Scientific Change And Theory Choice

Science does not stand still, and it does not simply accumulate.

9 frameworks, from 1920 to today · Philosophy Of Science · Philosophy

still a working view ended · each leads to its entry below

Overview

Science does not stand still, and it does not simply accumulate. Theories that once commanded near-universal assent—phlogiston, caloric, the luminiferous ether—have been abandoned, while upstarts like Copernican astronomy or plate tectonics displaced frameworks that seemed secure. The philosophy of scientific change and theory choice asks how and why this happens. What makes one theory replace another? Is there a rational method that governs the transition, or do social, psychological, and rhetorical forces do the real work? And when scientists choose between rivals, are they guided only by evidence, or also by values—political, ethical, aesthetic—that have no business in a purely epistemic decision? These questions matter because the credibility of science as a rational enterprise hangs on them. If theory change is arbitrary, science loses its claim to special authority; if it is rigidly rule-governed, it seems to deny what historians actually observe. The field's history is a long negotiation between those two poles.

Logical Empiricism and the Dream of a Rational Reconstruction

The story begins in the 1920s and 1930s with the Vienna Circle, a group of philosophers and scientists—Moritz Schlick, Rudolf Carnap, Otto Neurath, and others—who sought to make philosophy scientific. Their movement, logical empiricism (often called logical positivism), held that meaningful statements must be either analytic (true by definition, like mathematics) or empirically verifiable123. Metaphysics was dismissed as nonsense12. Within this framework, scientific change was imagined as the steady accumulation of confirmed generalizations, and theory choice as a matter of matching theory to observation.

The trouble was that observation is never pure. Carnap and Neurath themselves recognized that observation reports are theory-laden, and the verification principle proved embarrassingly hard to state without self-refutation. But the deeper problem emerged from the history of science: major theory changes did not look like accumulation. When Einstein replaced Newton, it was not simply that new facts were added; the very concepts of mass, space, and time were redefined. Logical empiricism had no good account of such conceptual upheaval.

Falsificationism: Popper's Asymmetry

Karl Popper, working in Vienna and later New Zealand and London, offered a decisive alternative. In The Logic of Scientific Discovery (first published in German in 1934, English 1959), he argued that verification is impossible—no finite number of observations can confirm a universal law—but falsification is logically clean. A single counter-instance refutes a universal claim. Science progresses, on Popper's view, by bold conjectures and attempted refutations; a theory is scientific only if it forbids something, making it vulnerable to test145. Popper famously contrasted Einstein's general relativity, which predicted a specific deflection of starlight, with psychoanalysis, which could accommodate any outcome5.

Falsificationism captured something real about scientific rigor, and it remains a touchstone. But it faced its own difficulties. Historians pointed out that scientists routinely retain theories in the face of apparent refutation, treating anomalies as problems to be solved rather than death sentences. And the Duhem–Quine thesis—that hypotheses are tested not singly but as whole networks, so that a failed prediction tells you only that something in the network is wrong—undermined the image of the clean, decisive refutation167. Popper's account described how science should proceed, but not how it does.

Kuhn and the Structure of Revolutions

The decisive rupture came in 1962 with Thomas Kuhn's The Structure of Scientific Revolutions18910. Kuhn, a physicist turned historian, argued that science is normally conducted within a paradigm—a package of shared assumptions, exemplary problems, and techniques that defines a field and tells practitioners what counts as a legitimate question. Normal science is puzzle-solving within the paradigm, not perpetual testing of fundamentals111. Anomalies accumulate; when they become severe enough, the field enters crisis, and a revolution occurs in which a new paradigm replaces the old1910.

Kuhn's most provocative claims concerned theory choice. Rival paradigms, he argued, are incommensurable: their proponents use the same words with different meanings, so that communication across a revolutionary divide is partial and translation is never complete812. There is no neutral observation language against which to adjudicate. Kuhn likened paradigm choice to a political revolution and suggested that conversion, not proof, is the operative mechanism. This seemed to many to threaten scientific rationality; Kuhn spent much of his later career qualifying and defending his position, insisting he was not claiming that might makes right.

The concrete case that anchored the debate was the Copernican revolution itself: the shift from a geocentric to a heliocentric cosmos, in which the meanings of "earth," "motion," and "center" all changed together. Kuhn's account made the history of science philosophically central, and it remains the most influential single work in the field.

