A collection of essays tracing the development and consequences of Kuhn’s own revolutionary philosophy of science
Preface: The Route from Structure
We begin in the Preface, where Kuhn establishes the occasion for this volume and orients the reader to the trajectory of his thinking since 1962. The book collects essays written between 1969 and 1987, following the publication of *The Structure of Scientific Revolutions*. Kuhn tells us he has chosen “the road since *Structure*” as the title because it captures both the continuity and the departures in his thought. The essays are arranged chronologically to show the evolution of his views, though he acknowledges that they do not form a single, unified argument.
Kuhn immediately confronts a central problem: many readers of *Structure* misunderstood his core concepts, particularly paradigm and incommensurability. He writes that critics often took him to be advocating for a form of relativism or irrationalism, when his actual aim was to describe how science actually operates. The Preface serves as a kind of apologia, but not a retraction. Kuhn insists that the core of his account remains intact: normal science operates within a paradigm; anomalies accumulate and precipitate crisis; revolutionary science introduces a new paradigm that is incompatible with the old one in fundamental ways.
What Kuhn clarifies here is that his interest has shifted over time. The early work focused on the sociological and psychological dimensions of scientific communities. The later work, represented in these essays, turns increasingly to questions of language, meaning, and the philosophy of mind. Kuhn speaks of a “linguistic turn” in his own thinking, and he previews the central claim that will animate many of the essays: that scientific revolutions involve changes in the lexicon or taxonomic structure available to a scientific community.
“The essays that follow represent the gradual crystallization of a point of view that, for me, has replaced the earlier one.”
Kuhn also explains the principle of selection. He has included previously published essays alongside one new piece, and he has avoided extensive revision of the older material, preferring to let readers see the development as it happened. He warns that the essays do not always agree with one another, and that earlier formulations sometimes give way to later, more precise ones. This is not a textbook but a record of inquiry.
The Preface thus establishes the work’s dual character: it is both a retrospective clarification of *Structure* and a forward-looking exploration of new philosophical ground. The reader is meant to follow Kuhn as he rethinks his own position, not to receive a settled doctrine. The stakes are high: understanding how science changes is understanding how rational thought itself changes over time.
Chapter 1: What Are Scientific Revolutions?
This chapter was first delivered as a lecture in 1981, and it represents Kuhn’s mature attempt to answer the question that had haunted his career. He begins by rejecting the image of science as a cumulative enterprise that simply adds new truths to an existing stock. Instead, he insists that scientific revolutions involve fundamental changes in the way scientists see the world.
Kuhn locates us in the aftermath of *Structure*’s publication, where he had introduced the concept of paradigm as the shared set of assumptions, methods, and exemplars that define a scientific community. But he now refines that concept into something more precise: a paradigm, he says, is primarily a disciplinary matrix — a constellation of commitments that includes symbolic generalizations, models, and exemplars. A scientific revolution, then, is a change in the disciplinary matrix, and this change is not merely additive but transformative.
The key argument of the chapter concerns incommensurability, which Kuhn now frames in terms of translation failure. When two paradigms compete, he argues, there is no neutral language into which both can be fully translated. The terms central to one paradigm cannot be mapped onto the terms of the other without loss or distortion. Kuhn offers the example of the Copernican revolution: the term “planet” meant something different to Ptolemaic astronomers (who included the sun and moon as planets) than it did to Copernicans. This is not a trivial terminological difference but a symptom of a deeper conceptual reorganization.
“The proponents of competing paradigms practice their trades in different worlds.”
Kuhn spends considerable time clarifying what he does *not* mean by incommensurability. He is not claiming that communication is impossible, only that it is difficult and incomplete. Scientists in different paradigms can communicate, but they must engage in a process of interpretation rather than simple translation. They become like travelers in a foreign culture, learning the meanings of terms by seeing how they are used in practice.
The chapter also introduces a distinction that will become central: between the textbook tradition, which presents science as a cumulative march toward truth, and the historical record, which shows science as a series of discontinuous transformations. Kuhn argues that textbooks systematically misrepresent the past by rewriting history to make the current paradigm appear inevitable. This is not a conspiracy; it is a pedagogical necessity. But it obscures the actual nature of scientific change.
What matters here is that Kuhn is moving the discussion away from the psychology of individual scientists and toward the structure of language and practice. Revolutions are not just changes in what scientists believe; they are changes in the very categories through which experience is organized. The reader begins to see why Kuhn’s later work takes a linguistic turn: incommensurability is fundamentally a problem about meaning, and meaning is embedded in the shared lexicon of a community.
