This blog post examines how Popper’s falsificationism and Kuhn’s paradigm theory can coexist within scientific progress.
The development of science began around the 16th-17th centuries when scientists like Galileo, Kepler, and Newton systematized classical physics. Later, in the 19th century, Einstein, Bohr, and Schrödinger created the new discipline of quantum mechanics, profoundly altering humanity’s understanding of the world. Meanwhile, viewing this scientific journey from a philosophical perspective raises fundamental questions about where and how science is progressing. Does truth exist in science? How does scientific progress occur? How did past theories fail or succeed to shape current science? By addressing these questions, we can contemplate the direction science should take in the future. Many philosophers of science have developed their own theories on this matter. Karl Popper’s ‘falsificationism’ and Thomas Kuhn’s ‘scientific paradigms’ are two representative theories that stand in opposition to each other. This essay briefly introduces Popper’s falsificationism and Kuhn’s paradigm theory, then then, from the perspective of scientific progress, I will describe how Kuhn’s ‘paradigm’ is correct in some aspects and incorrect in others, focusing on Popper’s ‘falsificationism’. Finally, I aim to show through examples that paradigms are subordinate to falsificationism, demonstrating that the two theories of Popper and Kuhn are concepts that can coexist.
First, to define scientific progress, we need to distinguish science from non-science, i.e., pseudoscience. Popper proposed a clear criterion for this: ‘falsifiability’. That is, if a hypothesis or theory has a structure that cannot be logically criticized, it is pseudoscience. The hypothesis ‘God exists’ cannot be falsified by any means, so it cannot be considered science. Conversely, the hypothesis that ‘the magnitude of Earth’s gravity is uniform at every point on Earth’ can be falsified by measuring gravitational force at two points on Earth with significant elevation differences. This is ‘scientific’ from Popper’s perspective. Thus, ‘scientific’ as defined by Popper does not mean distinguishing truth from falsehood, but rather signifies ‘meaningfulness’. Furthermore, for such hypotheses to be valuable, they must have the purpose of solving practical problems of the time; philosophical or scientific propositions lacking this purpose are nothing but meaningless nonsense. Meaningful and falsifiable conjectures undergo tests attempting to falsify them, and if proven false in these tests, the conjecture is discarded. Since such conjectures cannot be proven entirely true, Popper defines conjectures that remain unproven false after multiple tests as having high reliability. By repeatedly testing these meaningful conjectures and their falsifiability, science approaches the truth, the essence of all things, and this is termed scientific progress.
To summarize Popper’s argument on ‘falsificationism’: science develops through a cumulative process over history, repeating cycles of conjecture and falsification. In response, Thomas Kuhn introduced the concept of ‘paradigms’ to science, presenting a theory contrasting with Popper’s claims. The paradigm defined by Kuhn is somewhat ambiguous in its meaning and scope, but broadly speaking, it refers to the overall characteristics of science that define an era. For example, the ‘geocentric theory,’ accepted as truth before Galileo, can be seen as one paradigm. Just as this geocentric model is no longer accepted today, given the evidence proving the heliocentric model, a paradigm is formed by the beliefs of people at the time; it is a relative concept, not an absolute one. When problems that the existing paradigm cannot explain accumulate, a new paradigm takes hold. Kuhn defined this process of a new paradigm taking hold as a ‘scientific revolution,’ and this is the most contentious point of disagreement between Kuhn and Popper. Unlike Popper, who argued that scientific knowledge gradually accumulates through repeated cycles of making better conjectures based on various hypotheses and falsifying existing ones, thereby advancing toward scientific truth, Kuhn contends that such scientific revolutions entail the collapse of achievements built upon the existing paradigm and the establishment of a new one. Furthermore, while he did not deny that scientific revolutions directly lead to scientific progress because the new paradigm solves more problems than the old one, he rejected the notion that this development of science leads to any particular truth. This is because a paradigm shift is not about moving in one direction continuously, but about finding a better, new direction.
