This blog post examines whether scientific theories can transcend mere explanation to become descriptions of reality, focusing on Stephen Hawking’s black hole theory within the scientific realism debate.
In Zhuangzi’s butterfly story, known as the Butterfly Dream, Zhuangzi, having become a butterfly, could not distinguish whether he was dreaming or truly existing in reality. This is a fascinating thought experiment from classical literature. Christopher Nolan’s film “Inception” similarly unfolds around the theme of ‘dreams within dreams,’ garnering immense popularity in Korea by drawing approximately 5.9 million viewers. Thus, the question of whether the experiences we see and undergo are truly real has long been a core subject of philosophical debate, referred to as Philosophical Realism. Within the realm of science, a similar debate over scientific realism has unfolded, focusing on a thorough examination of the nature and status of scientific theories.
Today, highly developed sciences like physics, biology, and chemistry study subjects far beyond the range directly perceivable by human senses—from the birth of the universe to the forces operating within atoms. So, do objects we cannot directly observe, such as electrons, DNA, and black holes, truly exist? A crucial point to note here is that scientific realism, unlike the universal realism discussed in philosophy generally, already presupposes as a given fact that the observed object and the perceived existence do exist.
In his recently translated and published autobiography, “My Brief History,” theoretical physicist Stephen Hawking himself mentioned that his research was unlikely to win him a Nobel Prize during his lifetime. This is because his physical disability prevented him from directly participating in experimental physics, and his primary research subjects—black holes and quantum gravity theory—possess a nature that makes experimental verification in the near future difficult. The Nobel Prize in Physics is, in principle, awarded only for achievements that can be verified or observed through experimentation. This article, grounded in this critical perspective, aims to examine whether scientific theories are merely well-constructed tools for explaining phenomena or whether they can be understood as descriptions of reality, focusing on Hawking’s theories. Are scientists, like Zhuangzi who dreamed he was a butterfly, merely flying in a dream, or are they actually taking steps toward truth?
Scientific realism asserts that the objects studied by science actually exist. According to this position, scientific theories enable us to distinguish truth from falsehood, and the cause producing such results lies in a real world external to the human mind. In other words, the purpose of science is to provide a literally true account of how the world is. A core argument often presented by realists is the ‘miracle argument’. The miracle argument follows this logical structure: First, the development of scientific theories has enabled numerous predictions that were impossible in the past. Second, this success of science cannot be achieved merely by explaining observed results after the fact. Third, if scientific theories were merely explanatory tools, the phenomenon of such repeated accurate predictions would have to be regarded as miraculous. However, the assumption that miracles occur continuously in every field is unreasonable. Fourth, therefore, scientific theories must be understood not as mere explanatory tools but as descriptions of reality. Examples abound, such as manufacturing highly integrated semiconductors based on electron theory and developing new drugs based on theories of DNA and cellular processes.
In contrast, scientific non-realism views scientific theories as merely empirically adequate. Non-realists offer their own critiques of the miracle argument and claim numerous historical examples support their position. A prime example is the phlogiston theory. In the past, the combustion process was understood as the emission of particles called phlogiston. When an object to be burned was placed on a scale and ignited, a decrease in the object’s weight was observed after combustion. The phlogiston theory emerged to explain this phenomenon. However, today the phlogiston theory is clearly invalid. Therefore, the concept of ‘phlogiston’ does not exist, and scientific theories merely function as tools to explain phenomena, according to non-realists. Similarly, the theory that light propagates through an ether medium once held a dominant position, providing useful explanations and intuition in the wave-particle duality debate. However, it is now a scientifically established fact that no ether exists between the Sun and Earth. Therefore, ‘ether’ also does not exist. Thus, the core position of anti-realists is that a theory’s high explanatory power does not guarantee its truth. Anti-realists also point out that the argument from miracle commits the fallacy of affirming the consequent. That is, if the proposition “If p, then q” is true, it does not necessarily follow that the proposition “If q, then p” is also true. Inductive reasoning is prone to this fallacy when deriving general statements from observed cases. Some anti-realists also hold that scientific statements can only be falsified, not ultimately confirmed as true.
In response to these critiques, scientific realism can be defended more rigorously based on Leplin’s theory. Leplin proposed the ‘theory of novel predictions’. It is clear that the mere fact of being able to provide a post-hoc explanation cannot fully justify realism. However, when ‘novel’ predictions beyond the ordinary level are made, the scientific theory in question should be regarded as partially or approximately true. A prime example is the phenomenon of light bending by gravity, predicted by Einstein’s general theory of relativity. Newtonian mechanics, premised on the particle nature of light, could not explain this phenomenon under the principle of the constancy of the speed of light. In contrast, Einstein’s general theory of relativity introduced a new concept of spacetime, enabling the theoretical prediction of this phenomenon. This prediction was later verified through experiments measuring the deflection angle of starlight observed around the Sun during a total solar eclipse. Another example of a novel prediction is Fresnel’s diffraction experiment. During the intense debate over the wave-particle duality of light, Fresnel designed an experiment where light passed through a double slit into a dark box. The result was a bright spot at the center of the photosensitive film, along with diffraction patterns. This phenomenon could not be explained by existing optical theories and could only be precisely predicted in advance by Fresnel’s theory. At the very least, when a scientific theory presents novel predictions that go beyond common explanations, it is reasonable to view that theory as dealing with real entities.