Chapter 2: The Road since Structure
This essay, which gives the volume its title, was written in 1987 as a retrospective assessment of the twenty-five years since *Structure* appeared. Kuhn looks back at the reception of his work and identifies the central misunderstandings that have plagued it. The most persistent, he says, is the charge of relativism: the idea that if paradigms are incommensurable, then there is no rational basis for choosing between them.
Kuhn answers this charge directly. He argues that relativism presupposes a standpoint external to all paradigms from which they can be compared, and that no such standpoint exists. But this does not mean that paradigm choice is arbitrary. Scientists choose between paradigms based on values like accuracy, consistency, scope, simplicity, and fruitfulness. These values do not dictate a unique decision — different scientists may weigh them differently — but they provide a rational basis for debate. The mistake, Kuhn says, is to confuse the absence of algorithmic rules with the absence of rationality.
The essay then turns to the evolution of Kuhn’s own thinking about incommensurability. He admits that his earlier formulations were too vague, and that critics were right to press for clarification. He now offers a more precise account: incommensurability arises from the fact that the taxonomic categories of two paradigms do not map onto each other. When a scientific revolution occurs, the basic kinds into which the world is divided change. New categories appear; old ones disappear; and existing categories are redefined. This is not merely a conceptual change but a change in the lexical structure of science.
“What I am now calling lexical structures are the constitutive categories of the worldview within which normal science is practiced.”
Kuhn introduces the idea of a lexicon as a structured set of kind-terms that are learned together and cannot be defined independently. Children learning their first language acquire such a lexicon, and scientists learning a new paradigm must undergo a similar process. The lexicon determines what can be said, what can be asked, and what can be observed. A revolution changes the lexicon, and with it the range of possible scientific statements.
This chapter also contains Kuhn’s most explicit engagement with the philosophy of language. He draws on the work of W.V.O. Quine and Hilary Putnam, particularly Quine’s thesis of the indeterminacy of translation and Putnam’s arguments about the division of linguistic labor. But Kuhn distances himself from Quine’s behaviorism and from any form of linguistic idealism. He insists that the world exists independently of our categories, but that we have no access to it except through those categories.
The significance of this chapter lies in its consolidation. Kuhn is no longer merely defending *Structure*; he is building a new philosophical position. The road since *Structure* has led to a theory of meaning, a theory of scientific rationality, and a theory of conceptual change — all tied together by the notion of a lexicon. The reader is now prepared to see how these ideas will be developed in the essays that follow.
Chapter 3: Commensurability, Comparability, Communicability
This essay was originally published in 1982, and it takes up the most contentious concept in Kuhn’s philosophy — incommensurability — and subjects it to systematic analysis. Kuhn opens by noting that his critics have consistently misunderstood him: they think he denies the possibility of comparing competing theories, when in fact he only denies the possibility of a neutral, theory-independent language of comparison.
Kuhn distinguishes three senses of commensurability. Two theories are metrically commensurable if their claims can be expressed in the same units of measurement; they are taxonomically commensurable if they share the same system of categories; and they are translationally commensurable if the statements of one can be rendered without loss into the language of the other. His claim, he says, is only about the last two: competing paradigms are not fully translatable into one another, and their taxonomic categories do not overlap.
The central argument of the essay turns on the concept of meaning change. Kuhn argues that when a scientific revolution occurs, terms like “mass,” “force,” “element,” and “species” change their meanings in ways that are not captured by the standard account of definitional change. This is not just a matter of adding new criteria; the pattern of family resemblances that connects instances of the term is altered. A term like “mass” in Newtonian physics is not the same concept as “mass” in Einsteinian physics, even though they share a history.
“The members of a scientific community learn to identify the same objects, situations, and processes by applying criteria that are not fully reducible to rules.”
Kuhn illustrates this with the example of phlogiston theory. Phlogiston theorists and oxygen theorists did not simply disagree about the composition of air; they employed different systems of categories for classifying substances. The phlogiston theorist’s category of “dephlogisticated air” does not correspond to any single category in the oxygen framework. Attempts to translate between the two systems inevitably fail because the ontological commitments are incompatible.
