actually introduced in 1962 by both Kuhn and Feyerabend independently. Probably an outcome of their discussions, as the two were close friends. Kuhn suggests that the proponents of each paradigm see the world in their own way because of their scientific training and prior experience. They use a different conceptual framework and have different ideas about scientific standards.
One of the problems that I have with Kuhn’s theory is that normal science and revolutionary science are considered as distinct and apart from each other. One follows the other in succession. But science researchers know that normal and revolutionary work co-exist and interact at all times, even within the same research project. They are not distinct from each other, as there are revolutionary elements in normal research and normal elements in revolutionary research. In fact, no one knows how to distinguish normal from revolutionary research. Scientific breakthroughs have occurred by accident during normal research. Most scientists believe there is a revolutionary element in their own research and quite often the outcome proves them right.
Kuhn paints an overly conservative picture of scientists, who, in his view, are engaged in puzzle solving within the existing paradigm. But we know that successful scientists are able to think out of the box and inside the box at all times. Natural phenomena that are discovered and cannot be explained within the existing paradigm set the stage for the creation of a new theory, which cannot appear from nowhere but from conflict within the existing paradigm. The history of science shows that the seeds of revolutionary ideas are always present and it takes a brilliant mind to make them produce the fruits of great new knowledge. The outcome of a scientific effort determines whether the effort revolutionized science or not.
After all, what is a scientific revolution? Kuhn does not define the term. Its meaning in Kuhn is derived by implication from considering the entire theory. I would propose that a scientific revolution is simply a discovery that opens multiple paths of new research. We will discuss a few examples of scientific revolutions to see where they conform to Kuhn’s model and where they do not.
Newtonian physics and relativity are often cited by philosophers as being two competing theories. But Newtonian physics is not really in competition with relativity, nor is it contradictory to relativity. Relativity applies to objects moving at speeds close to the speed of light, while Newtonian physics applies to objects moving at much lower speeds. In our discussion of relativity we saw that relativistic time and length differ from their classical counterparts by the Lorentz factor, which is a correction factor that depends on the velocity of an object relative to the speed of light. The relative velocity of motion of two objects is the same concept in classical physics and relativity. The concepts are the same and the mathematical language is the same. It is not conceivable that a trained physicist who specializes in relativity will have a problem communicating with another trained physicist who practices mostly classical physics. There are no such distinct categories of physicists or scientific communities.
It would be extremely difficult, if at all possible, for a physicist to prove anything in relativity work if the use of classical concepts were disallowed. Let us think for a moment how Einstein’s brilliant and revolutionary mind developed E=mc2. The derivation is remarkably simple and uses concepts and formulas from Newtonian mechanics, such as momentum, kinetic energy and the laws of conservation of energy and momentum. That is as classical as physics can be. It also uses Planck’s and de Broglie’s laws, which may be considered as laws of the new paradigm, but are still expressed in mathematical language that is fully understood in the classical framework. And most importantly, Einstein’s derivation is based on the assumption that the laws of physics are the same in all reference frames! The speed of light is also the same in all frames! Einstein’s derivation shows clearly that relativity’s genetic origins are rooted in Newtonian physics. The common concepts and common mathematical language offer a standard for comparison and we cannot claim that the two paradigms are incommensurate.
We often read that the Newtonian theory of gravitation is not compatible with Einstein’s general relativity. Newton considers that gravity is caused by a force, while Einstein’s theory tells us that it is caused by a geometric distortion of spacetime. Does relativity reject the existence of a force? There are physicists who believe so, but there is hardly any physicist who rejects the existence of nuclear forces, the forces that keep the atomic nucleus together. The concept of force, therefore, is not obsolete in modern physics. The force of gravity is something that we experience every day. In fact, this force can produce work, which is the most quintessential element of the concept “force”. The apparent incompatibility between Newtonian gravity and relativity is resolved when we consider that the Newtonian gravitational force is an effect of gravity, which is a geometric distortion of spacetime caused by large masses.
The preceding sentence is a clear statement of two cause-and-effect relations: the force is caused by the property of gravity, and the property of gravity is caused by a geometric distortion of spacetime near large masses. We have two successive orders of causal relations and the idea of incompatibility of the two theories must be refuted.
Coming now to special relativity, if we consider the speed of an object as context, then the differences between Newtonian physics and relativity are contextual, or frame-related. Structures, buildings and machines were built successfully before anyone knew that F=ma or E=mc2. Satellites orbiting the earth were designed with the application of classical physics and without the knowledge of any relativistic principles. Those satellites were launched successfully and are still operating.
It is clear that Newtonian physics is not applicable to velocities that are of the same order of magnitude as the speed of light and relativity rules in those instances. But we must remember that relativity, although widely embraced by the scientific community, has not been fully validated, at least not in all of its different parts. For a full validation we need technology which is not available right now. We need the ability to accelerate objects to velocities that are comparable to the speed of light. These objects cannot be tiny particles. They must have size and mass such that the most minute changes in mass are measurable. There is no technology currently available that can do this. Certain parts of relativity, therefore, are unverifiable and unfalsifiable with current experimental capabilities. In any event, the simple fact that velocity offers a standard of comparison between Newtonian and relativistic physics means that the two paradigms are commensurate.
The comparison of heliocentrism and geocentrism is another example which does not confirm Kuhn’s theory. The Aristarchean heliocentric theory co-existed with Aristotle’s geocentrism for several centuries, albeit not with the same popularity as Aristotle’s system. The Copernican paradigm has many connecting threads both with Aristarchus and with Aristotle. The irony is that Einstein’s theory of relativity upset both models. New evidence has also shown that the solar system’s center of gravity is not the exact center of the Sun. This means that either model is acceptable regardless of the fundamental differences between the theories. Astronomers use both the heliocentric and geocentric models for research depending on which theory makes their calculations easier. They use reference frames with the origin in the center of mass of the earth, or the earth-moon system, the sun, or the sun-planets system. Astronomers will often mix in the same study heliocentric velocity and momentum with geocentric co-ordinates. Their selection of geocentric or heliocentric frames is merely a matter of convenience and, in the final analysis, it is only an approximation of the actual spacetime. Their choice is made for ease of computation and does not have any great philosophical implications. If astronomers can work in both paradigms in the same study with such ease in their daily work, then we may say that the paradigms are commensurate.
Kuhn’s work has raised both praise and skepticism. The traditional view of the evolution of science is that science progresses sometimes in an orderly fashion and other times in a haphazard fashion, constantly fusing old and new concepts. If there are accounts of reality in the old paradigm that cannot be reconciled with the ideas of the new paradigm, they will either be discarded or modified. If they are modified successfully, they will become part of the new paradigm. They will likely influence concepts of the new paradigm. Doesn’t this fusion process remind us of the dialectical triad thesis-antithesis-synthesis that we saw in our discussion of Hegel? The Hegelian view of scientific progress is more intuitive than Kuhnian theory and is generally supported by scientists, while Kuhnian theory has been quite popular among philosophers.
There is no doubt that Kuhn’s influential work has fostered the development of a new area of study, the philosophy of science. Canadian philosopher Ian Hacking asserts that one of Kuhn’s marvellous legacies is science studies as we know it today.