Paradigm and Paradigm Shift in the Philosophy of Science

Thomas Kuhn, in The Structure of Scientific Revolutions (1962), defined a paradigm as "the basic framework of assumptions, principles, and methods within which the scientific community operates." Paradigm sets the model of setting problems and typical solutions (so-called puzzles) for "normal science". During the period of normal science, scientists focus on solving problems within the framework of the accepted paradigm and usually do not question its fundamental postulates.

• Within the framework of normal science, anomalies accumulate - contradictions or unresolved problems that do not fit into the existing paradigm. Kuhn noted that "all paradigms encounter anomalies" that are ignored or explained in "sanitary" ways.
• When critical anomalies grow and are not resolved, a paradigm crisis arises. At this point, alternative hypotheses or models emerge. If a new theory is more convincing than the previous one, there is a revolution - a paradigm shift. Kuhn describes it as a process in which the old world order is no longer in use, giving way to a new one: “the community of scientists begins to lose confidence in the paradigm, a crisis comes ... At the same time, an alternative paradigm may appear, and if it conquers the community, a paradigm shift occurs — that is, a revolution.”
• After the revolution, the scientific community is organized around a new paradigm: the criteria for the significance of problems, methods and goals of research are changing. Kuhn emphasizes that a paradigm shift means a realignment of scientists' worldview (the so-called Gestalt switch). The new paradigm gives rise to a different “conceptual landscape” and with it the observed facts themselves change. Thus, he described the transition from Newtonian mechanics to Einsteinian mechanics as “a shift in the conceptual network through which scientists look at the world.”
• Kuhn’s concept of non-commensurability was significant: in his opinion, paradigms before and after a revolution often “do not have a common basis for comparison.” This means that the same set of empirical data can be interpreted differently, and proponents of different paradigms actually speak “different languages.” Kuhn insisted that after the revolution a new system of "concepts, methods and metaphysical assumptions" arose, incompatible with the old one.

Thus, according to Kuhn, science develops “jumpingly”: periods of normal science and revolutions alternate, when one paradigm is replaced by another.

2. Modern interpretations and criticisms of Kuhn's theory

Kuhn’s concept has caused a wide resonance and was actively discussed by critics. First, it was noted that the term "paradigm" he has multiple meanings (e.g., Masterman found more than twenty different uses in the "Structure"). Secondly, the conservatives of the scientific method criticized Kuhn for the radical nature of the conclusions. Thus, logical positivists and critics of the Popper type perceived the idea of noncommensurability as threatening relativism and supposedly irrational science. Kuhn did argue that it is impossible to say unequivocally that a later theory is "better approximated to the truth." This has given rise to accusations that scientific progress is turning into a wandering about the accepted paradigm.

In response, Kuhn refined and softened some of the wording. He stressed that the paradigm shift is still a scientific step forward and the "evolution" of knowledge, although not linearly striving for absolute truth. Kuhn himself wrote that paradigm shifts occur as a result of overcoming the limitations of old science, and even during the revolution continuity of many tasks remains. According to modern commentators, Kuhn "was shocked by the criticism", and then began to clarify his theses. Today, many philosophers note that Kuhn did not claim complete skepticism, but only pointed out the difficulty of comparing different worldviews.

In addition to criticism, alternative interpretations have appeared. Okay, E. Lakatos proposed to represent science in terms of scientific programs (with increasing or regressive modifications), weakening the binomial "revolution / normal science". L. Laudan developed the idea of research traditions by emphasizing rational "software heuristics" rather than sudden "revolutions". In the XXI century, philosophers also emphasize other aspects: for example, p. Hoyningen-Hyune treats Kuhn in a neo-Kantian way, where a paradigm shift changes the phenomenal world of the scientist, rather than some external reality. Others, inspired by Wittgenstein, view paradigms as families of similar examples or prototypes. Finally, Kunovsky’s emphasis on “examples” was developed by G. Margolis: Paradigm is not so much the rules as a model of successful research, forming the “habits of thinking” of scientists.

Thus, modern interpretations tend to suggest that Kuhn described an important socio-cognitive phenomenon (scientific consensus), but the initial strict consequences of his model (extreme relativism and irrationalism) were clamped down. The Kuhn model of paradigms became more flexible and was subjected to "constant descriptive refinement."

3. Development of ideas about paradigm shifts in XXI century

At the beginning of the XXI century, the philosophy of science is increasingly moving away from the idea of multiple one-time revolutions. There has been increased attention to the evolutionary evolution of science, similar to biology. Kuhn himself noted: "It was previously thought meaningless to say that a late theory is closer to the truth, but nevertheless he recognized scientific progress comparable to biological evolution." In other words, science is now often seen as a constantly expanding and complex system of knowledge, in which old ideas are transformed, and not repeatedly completely destroyed. Some philosophers even speak of a “flow” of scientific discoveries, not of separate “acts” of the revolution.

In addition, in the XXI century increased cognitive approach to science. We see the paradigm as a set of mental models and mental circuits in the minds of researchers. For example, H. Andersen et al. show that scientists store prototypical examples of successful research that set the framework for their thinking. This explains why communication skills within the scientific community — and not just formal logic — are important when changing paradigms. In this perspective, the paradigm is understood through the concept of "example", and not through a fully formalized set of rules.