The essay also addresses the worry that incommensurability leads to non-rationality in theory choice. Kuhn’s response is careful: he argues that theory choice is rational but not rule-governed. Scientists do not apply algorithms; they make judgments based on values that are themselves shaped by the paradigm within which they work. This does not make science irrational; it makes it human. Rationality, Kuhn suggests, is not a matter of following formal rules but of exercising good judgment within a community of inquiry.
What the reader gains from this chapter is a deeper understanding of Kuhn’s positive project. He is not a relativist; he is a pluralist about rationality. And he is not a skeptic about scientific progress; he simply denies that progress consists in approaching a fixed, final truth. The incommensurability thesis, properly understood, is a claim about language and categorization, not an invitation to irrationalism.
Chapter 4: The Natural and the Human Sciences
LOCATE: This chapter positions Kuhn’s analysis of scientific development within the larger debate about the distinction between the natural sciences and the human sciences (Geisteswissenschaften), responding to critics who have misunderstood his project.
PRESENT: Kuhn opens by acknowledging that his work has been appropriated by those who wish to blur the line between the natural and human sciences, or even to deny the special epistemic authority of the natural sciences. He finds this appropriation “largely a mistake.” The question he addresses is straightforward: is there a fundamental methodological or epistemological divide between the natural sciences and fields like history, literary criticism, or sociology?
Kuhn’s answer is nuanced. He argues that the differences between natural and human sciences are real but have been misdescribed. The traditional distinction, going back to Dilthey, holds that the natural sciences seek explanation (Erklären) through subsumption under general laws, while the human sciences seek understanding (Verstehen) through interpretation of meaning. Kuhn rejects this sharp dichotomy. He points out that the natural sciences themselves require interpretation—most notably in the sort of hermeneutic work needed to recognize and articulate anomalies that lead to crisis. “The practice of normal science,” he writes, “depends on the ability, acquired from exemplars, to group objects and situations into similarity sets which are primitive in the sense that the grouping can be done without answers to the question ‘similar with respect to what?'”
The crucial difference, Kuhn argues, lies not in the presence or absence of interpretation but in the consensus that characterizes mature natural sciences. Natural scientific communities display a degree of agreement about fundamentals—about what problems are worth solving, what methods are appropriate, and what counts as a solution—that is absent from the human sciences. This consensus is not imposed by logic or evidence alone; it is a sociological fact about the structure of the community. But it has profound consequences for the rate and direction of cognitive development.
“I now want to suggest that the sciences are more like the arts than I once believed, but I also want to argue that there is a sense in which they progress in a way that the arts do not.”
Kuhn then examines what it means for a field to be “scientific.” He rejects the idea that science is defined by its method, whether that method is falsificationism, verificationism, or some other algorithmic procedure. Instead, he offers a functional definition: a field is scientific to the extent that its practitioners share a sufficiently robust set of exemplars and commitments to allow for the puzzle-solving activity of normal science. The human sciences, by contrast, remain in a state of pre-paradigm debate, where foundational questions are perpetually reopened and no single set of commitments commands universal assent.
This does not mean the human sciences are worthless. Kuhn insists they produce genuine knowledge of a different kind. But he resists the relativist conclusion that all knowledge-producing enterprises are epistemologically equivalent. The natural sciences have achieved a form of cumulative progress—what he calls a “evolutionary advance”—that the human sciences have not. This progress is not toward a final truth but is nevertheless real in the sense that later scientific theories solve more problems, with greater precision, than their predecessors.
CLARIFY: What Kuhn offers here is not a methodological hierarchy but a sociological and historical typology. The natural sciences are distinguished by their community structure, not by their adherence to a special method. This allows him to account for the obvious cognitive successes of natural science without falling into positivism. At the same time, it blocks the easy relativist move that would collapse all inquiry into mere interpretation. The human sciences have their own integrity; they are simply organized differently, and that difference has consequences for how they develop over time.
Chapter 5: Afterwords
LOCATE: This chapter serves as Kuhn’s most systematic response to his critics, written for a 1990 volume of critical essays, where he revisits the central claims of *The Structure of Scientific Revolutions* and clarifies what he did and did not intend.
PRESENT: Kuhn begins by acknowledging that many misunderstandings of *Structure* arose from ambiguities in the original text. He takes responsibility for these, noting that he wrote the book in a “white heat” and sometimes expressed himself imprecisely. The terms paradigm and incommensurability, in particular, have been sources of endless confusion. He now wishes to clarify both.
