In this blog post, we will examine the arguments for and limitations of the theory of intelligent design—based on William Dembski’s ‘Intelligent Design’—and critically evaluate whether it can qualify as a scientific theory.
‘Intelligent Design’ is a book that argues the world was created through the design of an intelligent being. The author, William Dembski (hereinafter Dembski), argues that much of the controversy surrounding the existence of God has arisen because previous debates relied solely on theological methods. Accordingly, he claims to have proven the existence of God using a scientific method that anyone can examine. He also explains that the scientific method he refers to here is a concept distinct from the scientific method used by naturalists.
Demske’s core argument unfolds from Chapter 4 through Chapter 6. Chapter 4, as its title “Naturalism and Its Cure” suggests, argues that the naturalism currently dominating science is flawed and that intelligent design offers an alternative to overcome it. He contends that as long as science is premised on naturalism—which does not acknowledge supernatural beings and seeks the causes of natural phenomena solely within nature—Darwin’s theory of evolution is bound to prevail. Therefore, Demski argues that to break this circular logic, naturalism must be rejected, and intelligent design must be accepted as its alternative.
Chapter 5 explains how the claim that the world was created through the design of an intelligent being can be addressed within the realm of science. Demske uses the Search for Extraterrestrial Intelligence (SETI) program as an example to explain how traces of an intelligent being can be detected. He states that three characteristics must be identified to make a case for design: contingency, complexity, and specification.
Contingency means that while an object or event follows the rules that caused it to occur, it cannot be fully reduced to those rules alone. Demske uses ink printed on paper as an example. The ink must follow the physical and chemical laws governing the interaction between the paper and the ink, but the shape of the printed characters itself cannot be explained by those laws alone. Since the shape of the characters is ultimately determined intentionally by a person, it cannot be fully explained by physics and chemistry alone. However, this does not mean that the shape of the ink does not follow the laws of physics or chemistry. He argues that if an event is not indeterminate, it is a necessary event—that is, a naturally self-fulfilling event—and therefore cannot serve as evidence of design.
Complexity is a concept related to the probability of an event occurring. The more complex an event or object is, the lower the probability that it will occur by chance. Conversely, if an event is not complex and occurs very frequently, it can easily happen by chance alone, so it is difficult to view it as evidence of design. Therefore, complexity is also explained as an essential element required to argue for design.
However, even if an event possesses complexity, if it lacks specificity, it may still be the result of chance. For example, suppose a coin is flipped 100 times. If we record 1 for heads and 0 for tails, we obtain a binary sequence of length 100. Although the probability of such a sequence occurring is very low, we do not conclude that there is a special intention behind it. Therefore, Demske introduces the concept of specificity. Specificity refers to the property that allows us to determine, through supplementary information, that a certain event holds special significance. For example, a sequence of prime numbers from 1 to 100 expressed in binary can be viewed as information possessing specificity based on the supplementary information that they are “prime numbers.” In this case, the supplementary information and the event being interpreted must be independent of each other.
After explaining all three criteria, Demske explains why his criteria are valid. In this process, he raises the possibility of false positives (errors in which something is judged to be evidence of design when it is not) and false negatives (errors in which something is judged not to be evidence of design when it actually is). He argues that if evidence for design is found according to his criteria, the real issue is a type I error; furthermore, he claims that it can be proven that such a type I error cannot arise from inductive generalization.
In Chapter 6, he attempts to apply the criteria presented earlier to actual cases. Information theory plays a central role in this process. He explains that, according to information theory, information in a closed system can only be maintained or reduced by the laws of nature. Demske then argues that the only pathway through which new information can be generated is chance. However, he maintains that the amount of information that can be generated by chance cannot exceed the cosmic probability limit he has proposed. He asserts that if an event with an information content exceeding 500 bits is discovered, it is not the result of chance but rather information that proves intelligent design. Based on the preceding explanation, he defines this type of information as CSI (Complex Specified Information). Finally, he concludes that since bacterial flagella possess an information content of more than 500 bits, CSI actually exists, and therefore an intelligent designer exists.
My first thought after reading this book was that “there are many logical leaps.” Overall, rather than logically presenting evidence for his claims, this book places a heavy emphasis on explaining the reasons behind those claims. In other words, Demske appears to be confusing the context of discovery with the context of justification. The context of discovery refers to explaining the process through which the author arrived at his ideas, while the context of justification is the process of logically proving why his claims are valid. However, in presenting the new claim of “intelligent design,” Demske focuses on the context of discovery rather than the context of justification. Consequently, from the reader’s perspective, it is difficult to fully verify the book’s central claim that “intelligent design has been proven.” Yet even setting aside this logical leap, I believe that intelligent design is unlikely to become a scientific theory.
In other words, I intend to critique Demske’s argument from two perspectives. The first is an external critique: there is no reason for the scientific community to accept intelligent design at the cost of abandoning naturalism. The second is an internal critique, pointing out problems with the very structure of his logical argumentation.