Finally, scientists are increasingly focusing on the social and institutional drivers of paradigms. Philosophers of science of the XXI century take into account the multiplicity of scientific schools, the network structure of the modern scientific community (large projects, collaborations) and the role of technologies for the dissemination of knowledge. As a result, the idea of the paradigm has become more contextual and emergent: changes in approaches are recognized as a complex interaction of ideas, tools and institutions. These trends bring us closer to evolutionary or complex models of science, where a revolutionary "one-off" shift is complemented by a slow accumulation of changes and the related development of several directions.

4. Examples of Paradigm Shifts in Different Disciplines

• Physics. A classic example is the transition from Newtonian to relativistic mechanics in the early twentieth century. Kuhn wrote that this transition “illustrates particularly clearly a scientific revolution as a shift in the conceptual network through which scientists view the world”. In effect, it changed fundamental conceptions of space, time and mass: it was precisely a paradigm shift. Another example is the quantum revolution in the early twentieth century, when the emergence of quantum mechanics radically changed ideas about matter and energy.
• Biology. In biology, the principal paradigm shift is associated with the theory of evolution. The transition from a static view of species (Lamarckism) to Darwin’s evolutionary model marked a fundamental shift in the understanding of life. Kuhn mentions how Darwin “replaced Linnaeus’s static tree with a classification based on evolutionary kinship”. Later, in the twentieth century, Watson and Crick’s discovery of the structure of DNA (1953) began the “molecular revolution”: the merging of genetics, chemistry and biology into a new discipline. As contemporary reviews observe, “there is something paradigmatic and something revolutionary in molecular biology”: its rapid expansion and reorientation towards biochemical methods “came into conflict” with traditional approaches in evolutionary biology, although the Darwinian picture broadly survived.
• Cognitive science. An example of a paradigm shift here is the so-called cognitive revolution of the mid-twentieth century. For a long time, psychology and related disciplines were dominated by behaviourism: the study of behaviour without reference to internal processes. Beginning with the work of Noam Chomsky in the 1950s–60s and that of others, there was a fundamental reorientation towards studying the “black box” of consciousness, thought and perception. As noted on the websites, “the movement known as the cognitive revolution moved away from behaviourist approaches and began to regard cognition as central to psychology”. This shift gave rise to entirely new methods, such as memory experiments and neuroscientific research, and changed the status of language, consciousness and perception in science.
• Other (examples). Similar paradigm transitions are noted in other areas: for example, in chemistry, the transition from the classical scheme of the Mendeleev table to the quantum-chemical understanding of connections or in climatology - from individual explanations of weather to modern models of global change. In cognitive neuroscience, one can mention the shift to neuroimaging and exploring the brain as a computing machine. Each of these examples demonstrates a typical pattern: a radical reorganization of concepts, methods, and priority issues of discipline.

5. Interdisciplinarity, AI and Big Data as factors of new paradigms

New technological and organizational trends in science stimulate the emergence of "post-Kun" paradigms.

• Interdisciplinary. In recent decades, it is widely believed that the integration of different sciences - from bioinformatics to neuroeconomics - changes the architecture of scientific knowledge. Modern interdisciplinary initiatives are often described as challenging traditional disciplines and a kind of "scientific revolution" (in the spirit of Kuhn). However, as indicated by B. Politi, the very notion of interdisciplinarity, is now in a "preparadigmatic" phase: there is no single agreed definition, and there is no clear common paradigm within such projects. However, the synthesis of approaches (e.g., physics and biology in systems biology, computer science, and psychology in cognitive neuroscience) does form new paradigm complexes. These complexes are characterized by new methods (computational modeling, machine learning) and new issues (for example, modeling of complex systems) that go beyond traditional disciplines.
• Artificial Intelligence (AI). Emergence of modern technologies AI – Especially large neural network models are also considered as a source of a new paradigm stage. According to V. Dhara, "in 2023 St Petersburg. AI attracted the attention of the whole world with the appearance of pre-acquired models (such as GPT-3)This new opportunity has created a paradigm shift in AI, Turning it from an application problem into a universal technology that is customizable for all applications. According to Kuhn, the appearance of such powerful tools is a signal of the onset of a new period of normal science AI, where the focus of research will be on unlocking the possibilities of these models and overcoming new anomalies.
• Big Data (Big Data). Finally, the era of huge amounts of data gave rise to the idea of the so-called “fourth paradigm” of science. Rob Keechin points out that Big Data creates a fundamentally new epistemological approach: now scientists are not just testing hypotheses on a sample of data, but are trying to "extract insights directly from the data." Many researchers believe that this is a kind of data revolution with "long-range consequences for the production of knowledge." In this context, interdisciplinary collaboration (e.g., statisticians, computer scientists and subject matter experts) and new computing platforms are becoming important.

Taken together, these factors (the integration of disciplines, AI, Big Data, etc.) create an environment where new paradigms can be formed: they rely on flexible networks of knowledge and modern tools, and not only on classical theories. The researchers note that the development of such paradigms will be determined by the joint activities of different scientific communities and technologies, which generally continues the ideological line of Kuhn that science moves as a result of changing the “rules of the game” throughout the community.

Sources: original works and modern research of philosophers of science. In particular, Kuhn's key ideas are set out in his "Structure of Scientific Revolutions", and modern interpretations and examples are taken from the writings of contemporary authors and academic reviews.