In this blog post, we explore the necessity and limitations of scientific methodology through the theories of Feyerabend and Lakatos.
While Popper, Kuhn, and others offered various explanations about the methodology of science, comparisons and changes between theories in science, and the superiority of scientific knowledge, and various debates arose about this methodology and its limitations, Feyerabend became a hot topic by presenting an anarchistic and challenging epistemology. He opposed the methodology represented by the phrase “Anything Goes” and advocated for greater freedom for scientists. On the other hand, Lakatos advocated research program theory, arguing for the rationality of science while also proposing a methodology consistent with history. Unlike existing methodologies, he attempted to impose rules on the process and methods of science. Feuerabend viewed this attempt positively from his own perspective, and the two had such active exchanges that they planned to co-author a book. From this perspective, we will examine the theories of Feuerabend and Lakatos, which are different but have many similarities.
Feuerabend’s core argument can be summarized by his phrase, “Anything Goes.” First, he argues that all methodologies proposed so far have been unsuccessful. The history and process of science cannot be explained by a few methodological rules, and attempts to reinforce such methodological rules ignore the various conditions that influence science, making science inflexible and unrealistic, as well as unhelpful to science. Furthermore, most of the world that humanity seeks to explore is unknown, and in such a situation, it is not desirable to start with restrictions imposed by methodology. Feyerabend does not simply oppose methodology. He believes that a “humanistic attitude” should be introduced into science. His opposition to methodology is also intended to eliminate all coercion by methodology in science and increase individual freedom. In Feyerabend’s view, the way science is institutionalized in our society today is incompatible with a humanitarian attitude. Today’s society teaches science to its members as a matter of course and makes it compulsory for everyone to receive the same science education. Members of society have no right to demand to learn other knowledge instead of science. Teaching science in such a coercive and rigid manner prevents individual characteristics, such as each person’s intuition and way of thinking, from influencing scientific research, thereby hindering not only individual freedom but also the advancement of science. Furthermore, Feyerabend compares the absolute authority of science in modern society to the status of religion in the past, arguing that society must be liberated from science just as it was liberated from religion in the past. This will enable the creation of a society that is ideologically neutral, where science is not considered superior to other forms of knowledge.
Feuerabend’s “Anything Goes” is also appropriate for revealing his perspective on the problem of theory selection. He first asserts the impossibility of comparison. The impossibility of comparison means that different theories use different definitions, concepts, and standards, making it impossible to compare them logically. This implies that subjectivity is involved in the problem of theory selection in science. Feuerabend ultimately argues that the problem of theory selection goes beyond the involvement of subjectivity and depends entirely on each individual’s subjectivity, and therefore, no matter which theory is chosen, the scientist’s choice cannot be criticized as “irrational.” At this point, each theory attempts to persuade scientists through various propaganda activities.
While Feyerabend’s argument is consistent with anarchism, Lakatos’s theory differs greatly from Feyerabend’s in that it does not present a methodology, but rather pursues the rationality of science and presents explanations regarding the process of science and the selection and comparison of theories. Lakatos’ research program includes positive and negative discovery methods. The negative discovery method states that the core premises underlying the program, i.e., the “hard core” of the program, cannot be rejected or modified. If the hard core is rejected or modified, the scientist has already decided to abandon the research program. The positive discovery method consists of several guidelines for the research program. In order for a research program to be considered scientific, it must be consistent enough to derive a clear program for future exploration and, at least occasionally, succeed in predicting new facts. If such a research program continues to succeed in making new predictions, the theory becomes a progressive research program; conversely, if it fails to make new predictions, it becomes a regressive research program. This is also used to judge the relative value of different research programs when comparing them. Regressive research programs give way to progressive research programs. However, since even regressive programs can become progressive through research, and it can take a long time to predict new facts, Lakatos’ explanation does not tell us which theory a scientist should choose when there are two competing theories.
Feuerabend’s theory has been subject to various evaluations and criticisms, as it is bold and challenging. A.F. Chalmers accepts the essential parts of his theory but criticizes it in its application to reality. He responds positively to Feyerabend’s opposition to methodology, saying that his argument is successful, but regarding Feyerabend’s core theory of individual freedom, he says that Feyerabend uses Mill’s concept of freedom as freedom from all constraints, but overlooks the positive aspects of constraints. Since individuals are born into an already existing society, they do not have free choice, and individual freedom is determined by the position that individual occupies within the social structure. In order to improve science and society, one must ultimately start from the objective situation in which one currently finds oneself and rely on existing means, so Feyerabend’s ideal is not practical. Chalmers argues that if Feyerabend’s position is adopted, it is likely that those in power will continue to hold power, citing John Krieger’s interpretation that “anything goes actually means that everything stays the same,” and gives an overall assessment of Feyerabend’s position.