However, I argue that Popper’s falsificationism and Kuhn’s paradigm should coexist rather than be mutually exclusive structures. To make this argument, which contrasts with both scholars’ theories, persuasive, I will first explain paradigms and the phenomena they produce based on falsificationism. The most significant difference between Kuhn’s paradigm and Popper’s falsificationism lies in the question of whether knowledge accumulates. A paradigm shift implies the collapse of knowledge accumulated under the old paradigm. Furthermore, Kuhn’s paradigm theory introduces the Kuhn-loss problem: the new paradigm may fail to explain phenomena the old paradigm could account for. These issues arise from a flawed conception of scientific progress. Ultimately, since the new paradigm originates from research grounded in the old paradigm, even if a paradigm shift occurs through a scientific revolution, the claim that the new paradigm destroys the old is difficult to accept. Consider the ‘dual nature of light,’ which served as the foundation for quantum physics. Previously, light was regarded solely as a wave. However, phenomena like the photoelectric effect (where photoelectrons are emitted from a metal plate only when light of a specific frequency or higher is shone on it, regardless of the light’s intensity) arose that could not be explained by light’s wave nature alone. This led to the hypothesis that light is sometimes a wave, and sometimes a particle. This hypothesis emerged, placing the existing paradigm of light as a pure wave in crisis. Later, it was revealed that light is an entity possessing both wave and particle properties, and this became the new paradigm. However, even after the scientific revolution, research based on light’s particle nature undermined the existing paradigm’s hypothesis that light was a pure wave. Yet, since light’s particle nature often need not be considered, it is frequently treated simply as a wave. Research on the characteristics of electromagnetic waves conducted before the paradigm shift remains valid. This is a case where a new paradigm has taken hold, yet knowledge from the old paradigm remains intact rather than being completely dismantled. In other words, the scientific revolution achieves scientific progress not by establishing an entirely new paradigm after the collapse of the old one, but by improving the existing paradigm. The extent to which the old paradigm collapses varies depending on the degree of improvement. To illustrate with the model of tower construction: when a problem is discovered on the fifth floor of a tower built up to ten stories, the solution is not to demolish and rebuild the entire tower. Instead, only the problematic section and its associated parts—specifically, the floors above the fifth—are demolished and rebuilt.
Let’s interpret the example of light in light of Popper’s falsificationism. The existing paradigm of optics can be seen as a collection of hypotheses and theories that have accumulated credibility by successfully passing numerous falsification tests up to now, and the theory that light is a wave is one of these. The photoelectric effect was a successful falsification of the theory that light is a wave. This led to the hypothesis that light is both a wave and a particle. The theory that light is composed of photons possessing both wave and particle properties has survived subsequent falsification tests to this day. A paradigm shift does not mean the complete replacement of an existing paradigm with a completely different one; rather, it involves modifying people’s perceptions through falsification, and this is precisely what constitutes scientific progress.
The Kuhn-Roth problem arises from hasty conclusions about scientific revolutions and limiting scientific progress solely to explaining more phenomena immediately. We can consider two cases where a new paradigm fails to explain what the old one could. The first case occurs when the new paradigm is not yet sufficiently established. A prime example is the most significant paradigm shift in astronomy: the transition from the geocentric model to the heliocentric model. First, let’s examine this through Kuhn’s paradigm theory. The geocentric model could explain various celestial phenomena without contradiction based on measurements and calculations derived from observed data at the time. However, as time passed and celestial phenomena like the phases of Venus became difficult to explain within the geocentric framework, figures like Galileo and Copernicus began to question it. Centering on these individuals, a new paradigm—heliocentrism—emerged. Yet, early heliocentrism involved far more complex calculations of celestial motion compared to the widely accepted geocentric model, and it lacked sufficient evidence and persuasive power. Moreover, Tycho Brahe’s modified geocentric model provided a more accurate explanation for the errors in the geocentric model that the heliocentric theory sought to correct. The heliocentric theory only gained acceptance as the established model after its accuracy was proven through the theories of scientists like Kepler and Newton (Kepler’s laws, the law of universal gravitation, etc.). Consequently, the paradigm shifted from geocentrism to heliocentrism. However, before heliocentrism was fully established, the heliocentric theory proposed by Copernicus