On paradigms, Kuhn distinguishes two senses that were conflated in the original book. In the first, broader sense, a paradigm is a disciplinary matrix—the entire constellation of shared commitments that bind a scientific community together: symbolic generalizations, models, values, and exemplars. In the narrower, deeper sense, a paradigm is an exemplar—a concrete problem solution that serves as a model for further research. It is this second sense that Kuhn now regards as more fundamental. The exemplar is the unit of scientific education: students learn to see new problems as similar to paradigmatic solved problems, and this learned similarity relation is what gives normal science its direction and coherence.
On incommensurability, Kuhn is equally explicit. He never meant that successive theories are incomparable in all respects. They are comparable along many dimensions: they can be compared for empirical adequacy, for internal consistency, for scope, for simplicity, and for fruitfulness. The incommensurability is local and taxonomic. It arises because the basic categories of one theory cannot be mapped onto those of another in a straightforward, one-to-one way. When a scientific revolution occurs, some terms change their meaning because the taxonomy of natural kinds itself changes. For instance, after Einstein, the term “mass” does not mean the same thing it meant in Newtonian physics, because the theoretical relationships in which it participates have been fundamentally altered.
“Incommensurability thus becomes a sort of untranslatability, localized to the area in which taxonomic change occurred.”
Kuhn then addresses the charge of relativism. He insists he is not a relativist about truth or rationality. A later theory is always better than its predecessor in solving the problems that motivated the earlier theory. The sense in which scientific progress is real is that later theories provide more precise predictions, broader scope, and more effective puzzle-solving tools. But this progress is not toward a goal fixed by nature or logic. It is an evolutionary process in which the criteria of evaluation themselves evolve along with the theories.
He also clarifies his relationship to the sociology of scientific knowledge. He welcomes sociological studies of science but insists that his own project is primarily historical and philosophical. The sociology of science often treats scientific knowledge as a social construction in a way that denies its cognitive authority. Kuhn rejects this: “I am not a social constructivist.” The natural world imposes constraints on what theories can survive; not any belief can become a paradigm.
CLARIFY: This chapter is Kuhn’s most mature and careful statement of his position. He is trying to walk a narrow path between positivism and relativism. The key move is his insistence that the incommensurability of theories does not entail their incomparability. Theories can be rationally compared, but the comparison is not algorithmic; it involves judgment, persuasion, and the application of shared values. The result is a picture of science as a rational enterprise that is also deeply historical and social—a picture that avoids the Scylla of logical empiricism and the Charybdis of social constructivism.
Chapter 6: The Trouble with the Historical Philosophy of Science
LOCATE: This chapter, originally the 1991 Taft Memorial Lecture at the University of Cincinnati, finds Kuhn reflecting on the reception and misreading of his work, specifically the tension between his historical approach and the philosophical expectations it challenged.
PRESENT: Kuhn opens by acknowledging that his work has been “widely misunderstood” and that the very term “Kuhn’s historical philosophy of science” is itself a source of trouble. He identifies the core problem as stemming from the fact that his approach was “deeply subversive” of the received view of science, and that many readers wanted to find a relativist or irrationalist position where none existed. The trouble, he explains, arises from the collision between two different conceptions of what philosophy of science should be. The traditional view, descending from logical positivism, sees philosophy as providing a “rational reconstruction” of scientific knowledge, one that is normative and ahistorical. Kuhn’s historical approach, by contrast, treats science as a “human enterprise” that must be understood in its concrete historical development.
Kuhn then lays out the central difficulty: “How can history of science be a philosophical discipline?” He argues that the historian of science cannot simply apply philosophical categories to past science; rather, the historian must enter into the thought world of the past, understanding scientific ideas on their own terms. This creates a fundamental tension. The philosopher wants to judge past science by present standards, to separate good science from bad. The historian wants to understand past science as a coherent intellectual enterprise, which means suspending judgment about its truth or falsity. Kuhn points out that this tension is not accidental but essential: “The historian of science must be able to see old science as it was seen by its practitioners.” This does not mean abandoning standards of rationality, but it does mean recognizing that those standards have themselves changed over time.