Furthermore, it is necessary to establish a more universal standard for ‘novelty’. According to Choi Seong-ho (2006), the criteria for strong novelty are as follows. First, there is the condition of independence: the observation must be deducible using only that specific scientific theory. Second, the uniqueness condition requires that, at the time, only that scientific theory could provide a convincing basis for the prediction. Einstein’s refraction of light and Fresnel’s dark box experiment, mentioned earlier, satisfy both conditions. Einstein could deduce the refraction of light by the Sun’s gravity through his theory of relativity, which Newtonian mechanics at the time could not explain. Fresnel, too, could deduce the patterns appearing on photosensitive film based on the dual nature of light, which existing theories—viewing light as possessing only a single property, either wave or particle—could not account for. While cases satisfying both independence and uniqueness conditions are rare, they do exist in the history of science. Therefore, ‘novel predictions’—cases satisfying both independence and uniqueness conditions—can serve as sufficient conditions for judging that a scientific theory describes reality.
My position on scientific non-realism is as follows. Scientific realism and non-realism can be seen as focusing on different aspects of scientific theories. Realism emphasizes the predictive power of science, while non-realism stresses the explanatory power of scientific theories, arguing that such explanatory power does not necessarily correspond directly to reality. However, scientific theories provide excellent explanations while simultaneously enabling prediction. Scientific theories are not mere collections of descriptive sentences or mathematical propositions; they possess both explanatory power regarding the world of existence and predictive power concerning future phenomena. If the terms of a scientific theory performed only metaphorical functions, or if the explanations it offers were merely structural models, then the reason to call it an empirical science would also vanish. As seen in the novel predictive theory examined earlier, the conditions of independence and uniqueness can serve as criteria for judging the nature of a scientific theory. Unlike general realism debates, the scientific realism debate sees both sides agreeing on the existence of the object itself; the point of contention lies in the nature of the explanation. If the explanation possesses the power of unique, novel prediction, this signifies it is dealing with reality.
Counterarguments are also possible against non-realists who cite the repeated overthrow of existing theories in the history of science as evidence. Non-realists argue that statements about reality should be irreversible, but the fact that science has undergone multiple revolutionary changes does not justify this claim. Even if the explanatory framework changes, the fact that scientific theories refer to reality itself remains intact. For example, the phlogiston theory is no longer accepted as an explanation for combustion. However, the phenomenon of mass loss during combustion that phlogiston sought to explain is now accounted for through the evaporation of water vapor and its chemical combination with oxygen. Modern chemical theory excludes the old, erroneous explanation while more precisely encompassing the reality that the phenomenon pointed to. Similarly, Newtonian classical mechanics is no longer accurate when an object’s velocity approaches the speed of light. Yet under everyday conditions, most objects move very slowly compared to light speed (v≪c), and under these conditions, Newtonian mechanics is subsumed as a special case of relativity theory through the Lorentz transformation. The world described by Newtonian mechanics can be understood not as a mere abstraction, but as a part of reality, or as a three-dimensional approximation of four-dimensional spacetime. That is, scientific theories provide partially true intuitions about reality, and through scientific progress, we gradually approach reality.
The non-realist’s other claim—the limitations of human experience and the incompleteness of cognitive abilities—also faces criticism. Extreme relativism or skepticism cannot replace science. Even extreme relativists rely on rationality and reason in daily life. The assertion that all belief systems are relative or incomparable borders on evading verification and is hardly a fair argument. At the very starting point of the scientific realism debate, the existence of the observed object, the possibility of explanation, and the possibility of prediction are already presupposed. Of course, prevailing scientific theories or the observer’s psychological state can influence experimental design and data collection. Nevertheless, the very attempt to approach reality through observation and experimentation is an inherent characteristic of science. Scientific theories undergo global verification processes to establish their rigor. Even if they lack the pure deductive systems of mathematics or logic, they progressively approach truth and reality through experience.
Based on this discussion, we can examine Hawking’s black hole theory as a case that has faced criticism from scientific non-realists as merely a constructed theory. According to non-realists, Hawking’s black hole theory and quantum gravity theory do not deal with reality; they are merely mathematical devices introduced to explain the motion of the universe. However, Hawking’s cosmic theory can be seen as dealing with real objects when judged against the novel prediction criterion proposed by Replen. Specifically, black holes form an extremely strong gravitational field by absorbing mass, creating a region from which even light cannot escape. This boundary is called the edge of the black hole, or the event horizon. According to Hawking’s theory, quantum effects near the event horizon cause a faint emission of energy, known as Hawking radiation. This radiation is extremely faint and occurs at great distances, making it extremely difficult to observe with current technology. However, if experimental physics, including radio detection technology, advances sufficiently, or if equipment capable of detecting Hawking radiation is established in outer space beyond Earth, Hawking’s black hole theory could be empirically verified. Furthermore, Hawking’s theory can theoretically deduce the form and distribution of this radiation, satisfying the independence condition. Moreover, Hawking radiation is expected to contain information related to the star’s formation prior to its absorption by the black hole. No theory other than Hawking’s exists that can interpret this information. This satisfies the uniqueness condition. Therefore, Hawking’s black hole theory can be seen as presenting novel predictions that satisfy both the independence and uniqueness conditions. Although the fact that experimental equipment to directly verify these predictions did not emerge during Hawking’s lifetime is a separate issue, the fact that this theory enables novel predictions cannot be denied. Therefore, Hawking’s black hole theory can be evaluated as a theory dealing with reality. Even if it is not a complete theory of black holes, the existence of an entity emitting energy in space is at least undeniable.
In conclusion, through Hawking’s major research achievement—black hole theory—we can reasonably infer that entities related to quantum gravity exist in the universe. Scientific progress will gradually reveal this reality, leading human understanding to deeper and more sophisticated levels. Scientists can be seen not as beings wandering in dreams, but as those who approach reality through imperfect yet accumulating knowledge.