Lakatos and the Methodology of Research Programmes

Imre Lakatos, a Hungarian philosopher who fled to Britain, sought a middle path between Popper and Kuhn. In his Methodology of Scientific Research Programmes (essays collected posthumously in 1978), he argued that the unit of appraisal is not the isolated theory but the research programme: a hard core of assumptions protected from refutation by a belt of auxiliary hypotheses, plus a set of heuristic rules for extending the programme. A programme is progressive if it predicts novel facts that are subsequently confirmed; it is degenerating if it only patches up anomalies after the fact. Scientists should stick with a progressive programme and abandon a degenerating one—but Lakatos admitted that judging whether a programme is progressive may take decades, and that there is no mechanical rule for when to give up.

Lakatos's scheme preserved rational standards while accommodating Kuhn's observation that theories are never abandoned at the first anomaly. His historical case studies, notably of the Bohr model of the atom, showed how a programme can survive apparent refutation and later vindicate itself.

Research Traditions and Problem-Solving

Larry Laudan offered a further refinement in Progress and Its Problems (1977). He replaced paradigms and research programmes with research traditions: sets of assumptions about the entities and methods appropriate to a domain. Traditions are not simply true or false; they are better or worse at solving problems. Laudan distinguished empirical problems (explaining observed phenomena) from conceptual problems (internal inconsistencies, conflicts with other accepted theories), and argued that progress consists in the increasing problem-solving effectiveness of a tradition. This gave a more flexible, pluralistic picture: traditions can coexist, and a scientist may rationally work within one while acknowledging another's strengths.

The Strong Programme: Sociology Enters

If Kuhn made room for social factors, the Edinburgh school drove straight through the opening. In the 1970s, Barry Barnes and David Bloor proposed the Strong Programme in the sociology of scientific knowledge, most programmatically in Bloor's Knowledge and Social Imagery (1976). Its central claim was that sociological explanation should apply symmetrically to true and false beliefs alike: one should not explain a belief's acceptance by its truth, but by the same kinds of social causes that explain error13. The programme's four tenets—causality, impartiality, symmetry, and reflexivity—made it a radical challenge to the philosophy of science. Bloor's own case study of the controversy over Pythagorean numerology and the work of the Edinburgh school on the "science wars" of the 1990s made the position famous and controversial. Critics charged that symmetry collapses the distinction between warranted and unwarranted belief; defenders replied that philosophers had simply assumed what needed explaining.

Bayesian Confirmation and Decision Theory

Running alongside these historical and sociological developments was a formal tradition. Bayesian confirmation theory treats scientific reasoning as probabilistic updating: a scientist begins with prior probabilities for hypotheses and revises them by Bayes' theorem as evidence arrives1415. The approach, developed by philosophers including Rudolf Carnap, Richard Jeffrey, and later Jon Dorling and Colin Howson, offers a precise account of how evidence can favor one theory over another even when neither is certain. It also connects theory choice to decision theory: choosing a theory is choosing an action under uncertainty, and one can ask which choice maximizes expected epistemic utility.

Bayesianism has real explanatory power—it captures the intuition that surprising predictions confirm more strongly than expected ones—but it faces the problem of prior probabilities: where do they come from, and can they be justified objectively? The debate over Bayesianism vs. frequentism and over the objectivity of priors remains live.

Values in Theory Choice

A distinct strand asks whether non-epistemic values—ethical, political, social—legitimately enter theory choice. The traditional answer was no: science should be value-free. But philosophers including Helen Longino, in Science as Social Knowledge (1990), and later Heather Douglas, in Science, Policy, and the Value-Free Ideal (2009), argued that values inevitably shape which hypotheses are pursued, which evidence is gathered, and how uncertainty is characterized. Douglas distinguished the direct role of values in choosing what to study from their indirect role in assessing evidence, and argued that a rigid value-free ideal is neither achievable nor desirable. The debate over inductive risk—the risk of accepting a false hypothesis versus rejecting a true one—shows how value judgments enter even the most technical inference.

Social Epistemology of Science

The most recent major development is the social epistemology of science, which treats scientific communities, rather than individual scientists, as the primary knowers. Work by Philip Kitcher (The Advancement of Science, 1993), Alvin Goldman, and Miriam Solomon examines how the division of cognitive labor, peer review, dissent, and consensus formation affect the reliability of scientific knowledge. Formal models of network epistemology, such as those developed by Kevin Zollman, ask when communication structures help or hinder the community's convergence on truth. This tradition descends from both Kuhn and the sociology of science but retains a normative, philosophical orientation: it asks not just how science works but how it ought to be organized.