A key example Kuhn uses is the transition from Ptolemaic to Copernican astronomy. The standard philosophical account treats this as a case where scientists rejected a false theory in favor of a truer one. The historical account, however, shows that the Ptolemaic system was not simply a “false” theory that was replaced by a “true” one; it was a sophisticated and successful research program that had solved many problems. The Copernican revolution, Kuhn insists, was not a simple matter of evidence trumping error. Rather, it was a “gestalt switch” in which the entire framework of astronomical explanation was transformed. This transformation cannot be captured by the standard philosophical categories of confirmation and falsification. This leads Kuhn to his central claim: the problem with the historical philosophy of science is that it cannot be assimilated to the traditional philosophical project of providing a “logic of scientific discovery” or a “rational reconstruction” of scientific method.
“The historical philosophy of science is a product of the last generation, and it has yet to find a secure place within philosophy. Its practitioners are still trying to justify its existence and to clarify its methods.”
CLARIFY: This chapter is both a defense and a diagnosis. Kuhn is defending the legitimacy of the historical approach while also diagnosing why it has been so controversial. The trouble is not that the historical approach is wrong, but that it challenges deeply held philosophical assumptions about the nature of rationality and the possibility of ahistorical standards of judgment. Kuhn is insisting that the philosophy of science must be naturalized—it must take account of the actual historical development of science—without being reduced to a purely descriptive enterprise. The historical approach, in his view, offers the only way to understand science as a rational activity that is also a human and historical one.
Chapter 7: The Road since Structure (Part II: The Nature of Scientific Development)
LOCATE: In the latter portion of this previously discussed chapter, Kuhn deepens his account of scientific development by introducing the concept of “speciation” as a biological metaphor for the emergence of new fields and specialities.
PRESENT: Kuhn argues that the development of scientific knowledge is not like the cumulative growth of a single tree; rather, it is more like the evolutionary process of speciation, where new species branch off from existing ones. He writes, “Scientific development, like biological evolution, is unidirectional and irreversible. One theory is not simply replaced by another; instead, the scientific community splits, with some scientists pursuing the new theory and others remaining with the old.” This process is not a matter of one theory being proven false and another true; it is a matter of the scientific community reorganizing itself around new problems and new standards.
Kuhn elaborates this idea by describing how scientific revolutions often create new specialties rather than simply replacing old ones. For example, the Copernican revolution did not simply eliminate Ptolemaic astronomy; it created a new branch of astronomy that eventually became modern astrophysics. Ptolemaic astronomy, as a practical system for navigation and calendar calculation, continued to be used long after Copernicus’s work. Similarly, the Newtonian revolution did not simply replace Aristotelian physics; it created a new discipline—classical mechanics—that coexisted with other forms of natural philosophy for generations. This process, Kuhn argues, is not a sign of irrationality but of the complex social structure of science. Scientists are trained into communities, and those communities have their own standards, problems, and instruments. A revolution changes those standards, but it does not necessarily eliminate the older community’s work; it may simply render it irrelevant to the new research program.
Kuhn then introduces a crucial distinction between “normal science” and “extraordinary science.” Normal science is the puzzle-solving activity that occurs within a paradigm; it is cumulative, incremental, and directed by the paradigm’s rules. Extraordinary science occurs when a paradigm is in crisis, when the puzzle-solving machinery breaks down. During extraordinary science, scientists begin to question fundamental assumptions and to experiment with new approaches. This is the period of “revolutionary science” that Kuhn described in *Structure*. However, Kuhn now adds that the transition from extraordinary to normal science is not simply a matter of one paradigm replacing another; it is a matter of the scientific community reorganizing itself. The new paradigm creates a new community, with its own journals, conferences, and training programs. The old paradigm may continue to exist, but it is no longer at the center of the scientific enterprise.
“The outcome of a scientific revolution is not a new version of the old science, but a new science, one that is incommensurable with its predecessor.”
CLARIFY: This chapter represents Kuhn’s mature thinking about the dynamics of scientific change. The speciation metaphor allows him to account for the fact that scientific development is not a simple linear progression but a branching process in which old theories can persist alongside new ones. This is not relativism; it is a more accurate description of how science actually develops. The key insight is that scientific revolutions are not just changes in theory; they are changes in the structure of the scientific community itself. The result is a picture of science as a deeply social enterprise that is also rational and progressive.
Chapter 8: The Problem of the History of Science
LOCATE: In this chapter, originally a lecture delivered at the University of California, Berkeley, in 1979, Kuhn addresses the methodological problem of how to write the history of science without imposing present-day categories on past science.