faced a Kuhn-Ross problem. This occurred because the new paradigm, heliocentrism, was less persuasive than geocentrism. Let’s reinterpret this based on falsificationism. When Copernicus proposed the heliocentric theory, the geocentric model still held sway as the dominant theory. At that time, the heliocentric theory could not yet be considered a new ‘paradigm’; it was merely a hypothesis attempting to point out the contradictions of the geocentric model. This hypothesis was falsified by the cumbersome nature of the existing geocentric model and Tycho Brahe’s modified geocentric model, and it failed to spark a scientific revolution. Later, with the advancement of physics, a more systematic heliocentric theory reemerged as a hypothesis. This new theory successfully refuted the geocentric model by explaining everything the geocentric model could explain without issue, while simultaneously resolving the contradictions inherent in the geocentric model. Only then could the heliocentric theory be considered established as a new paradigm. If a new hypothesis cannot explain phenomena that should be explainable within the existing paradigm (phenomena directly observable in real life), this does not mean the scope of explainable phenomena has shrunk since the scientific revolution; rather, it represents an attempt toward a scientific revolution. If this hypothesis gains persuasive power to falsify or surpass the existing hypothesis, a paradigm shift occurs. Conversely, if the new hypothesis fails to gain persuasive power, no scientific revolution occurs, and the existing paradigm persists.
The second case occurs when a paradigm established through the first case fails to explain what the existing paradigm could explain. Before quantum mechanics developed, when Newtonian mechanics encompassed all of mechanics, it was believed that any motion could be accurately predicted if the physical quantities of an object were sufficiently known, no matter how small the object. This belief held true even for the motion of atoms. However, as research into the microscopic world (the small world invisible to the naked eye) continued, quantum mechanics overturned the paradigm of classical mechanics. Subsequently, the theory that in the microscopic world, it is impossible to accurately predict an object’s physical quantities nor to directly verify them, due to the principle of uncertainty, was accepted. In other words, the motion of particles that classical mechanics could explain became inexplicable under the newly established quantum mechanics. However, in quantum mechanics, this ‘inability to explain’ itself becomes the core of scientific progress and the new paradigm.
Kuhn also proposed the ‘incommensurability’ principle, stating that different paradigms lack a common measure. As one example, the Pythagorean paradigm, which held that all things are composed of rational numbers, became incompatible with the new paradigm of numbers established later when it was discovered that the ratio of the hypotenuse to the other sides of a right isosceles triangle could not be expressed as a rational number. This incompatibility established the principle of incommensurability. Kuhn argues that incommensurability makes comparison between two paradigms impossible. However, if we view scientific revolutions as improvements to paradigms, there is no need to consider ‘incommensurability’. If there are incommensurable aspects between the old and new paradigms, those aspects represent areas where problems were discovered through falsification within the old paradigm, leading to its abandonment and the establishment of new hypotheses and theories. This knowledge is distinct from that inherited from the old paradigm. Once it was discovered that the ratio of the hypotenuse to the other sides of a right isosceles triangle cannot be expressed as a rational number, there is no need to compare this with the Pythagorean theory that previously attempted to express it as a rational ratio. Conversely, there is also no need to discard the Pythagorean theories about right triangles that were previously explainable using rational numbers. Knowledge that could be well explained by the existing paradigm need not be dismantled; thus, the irreconcilable aspects are knowledge that collapsed during the scientific revolution. Comparing it to a new paradigm that withstood falsification tests and gained higher credibility is unnecessary.
In conclusion, viewing a paradigm of an era as a set of falsifiable hypotheses and theories from a falsificationist perspective, a scientific revolution can be seen as the ‘emergence of an improved paradigm’ accompanied by new theories and hypotheses that explain the contradictions discovered through falsification of the existing paradigm. This demonstrates how scientific revolutions achieve scientific progress without needing to consider the incommensurability between paradigms. This allows us to clarify the relationship between scientific progress and scientific truth. Truth is the very essence of all things, and humans have no way to know with certainty whether this essence exists or whether we have reached it. However, as the falsification of an existing paradigm and the improved new paradigm that follows can explain more than before and describe more about the world, science can be seen as advancing toward that truth.