Certainly, it is not easy to reject the essential parts of Fireball’s theory. Chalmers criticized Fireball’s theory, saying that science would flow in a direction that maintains the status quo of power, but this risk is more likely to appear in other theories that have methodologies. If Copernicus and Einstein had not broken with the existing direction of research and proposed the heliocentric theory and the theory of relativity, respectively, and if they had not continued their research despite the errors and unresolved problems that existed when their theories were first proposed, would the geocentric theory and Newtonian mechanics have remained dominant for longer? In this respect, Feyerabend’s theory of allowing scientists to explore freely is appealing. However, there is an extreme side to his argument. Feuerabend’s “Anything Goes” seems to apply not only to methodology but also to his view of disciplines other than science. He argues that there is no reasonable way to say that science is superior to any other discipline, and that people should be able to study astrology instead of science if they wish. In other words, he allows all “pseudosciences.” However, not all disciplines that are not science are called pseudoscience. “Pseudoscience” literally means “pretending to be science,” and it remains pseudoscience because it lacks what is necessary to be science. Should such pseudoscience be evaluated on the same level as science? Of course, as Feyerabend says, this may be because we lack research and knowledge about pseudoscience. However, there seem to be factors that explain why these pseudosciences are not accepted in academia or society, even if they cannot justify or explain their distinction from science. Furthermore, regarding Feyerabend’s impossibility of agreement, he argued that since the two theories cannot be compared, it is impossible to say which theory is superior, and that the choice of theory depends on the preferences of individual scientists and the propaganda of each theory. However, even if theories cannot be compared, the performance of those theories can be compared. This data cannot predict the future, and therefore the results may change, so it cannot justify the choice of theory, but it can explain why more people prefer certain theories or why it is more natural to prefer certain theories.
Regarding Lakatos, Chalmers points out that Lakatos fails to present clear criteria for rejecting certain research programs or selecting one of competing research programs. Since it is only possible to determine whether a research program is progressive or regressive after a period of time, it is impossible to give advice to scientists who are faced with situations that require immediate decisions. Lakatos sought to establish rules that would prevent intellectual corruption by pursuing the rationality of science, but as he himself acknowledged, his theory cannot provide advice to individual scientists. However, this point is also the reason why Feyerabend evaluates Lakatos’ theory positively, unlike empiricism, falsificationism, and paradigm theory. Feuerabend argues that Lakatos’ theory does not impose rules on scientists, but merely provides criteria to help them evaluate their own situations. Unlike other theories, Lakatos’ theory does not criticize the choice of a regressive program, but also mentions the possibility that a regressive program may become progressive again. He evaluates Lakatos’ theory as anarchism disguised as something other than anarchism. However, he strongly criticizes the fact that Lakatos uses the premise that science, especially physics, is superior to other disciplines as if it were a proven fact.
Thus, Lakatos’ theory and Feyerabend’s theory seem to be different, but from Feyerabend’s point of view, they are similar. He goes so far as to liken Lakatos’ theory to a Trojan horse for anarchism. If so, I would like to combine Lakatos’ theory with Feyerabend’s theory and use Lakatos’ theory as a restraint so that Feyerabend’s theory does not appear too idealistic and extreme. First, I will maintain Feyerabend’s theory as a whole, but introduce Lakatos’ research program. This will alleviate the complaint that Feyerabend does not provide any explanation of the scientific process. However, unlike Lakatos, we are not introducing a research program in order to pursue “rational change.” Here, the research program is merely a means of explaining phenomena and providing another source of reference for scientists when choosing between competing theories. The performance of this research program may explain which program scientists prefer. Furthermore, using the research program, we question the implication of Feyerabend’s claim that there is no rational basis for saying that science is superior to other disciplines and that it should therefore be treated equally with all other systems of knowledge. According to Feyerabend, each discipline must have a well-defined purpose and a means of achieving that purpose, and according to Lakatos’s research program, in order to be a “scientific” research program, it must be consistent enough to derive a program and at least occasionally derive new facts. Based on this, it cannot be said that science is superior to other disciplines, and therefore it is not inherently natural to give scientific knowledge more authority than the knowledge of other disciplines. Before hastily judging other disciplines to be inferior to science, we must first explore the field, find its goals and methods for achieving them, and find the facts that are being derived from the research program in that field. At this point, fields classified as “pseudoscience” will lack one or more of the following: a well-defined purpose, a method for achieving that purpose, consistency, or the derivation of new facts. If the derivation of new facts is lacking, in other words, if the research program is extremely regressive, it cannot be said that choosing this research program is unreasonable. This is because even a regressive program may lead to new discoveries at the end of the research. However, in the case of extremely regressive programs, fewer people will choose them, and similarly, research support and publication of papers will also decrease. Furthermore, the fact that a program is regressive will greatly undermine efforts to promote it and make it easy for other programs to gain popularity. This may explain why society and other individuals are not interested in or avoid that field. In other words, even if it is not unreasonable for an individual to study astrology, the fact that society does not place much value or authority on astrology can be explained by this. In other words, “pseudoscience” is generally explained as being determined by social/historical reasons rather than by objective criteria.
Ultimately, the core of Feyerabend’s argument is that science should be given more freedom, and society should also be given freedom from science, so that members can learn about each other’s fields of study, evaluate other fields, and systematically learn about other fields before using them, in order to understand their goals and methods. Here, scientists should find methods appropriate to the situation in science based on their experience, knowledge of other fields, and personal judgment. However, unlike Feyerabend, who accepts pseudoscience, Lakatos introduced his research program to provide materials and criteria to help exclude pseudoscience and select theories. This does not completely exclude pseudoscience. This is because it is difficult to judge whether a scientist’s choices in research are rational or irrational. However, this seems to be helpful in selecting programs that are more likely to provide beneficial results. In modern society, there are frequent attempts at “convergence” and “integration” between academic disciplines. Subjectivity can also be involved in science, and a free and open epistemology/approach that is not bound by methods can reduce the sense of familiarity between science and other disciplines and the awe of science, thereby facilitating communication/convergence between science and other disciplines.