PRESENT: Kuhn begins by acknowledging that the history of science is a paradoxical enterprise. The historian tries to understand past science on its own terms, but the very tools the historian uses—the concepts, categories, and methods of modern science—are themselves products of the history being studied. This creates a hermeneutic circle: the historian must use modern categories to understand past science, but those categories may distort what they are trying to understand. Kuhn proposes a solution: the historian must engage in a kind of “bilateral hermeneutics” in which she moves back and forth between her own categories and those of the past. This is not a one-way process; the historian’s own categories are transformed by the encounter with the past.
Kuhn then describes the practical difficulties this creates. The historian of science cannot simply read past scientific texts as if they were addressing modern problems. The historian must learn the language, the instruments, and the social context of the past science. This means reading not only the canonical texts but also the textbooks, the laboratory notebooks, and the correspondence of past scientists. It also means understanding the non-scientific factors—the religious, philosophical, and political contexts—that shaped scientific practice. Kuhn gives the example of the seventeenth-century debate about the vacuum. To understand why this debate was so intense, one must understand the theological implications of the concept of a void, which challenged the Aristotelian doctrine that nature abhors a vacuum. This is not a matter of irrationality; it is a matter of seeing the debate in its full historical context.
However, Kuhn warns against the temptation of relativism. The historian’s goal is not to judge that past science was “as good as” modern science, but to understand why past scientists believed what they did. This requires a kind of “historical empathy” that is disciplined by evidence. The historian must be able to enter into the thought world of the past while also maintaining critical distance. This is not easy, and Kuhn admits that it is a skill that is learned through practice. The goal is not to produce a “neutral” history but a history that is faithful to the evidence while also being accessible to modern readers.
“The historian of science must be bilingual: he must be able to speak the language of the past and the language of the present, and to translate between them without reducing one to the other.”
CLARIFY: This chapter is a methodological manual for the historian of science. Kuhn is insisting that the history of science is a genuine discipline with its own methods and standards. It is not simply the application of philosophical or sociological theories to past science. The historian must engage in a kind of dialogical encounter with the past, one that transforms the historian’s own understanding. This is not relativism; it is the recognition that understanding the past requires a kind of imaginative reconstruction that is both disciplined and creative. The result is a picture of the history of science as a “human science” in the sense that it requires the same skills of interpretation and empathy that are needed to understand any human activity.
Chapter 9: The History of Science as an Independent Discipline
We are in the latter portion of Kuhn’s collection, where he turns to methodological questions about the discipline of history of science itself. This chapter presents Kuhn’s arguments for the autonomy and distinctive character of the history of science as a scholarly practice.
Kuhn begins by asserting that the history of science is not merely a subfield of history or a handmaiden to philosophy of science, but an independent discipline with its own methodology and standards. He traces the origins of this discipline to the mid-twentieth century, when a generation of scholars began to approach scientific texts from the past with a new kind of attention. These scholars, including Kuhn himself, rejected the Whig history approach that read the past solely as a prelude to the present.
The central methodological principle Kuhn articulates is that the historian must strive to see the past through the eyes of the actors involved. This requires a suspension of present-day scientific knowledge, or at least a disciplined awareness of how that knowledge can distort historical understanding. Kuhn writes that the historian must attempt to become “bilingual” in the language of the period under study and the language of modern science, moving between them with care and attention to the differences.
Kuhn then addresses the relationship between internalist and externalist approaches to the history of science. Internalist histories focus on the conceptual and technical content of science itself, while externalist histories examine the social, cultural, and institutional contexts in which science develops. Kuhn argues that these approaches are not competitors but mutually necessary. The most illuminating historical work integrates both perspectives, showing how conceptual developments are shaped by and in turn shape their social settings.
This chapter matters because it crystallizes Kuhn’s vision of what the history of science should be. It is not a mere chronicle of discoveries or a repository of case studies for philosophers. It is a disciplined inquiry into how human beings have come to understand the natural world, an inquiry that requires its own methods and standards. The historian of science, in Kuhn’s view, is not a philosopher who happens to work on historical examples, but a practitioner of a distinct craft with its own integrity.
Chapter 10: The Road since Structure (Part III: Rationality and Theory Choice)
We now move to what is essentially the third major discussion of themes from *Structure*, focusing on the problem of rationality in scientific development and the process of theory choice. Kuhn is responding to critics who accused him of making science irrational or merely a matter of mob psychology.