Before presenting the first critique, we must first examine the process by which a hypothesis is accepted as a scientific theory. What is a scientific theory? The Oxford English Dictionary defines a scientific theory as “a hypothesis or general law that has been confirmed or verified through observation or experimentation and is recognized as providing the best explanation of known facts.” In other words, a scientific theory can be described as a framework that effectively explains what we know about the universe.
This implies two important facts about scientific theories. First, “explaining the world well” means that a theory must be able to explain or predict not only facts observed so far but also those yet to be observed. Second, a scientific theory “explains the world well”; it does not represent an absolute truth that perfectly explains every fact about the world.
Since the past consists of events that have already occurred within a finite span of time, we can explain a significant portion of it even without scientific theories. For example, even without the theory of evolution, we can verify to some extent which organisms have existed in Earth’s history through fossils. However, we can predict which organisms will emerge in the future based on scientific theories such as the theory of evolution. Newtonian mechanics, which humanity has utilized for a long time, has not only explained planetary motion to date but has also been able to predict future planetary motion with high accuracy.
The important point here is that Newtonian mechanics is not an absolute truth. While Newtonian mechanics explains planetary motion quite accurately, Einstein’s theory of relativity is necessary for more precise predictions. Furthermore, quantum mechanics is required to fully explain the microscopic world even within the framework of the theory of relativity. Thus, scientific theories are accepted because they provide powerful frameworks for understanding the world, not because they are absolute truths that perfectly explain it.
This point is easier to understand by borrowing Plato’s philosophical perspective on the relationship between the mathematical world and the physical world. Scientific theories can be viewed as metaphysical models that use the language of mathematics to describe the physical world. In fact, both Newtonian mechanics and the theory of relativity are defined within mathematical frameworks. While these metaphysical theories are extremely helpful in understanding the real world, they cannot perfectly replicate reality. If we define “truth” as a complete understanding of the physical world, then scientific theories are not truth itself, but rather models that provide a highly sophisticated approximation of reality. Therefore, for a hypothesis to be accepted as a scientific theory, it must undergo verification regarding how effectively it can explain and predict the real world. This is referred to as the explanatory power of a scientific theory.
It is important to note here that the question of absolute truth is not the central issue in this process. Because human cognitive abilities are limited, it is practically impossible to fully grasp absolute truth. Furthermore, discussions regarding truth itself fall more within the realm of philosophy or religion than science. Since Demski also argues that intelligent design is a separate discussion from theology, it is appropriate to set aside the debate over truth and instead judge a scientific theory’s explanatory power based on how well it explains the facts observable by humans.
Now, let’s examine why intelligent design cannot be a scientific theory. Intelligent design is severely lacking in this very explanatory power required of a scientific theory. Demske argues that there are still many biological phenomena that evolutionary theory cannot explain, and that intelligent design can account for these gaps. In other words, he believes that intelligent design enhances the explanatory power of a scientific theory more than naturalistic evolutionary theory does. However, as mentioned earlier, a scientific theory is not a truth that perfectly explains the universe. Therefore, it is only natural that there are aspects evolutionary theory has yet to explain. In fact, evolutionary biology has been constantly revised and refined based on new evidence and research findings. Demske argues that, in the course of these revisions and refinements, we must abandon naturalism itself. Ultimately, the question we must address is whether scientific explanatory power actually improves when the scientific community abandons the methodology of naturalism and acknowledges an intelligent designer. In other words, we must examine whether abandoning naturalism enhances the explanatory power of scientific theories.
Let us assume that we accept the theory of intelligent design. Let us suppose that this theory can explain how all life has reached its current form through a process of creation by an intelligent designer. What, then, will happen in the future? The theory of intelligent design does not provide a framework capable of predicting how biological phenomena will change in the future. This is because the intelligent designer is assumed to be an entity not bound by any natural laws. Even if we assume that the intelligent designer’s intentions can be known, if that entity were to follow certain laws, it would already become an entity explainable within the framework of natural laws. Therefore, under the premise that the intelligent designer’s intentions cannot be predicted, it is also impossible to predict what biological phenomena will emerge in the future. In other words, while the theory of intelligent design may claim to explain past phenomena, it fails to provide any insight or prediction regarding how life will change in the future. Once the theory of intelligent design is accepted, it is difficult to secure the ability to make predictions about the future, no matter how much the theory is developed. In this respect, its scientific explanatory power is inevitably very limited.
In contrast, the theory of evolution has explained a significant portion of biological diversity from the origin of life to the present, and it also provides a framework for predicting how biological populations might adapt and evolve as the environment changes in the future. Furthermore, evolutionary biology has continuously expanded its explanation of the history of past life forms through new fossil discoveries, genetic research, and molecular biology studies. The past is a finite span of time spanning approximately 4.6 billion years, while the future is an ever-continuing flow of time. While intelligent design theory can offer explanations for the past, it fails to provide a framework for explaining the future. In contrast, evolutionary theory—though not perfect—can continuously expand its explanatory power to encompass both the past and the future. Therefore, abandoning naturalism in favor of intelligent design theory could significantly weaken the ability of scientific theories to predict the future.