The Present

Today the field is pluralistic. Bayesian formal epistemology, social epistemology, and values-in-science debates are all active, often in conversation with each other. The Kuhn–Popper–Lakatos axis remains the historical backbone, but the questions have broadened: How should scientific communities be structured to avoid groupthink? When is it rational to dissent from consensus? How should values be managed rather than excluded? And the old puzzle—whether there is a universal, ahistorical method of theory choice—has largely given way to a more contextual, empirically informed picture in which rational standards are real but historically situated. What remains open is how to reconcile the normative ambition of philosophy with the messy, contingent record that history and sociology reveal.

References

  1. Philosophy of science (Logical positivism, Defining science, Coherentism, introduction, Thomas Kuhn), Wikipedia
  2. Logical positivism, Wikipedia
  3. Vienna Circle, Wikipedia
  4. Falsifiability, Wikipedia
  5. Karl Popper (introduction, Background to Popper's ideas), Wikipedia
  6. Duhem–Quine thesis, Wikipedia
  7. Confirmation holism, Wikipedia
  8. Thomas Kuhn, Wikipedia
  9. Paradigm shift, Wikipedia
  10. The Structure of Scientific Revolutions (introduction, History), Wikipedia
  11. Normal science, Wikipedia
  12. Commensurability (philosophy of science), Wikipedia
  13. Strong programme, Wikipedia
  14. Bayesian inference, Wikipedia
  15. Bayesian probability, Wikipedia

The frameworks of Scientific Change And Theory Choice

Each framework is a way of working the field has used: when it held, and the question it answers.

  1. 1920-1960

    Logical positivism

    Which statements are meaningful?

    Fly to its tower in Nousfarer

  2. 1934-1980
  3. 1950-Present

    Bayesian epistemology

    How should belief change with evidence?

    Fly to its tower in Nousfarer

  4. 1962-Present

    Paradigm Theory

    How does science change through revolutions?