Kuhn begins by clarifying that he never denied the role of rationality in science. What he challenged was a particular, narrow conception of rationality that required a logical algorithm for theory choice. Such an algorithm would specify rules that, if followed mechanically, would lead any rational agent to the same decision. Kuhn argues that no such algorithm exists or could exist, because theory choice involves values that cannot be reduced to a simple formula.
The values that guide theory choice include accuracy, consistency, scope, simplicity, and fruitfulness. These values are shared by the scientific community and play a crucial role in decisions between competing theories. However, they are ambiguous in their application. Different scientists may interpret the same value differently, or weight the values differently in a particular case. One scientist may prioritize simplicity, another may prioritize accuracy. This does not make their decisions irrational, because both are applying the same set of shared values, just in different ways.
Kuhn then introduces the concept of incommensurability as it relates to theory choice. Incommensurability does not mean that theories cannot be compared at all. It means that there is no common measure, no neutral language or set of criteria that can decide between them in a mechanical way. The comparison must be holistic and contextual. Scientists must weigh the strengths and weaknesses of competing theories across multiple dimensions, and reasonable people may disagree about the outcome.
This chapter is crucial because it presents Kuhn’s positive account of scientific rationality. Science is rational not because it follows a fixed method but because it is governed by shared values that are applied through deliberation and persuasion within a community. The process of theory choice is more like a judgment than a calculation, but it is a judgment disciplined by the values of the scientific community and constrained by the evidence. This is Kuhn’s answer to the charge that his account makes science subjective or relativistic.
Chapter 11: Possible Worlds in the History of Science
In this chapter, Kuhn engages with philosophical discussions about possible worlds and their relevance to understanding the history of science. He is responding to work by philosophers such as David Lewis and Saul Kripke, who had developed theories of possible worlds as a tool for analyzing modality—necessity and possibility.
Kuhn begins by noting that the concept of possible worlds has intuitive appeal for historians of science. When we study a past scientific theory that has been abandoned, we are in some sense considering a world that might have been—a world in which that theory was true. The Aristotelian universe, the phlogiston theory, the caloric theory of heat—these are not just false theories but entire ways of seeing the world that no longer exist.
However, Kuhn argues that the standard philosophical treatment of possible worlds is inadequate for the historian’s purposes. The standard approach treats possible worlds as alternative states of the actual world, differing only in the distribution of properties and the truth of propositions. But the shift from one scientific paradigm to another involves more than a change in which propositions are true. It involves a change in the very categories and concepts through which the world is described. The world of Aristotelian physics is not just a different distribution of properties from the Newtonian world; it is a world structured by different kinds of entities, different laws, different modes of explanation.
Kuhn then develops his own account of what it means to understand a past scientific theory. To understand Aristotelian physics is not simply to know that certain propositions about motion and place were asserted. It is to reconstruct the entire conceptual scheme within which those propositions made sense. This requires a kind of hermeneutic engagement, where the historian must learn to see the world through the categories of the past theory.
This chapter is significant because it deepens Kuhn’s account of incommensurability by engaging with contemporary analytic philosophy. Kuhn is not content to simply assert that past theories are different from present ones. He is trying to give a precise philosophical account of what that difference consists in. The historian’s encounter with past science is not a matter of comparing lists of propositions but of entering into a different conceptual world, and this requires a more complex understanding of meaning and reference than standard philosophical theories provide.
Chapter 12: The Trouble with the Historical Philosophy of Science (II)
We return to a theme that Kuhn has touched on throughout the collection: the tensions between history and philosophy of science as academic disciplines. This chapter extends the argument of the earlier “Trouble” chapter, focusing on the institutional and methodological dimensions of the relationship.
Kuhn begins by noting that the historical philosophy of science emerged in the 1960s and 1970s as a promising synthesis of two fields. Philosophers would use historical case studies to test and refine their theories of scientific method, while historians would draw on philosophical concepts to structure their narratives. This seemed to offer the best of both worlds: the empirical grounding of history combined with the conceptual rigor of philosophy.
However, Kuhn argues that this synthesis has proven more difficult than anticipated. The problem is that history and philosophy have different goals and standards of evidence. Philosophy aims at universal claims about the nature of scientific knowledge, claims that are supposed to hold for all times and places. History aims at particular explanations of specific events, explanations that are sensitive to context and contingency. The philosopher wants to abstract from the details to find general patterns; the historian wants to immerse in the details to understand what actually happened.