The second criticism concerns the substance of Demski’s argument. In particular, there is a serious logical leap in the line of reasoning that claims the complexity-specificity criterion can successfully avoid false positives. Demski argues, “Whenever the complexity-specificity criterion determines that design is present, and the intrinsic causal processes of that case are subsequently elucidated, it turns out that design actually exists. Therefore, whenever the complexity-specificity criterion indicates design, design actually exists.” However, he does not provide a concrete explanation of which specific cases actually support this conclusion. Furthermore, while he describes his logic as inductive generalization—a method of deriving a general proposition through a logical process that connects multiple individual cases— However, in Demski’s argument, such an inductive process is not sufficiently presented.
The part where he explains the world by analogy with a function is also difficult to accept. Demski argues that if the world is governed by laws, those laws are deterministic, so the total amount of information remains constant or decreases. However, can we really say that the real world is completely deterministic?
The mathematician Laplace proposed a hypothetical entity known as “Laplace’s Demon.” Laplace’s Demon is assumed to know all the information about the entire universe at any given point in time. If so, this entity would be able to accurately calculate the state of the universe after a certain period of time by applying all the laws governing it. In other words, this means it could perfectly predict the future. However, modern physics—particularly quantum mechanics—demonstrates that even with full knowledge of all information, perfectly predicting the future is fundamentally impossible. In quantum mechanics, the state of a particle is described probabilistically, and certain physical quantities cannot, in principle, be precisely determined simultaneously. Furthermore, quantum field theory explains that even the vacuum state is in a state of constant flux due to quantum fluctuations. Therefore, the claim that “the total amount of information is always conserved or decreases”—a core premise of Demski’s logic—is difficult to accept as is from the perspective of modern physics.
Meanwhile, a former student assessed in an essay that Demski’s logic could be mathematically valid. However, the student argued that intelligent design is not recognized as a scientific theory because it has not undergone a broad verification process within the academic community. Furthermore, just as quantum mechanics, string theory, and the theory of parallel universes are accepted as scientific theories—even though their central concepts cannot be directly proven—because their surrounding hypotheses and predictions have been verified, the student argued that intelligent design could also be recognized as a scientific theory if its surrounding hypotheses are verified, even if its core hypothesis cannot be directly proven. As an example, the student cited the actual existence of CSI.
In fact, even in Newtonian mechanics, the central concept of force itself cannot be directly proven. However, by assuming the relationship F = ma, it has very successfully explained numerous natural phenomena, including the motion of the solar system, and is therefore widely accepted as a scientific theory. Similarly, quantum mechanics is recognized as a scientific theory not because it directly proves the concept that particles exist probabilistically, but because the theory explains and predicts numerous experimental results with a very high degree of accuracy. String theory has not yet undergone sufficient experimental verification; as a result, it is often regarded as a research program in theoretical physics or a hypothetical theory rather than an established scientific theory. Furthermore, while various forms of the parallel universe theory have been proposed, it has not yet been accepted as an experimentally established scientific theory. In contrast, the theory of relativity has had several of its predictions—such as time dilation, the bending of light by gravity, and gravitational waves—repeatedly verified through observations and experiments. In comparison, the theory of intelligent design has faced criticism for failing to present sufficient verifiable research results that independently predict or explain natural phenomena.
However, as stated earlier, a scientific theory must provide a good explanation of the world. As suggested by the view introduced earlier, the verification of related hypotheses can imply that the theory effectively explains the real world. Nevertheless, a scientific theory must not merely explain current phenomena; it must also be able to offer testable predictions regarding new observations and experimental results. Even if some of the peripheral hypotheses of intelligent design were to be examined, I still believe that intelligent design is unlikely to be recognized as a scientific theory because it fails to provide predictive power for the future. Quantum mechanics, as well as the theory of relativity, have evolved by predicting new phenomena and having those predictions verified. Research is also ongoing in string theory and the theory of parallel universes to establish this predictive capability. In this regard, I judge that intelligent design theory differs in nature from these fields.
So far, we have examined the criticisms that can be raised both from within and outside the intelligent design theory. It is difficult to view intelligent design theory as a theory that sufficiently explains the universe, including biological phenomena, and it is also difficult to consider it as possessing the explanatory power and predictive capability required of a scientific theory. Furthermore, there are several leaps in logic and areas requiring further scrutiny even within its internal line of reasoning.
Researchers who support the theory of intelligent design consistently point out the limitations of evolutionary theory and present their claims as a scientific alternative. In fact, in South Korea, some creation science-related organizations have in the past filed petitions arguing that content related to evolution in the curriculum and science textbooks should be revised or that content related to creationism or intelligent design should be included. However, these attempts failed to gain broad consensus within the academic community, and South Korea’s current science curriculum continues to maintain the theory of evolution as a core component, based on the research findings of modern biology. For the theory of intelligent design—which has fundamental limitations—to be recognized as a scientific theory, it must present not only philosophical arguments but also testable predictions and reproducible scientific evidence.