    About Paradigm Theory
    Kuhnian Paradigm Theory transformed the philosophy of science by asking a deceptively simple question: how do scientific communities actually change their most fundamental beliefs? Before Kuhn, most philosophers assumed that science progresses by steadily accumulating truths through a universal method. Kuhn argued instead that science alternates between long periods of puzzle-solving under a shared paradigm and brief, disruptive revolutions where the paradigm itself is replaced. This account challenged the idea that theory choice is a purely rational, rule-governed process and introduced historical and sociological dimensions into the heart of philosophy of science.
    Core Curriculum
    The following concepts form the teaching spine of Kuhnian Paradigm Theory in the philosophy of scientific change:
    1. Paradigm – A paradigm is the constellation of beliefs, values, techniques, and exemplars shared by a scientific community. It defines what counts as a legitimate problem and method during normal science.
    2. Normal Science – The routine, puzzle-solving activity conducted under a paradigm. Scientists work to extend and articulate the paradigm, not to test it fundamentally.
    3. Anomaly – A persistent failure of the paradigm to solve a problem it was expected to handle. Anomalies are initially set aside or explained away, but accumulating anomalies can trigger a crisis.
    4. Crisis – A period of pronounced professional insecurity when the paradigm's ability to guide research is widely questioned. During a crisis, alternative approaches proliferate.
    5. Revolution – A non-cumulative episode in which an older paradigm is replaced in whole or in part by an incompatible new one. Revolutions are the mechanism of scientific change.
    6. Paradigm Shift – The process by which a community transitions from one paradigm to another. This is not a gradual accumulation but a gestalt-like transformation.
    7. Incommensurability – The claim that successive paradigms lack a common measure. They differ in their concepts, standards, and even the world they perceive, making direct point-by-point comparison impossible.
    8. Scientific Community – The social group that shares a paradigm. For Kuhn, the community is the locus of scientific authority; paradigm change is a community decision, not an individual one.
    9. Exemplar – A concrete problem-solution that the community treats as a model for further work. Exemplars are a key way paradigms are transmitted and learned.
    10. Disciplinary Matrix – Kuhn's later, more precise term for what he earlier called a paradigm. It includes symbolic generalizations, models, values, and exemplars.
    11. Puzzle-Solving – The characteristic activity of normal science. Puzzles are problems guaranteed to have a solution within the paradigm, unlike genuine novelties.
    12. Gestalt Switch – A metaphor for the perceptual and conceptual shift that occurs during a revolution. After the switch, scientists see the world differently.
    13. Textbook Science – The sanitized, retrospective version of science presented in textbooks. It obscures the revolutionary history of the field by presenting the current paradigm as the natural outcome of cumulative progress.
    14. Pre-paradigm Science – An early, immature stage of a field before a consensus paradigm emerges. Multiple schools compete without a shared framework.
    15. Extraordinary Science – The period of crisis and revolution when normal science breaks down and scientists explore novel theories and methods.
    Intellectual History
    Kuhnian Paradigm Theory emerged from a specific historical moment in the philosophy of science. In the early 1960s, the dominant framework was Karl Popper's Critical Rationalism, which held that science progresses by bold conjectures and severe falsification tests. Thomas Kuhn, a physicist turned historian of science, found this account inconsistent with the historical record. His landmark work, The Structure of Scientific Revolutions (1962), offered an alternative rooted in the actual behavior of scientific communities.
    Kuhn's key insight was that scientists rarely test their fundamental assumptions. Instead, they work within a paradigm, solving puzzles and extending its reach. Only when anomalies accumulate to the point of crisis does the community consider alternatives. The choice between paradigms, Kuhn argued, is not fully rational in the Popperian sense because paradigms are incommensurable: they differ in their standards, concepts, and even the phenomena they recognize. Persuasion, rhetoric, and generational turnover play crucial roles in paradigm shifts.
    The book provoked immediate and intense debate. Critics charged that Kuhn had reduced science to mob psychology or irrational consensus. Kuhn responded by refining his concepts, introducing the term "disciplinary matrix" in the postscript to the second edition (1970) to clarify what he meant by paradigm. He also emphasized that incommensurability was partial, not total, and that theory choice involved shared values like accuracy, consistency, scope, simplicity, and fruitfulness—even if those values could be interpreted differently by adherents of rival paradigms.
    Kuhn's framework coexisted with and influenced several successor and parallel frameworks. Imre Lakatos's Methodology of Scientific Research Programmes (1970) attempted to reconcile Kuhn's historical sensitivity with Popperian rationality by treating science as a series of competing research programs with hard cores and protective belts. Paul Feyerabend's Epistemological Anarchism (1975) radicalized Kuhn's insights into a full-blown methodological pluralism. Larry Laudan's Research Traditions (1977) offered a more fine-grained account of scientific change. The Sociology of Scientific Knowledge (1970s onward) took Kuhn's emphasis on community and turned it into a systematic empirical program studying how scientific knowledge is socially constructed.
    Kuhn's work also had a profound impact beyond philosophy of science, influencing fields as diverse as sociology, political science, economics, and management studies. The term "paradigm shift" entered popular culture as a label for any fundamental change in thinking. Within philosophy of science, however, Kuhnian Paradigm Theory remains a touchstone for debates about rationality, progress, and the nature of scientific change. Its core concepts—paradigm, normal science, revolution, incommensurability—continue to structure introductory courses and shape research agendas.
    The 15 concepts its tower teaches
    • Anomaly
    • Crisis
    • Disciplinary Matrix
    • Exemplar
    • Extraordinary Science
    • Gestalt Switch
    • Incommensurability
    • Normal Science
    • Paradigm
    • Paradigm Shift
    • Pre-paradigm Science
    • Puzzle-Solving
    • Revolution
    • Scientific Community
    • Textbook Science

    Fly to its tower in Nousfarer

  5. 1970-1990
  6. 1976-2000

    Strong Programme

    What social causes shape scientific belief?

    Fly to its tower in Nousfarer

  7. 1977-2000
  8. 1990-Present

    Values in Science

    Should values influence scientific choice?

    Fly to its tower in Nousfarer

  9. 1993-Present

    Social epistemology

    How do communities produce knowledge?

    Fly to its tower in Nousfarer

How the ideas connect

Each arrow runs from an earlier idea to the later one it shaped. The list below says what passed between them.

Logical positivismFalsificationismBayesian epistemologyParadigm TheoryMethodology of Scientific Research Pr…Strong ProgrammeResearch Traditions ModelValues in ScienceSocial epistemology

Beyond the field

Where Scientific Change And Theory Choice's ideas came from in other fields, and what they led to there: the strongest 12 of 44 links.

A city in Nousfarer: a red-carpeted stair climbing terraces to a crowned tower on an acropolis, more towers round it, forest and snowy mountains beyond.
A city in Nousfarer, the game played in the browser.

Scientific Change And Theory Choice is a planet in Nousfarer

Its city has a tower for each of these 9 frameworks. Land, walk in, and climb one concept at a time.

Land on Scientific Change And Theory Choice

Related fields

Fields its ideas are linked to

More in Philosophy Of Science

See where Scientific Change And Theory Choice sits on the Map of Human Knowledge