Kuhn then discusses the problem of normativity in the history of science. Philosophers want their accounts of science to have normative force—to tell us what good science should look like. Historians, by contrast, are primarily concerned with descriptive accuracy—telling us what science actually has been. The attempt to derive normative conclusions from historical cases is fraught with difficulty, because the cases themselves are ambiguous and open to multiple interpretations.
This chapter reveals Kuhn’s growing skepticism about the possibility of a unified historical philosophy of science. He does not reject the value of interdisciplinary work, but he insists that the two disciplines must maintain their distinct identities and methods. The historian cannot simply serve as a data provider for the philosopher, and the philosopher cannot simply impose a priori categories on the messy reality of history. The most productive relationship is one of mutual challenge and dialogue, where each discipline forces the other to confront its assumptions and limitations.
Chapter 13: The Natural and the Human Sciences (II)
This chapter revisits and extends the discussion from the earlier chapter on the same topic, focusing on the question of whether the human sciences can or should emulate the methods of the natural sciences. Kuhn is engaging with a long-standing debate in the philosophy of social science.
Kuhn begins by reiterating his commitment to the idea that the natural sciences provide our best model of knowledge. They are not perfect, but they are the most successful intellectual enterprise humans have ever devised. However, he immediately qualifies this by noting that the success of the natural sciences depends on specific features that may not be present in the human sciences.
The key feature of natural science, Kuhn argues, is its ability to generate paradigms that are accepted by the relevant community and that provide a framework for normal science research. The human sciences, by contrast, are characterized by a multiplicity of competing frameworks, none of which commands universal assent. This is not necessarily a defect—it may reflect the nature of the subject matter. Human behavior and social phenomena may be inherently more resistant to the kind of puzzle-solving that characterizes normal science.
Kuhn then introduces the concept of interpretive communities as an alternative model for the human sciences. Instead of aiming for universal laws or predictions, the human sciences might focus on the interpretation of meaning within specific communities. This would make them more like history or anthropology than like physics. The goal would not be to explain human behavior in terms of covering laws but to understand it in terms of the meanings and intentions that give it shape.
This chapter is important because it shows the limits of Kuhn’s model of science. The natural sciences, as described in *Structure*, may not be the appropriate model for all forms of inquiry. Kuhn is not arguing that the human sciences are inferior or that they should try to become more like physics. He is arguing that they have their own distinctive goals and methods, and that these should be articulated and defended on their own terms. The human sciences are not failed natural sciences; they are different enterprises altogether.
Chapter 14: Afterwords (II)
This is the second part of the retrospective reflection that closes the collection. Kuhn is responding to critics and clarifying his current positions on key issues that have been debated since the publication of *Structure*. The tone is more personal and reflective than in the other chapters.
Kuhn begins by addressing the charge of relativism once again, but with a new angle. He notes that many of his critics have assumed that incommensurability undermines the possibility of scientific progress. But this assumption, Kuhn argues, depends on a particular conception of progress that is itself tied to a discredited view of science. If progress means getting closer to the truth, in the sense of a correspondence between theory and reality, then incommensurability may indeed threaten it. But there is another conception of progress: increasing the ability to solve puzzles, to predict phenomena, to control nature. On this conception, progress is real and measurable.
Kuhn then discusses his evolving view of truth. He acknowledges that the concept of truth has been problematic in his work and that he has sometimes avoided it. He now wants to say that truth is a legitimate concept, but it must be understood in a context-dependent way. A statement is true within a particular conceptual scheme if it meets the standards of evidence and reasoning that are operative in that scheme. This is not a relativist position, because the standards themselves have a history and can be evaluated in terms of their success in solving problems.
The chapter concludes with Kuhn’s reflections on the future of science studies. He expresses concern that the field has become too fragmented and that the interdisciplinary dialogue he hoped for has not fully materialized. Philosophers, historians, and sociologists of science often talk past one another, using different methods and pursuing different goals. Kuhn calls for a renewed effort at genuine interdisciplinary communication, one that respects the distinct contributions of each discipline while seeking common ground.
This final substantive chapter matters because it shows Kuhn engaging with the ongoing legacy of his own work. He is not simply defending his earlier positions but developing and refining them in light of criticism and further reflection. The picture that emerges is of a thinker who is deeply committed to the project of understanding science as a human activity, and who is willing to change his mind when the evidence demands it. The *Road since Structure* is not a final statement but an ongoing conversation, and Kuhn is one of its most active participants.
