Philosophy Of Science

Philosophy of science asks what science is, how it works, and what it can tell us about the world.

30 frameworks, from 350 BCE to today · Philosophy Of Science · Philosophy

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

Overview

Philosophy of science asks what science is, how it works, and what it can tell us about the world12. It is not a branch of science but a reflective discipline that stands beside it, examining the logic of evidence, the structure of theories, the meaning of explanation, and the social conditions under which knowledge is made1. The stakes are practical as well as abstract: if we can say what makes a scientific claim credible, we can defend science against superstition and pseudoscience—and also understand its limits. The field's history is a long argument about whether there is a single scientific method, what that method is, and whether science progresses toward truth or merely changes its fashions13.

Aristotle and the Origins of Demonstrative Science

The Western tradition begins with Aristotle (384–322 BCE), who set out in the Posterior Analytics a model of what he called demonstrative science (epistēmē apodeiktikē). Scientific knowledge, for Aristotle, was knowledge of why things are as they are, derived by valid deduction from premises that are themselves necessary and explanatory. A demonstration is a syllogism whose premises state the causes of the conclusion: to know scientifically that a triangle has angles equal to two right angles, one must grasp the cause, not merely the fact. The premises must be true, primary, and better known than the conclusion. This ideal shaped natural philosophy for nearly two millennia. It also created a problem: where do the first premises come from? Aristotle answered that they are grasped by nous, an intellectual intuition of essences, which left empirical inquiry in an ambiguous position—observation could suggest universal principles but could not prove them.

The Mechanical Philosophy and the New Experimentalism

The Aristotelian synthesis broke down in the sixteenth and seventeenth centuries4. The mechanical philosophy—associated with René Descartes, Robert Boyle, and others—replaced Aristotelian forms and final causes with matter in motion: all natural phenomena were to be explained by the size, shape, and motion of particles. This was a metaphysical program as much as a methodological one, and it raised a new question: if the underlying corpuscles are invisible, how can we know hypotheses about them are true? Boyle's answer was the experimental method: carefully reported experiments, witnessed by credible gentlemen, could establish matters of fact even when the underlying causes remained conjectural.

Francis Bacon (1561–1626) supplied an influential methodological vision in the Novum Organum (1620)1567. His Baconian inductive method proposed that the investigator should ascend from particulars to general axioms by a graduated process, using "tables of discovery" to organize observations and eliminate false generalizations7. Bacon was sharply critical of the deductive ideal, arguing that it bred endless verbal disputes. His program was more a manifesto than a working method—he did not himself make major discoveries—but it gave lasting prestige to the idea that science should proceed by systematic observation and induction rather than by speculation.

Newton and the Problem of Method

Isaac Newton's Principia (1687) became the model of successful science, and Newton himself offered methodological pronouncements that shaped later debate14. His famous phrase hypotheses non fingo—"I frame no hypotheses"—expressed a refusal to speculate about the cause of gravity while insisting that the mathematical description was derived from phenomena. Newton's methodology combined mathematical deduction from observed regularities with a cautious empiricism about causes. His "Rules of Reasoning in Philosophy" included the principle that we should admit no more causes than are true and sufficient to explain the phenomena, and that like effects should be assigned like causes. This was read by many as a license for inductive generalization from observation, and it set the terms for the next century's debates about whether scientific reasoning could be reduced to induction.

Whewell, Mill, and the Inductive Debate

In the nineteenth century the question of method became a live philosophical controversy1. William Whewell (1794–1866) argued in The Philosophy of the Inductive Sciences (1840) that discovery involves what he called the colligation of facts: the mind imposes a conception—a unifying idea—on otherwise disconnected observations, and the fit between conception and fact is judged by a "consilience of inductions," when a theory successfully predicts phenomena of a kind not used in its construction. Whewell's account was Kantian in spirit: the mind contributes the conceptual framework. John Stuart Mill (1806–1873), in A System of Logic (1843), defended a more austere empiricism. His methods of experimental inquiry—the methods of agreement, difference, concomitant variation, and residues—were canons for inferring causal relationships from observed correlations. Mill held that all inference is ultimately inductive and that the uniformity of nature is the ultimate premise. The Whewell–Mill exchange framed the central tension that would persist: does science discover facts or construct interpretations?

Positivism and the Logical Empiricists

Auguste Comte (1798–1857) gave the name positivism to the view that genuine knowledge is confined to observable phenomena and their laws, and that metaphysical speculation belongs to an earlier stage of human development18. This outlook was radicalized in the early twentieth century by the Vienna Circle, a group of philosophers and scientists including Moritz Schlick, Rudolf Carnap, Otto Neurath, and Hans Hahn. Logical empiricism (or logical positivism) combined empiricism with the new logic of Frege and Russell1910. Its central project was the logical analysis of scientific language: to show how theoretical statements are related to observation, and to eliminate metaphysics by a criterion of meaningfulness1. The verification principle—that a statement is meaningful only if it can be verified by experience—was its most famous weapon, though it proved difficult to state without self-refutation.

The logical empiricists made lasting contributions: the distinction between the context of discovery and the context of justification (Reichenbach), the analysis of confirmation, and the project of formalizing scientific theories as axiomatic systems. But their program faced internal difficulties19. The verification principle excluded universal laws, which cannot be conclusively verified. The analytic–synthetic distinction, on which their account of logic depended, was attacked by W. V. O. Quine in "Two Dogmas of Empiricism" (1951)11112. And the attempt to reduce theoretical terms to observation terms proved elusive.

Falsificationism and the Demarcation Problem

Karl Popper (1902–1994) offered a response that reshaped the field. In The Logic of Scientific Discovery (first published in German in 1934), Popper argued that falsificationism solves the problem of demarcating science from pseudoscience: a scientific theory is one that forbids certain observations, and is therefore testable. Einstein's relativity made risky predictions; Freudian psychoanalysis, Popper claimed, could accommodate any observation and so was not scientific. Science proceeds by conjecture and refutation: we propose bold hypotheses and try to falsify them13. Popper's account captured something important about the critical spirit of science, but it faced the Duhem–Quine problem—a failed prediction never refutes a single hypothesis in isolation, because auxiliary assumptions can always be adjusted114. And it seemed to leave no room for the ordinary confirmation that scientists rely on.

Explanation: The Deductive-Nomological Model and Its Rivals

A parallel tradition asked not how theories are confirmed but what it means to explain1. Carl Hempel and Paul Oppenheim's deductive-nomological model (1948) held that an explanation is a deductive argument whose premises include at least one law of nature and whose conclusion is the event to be explained. To explain why a particular pendulum has a certain period, one deduces it from the law of pendulum motion and the initial conditions. The model was elegant and unified explanation with prediction, but it had counterintuitive consequences: a flagpole's shadow can be deduced from the law of light propagation and the flagpole's height, yet we do not think the shadow explains the height15. Wesley Salmon and others pressed this asymmetry problem, and alternative accounts emerged116. The causal-mechanical approach, developed by Salmon and later by Peter Machamer, Lindley Darden, and Carl Craver, holds that to explain is to identify the mechanism that produces the phenomenon—to show how the parts and their activities bring it about. This view proved especially congenial to biology and neuroscience, where explanations often cite mechanisms rather than laws.

Kuhn and the Historical Turn

The most disruptive intervention came from Thomas Kuhn, a historian of science, whose The Structure of Scientific Revolutions (1962) introduced the concept of the paradigm: a shared body of exemplars, methods, and assumptions that defines a scientific community's problems and standards11718. Kuhn argued that science alternates between "normal science," in which researchers solve puzzles within an accepted paradigm, and revolutionary periods, in which anomalies accumulate and a new paradigm displaces the old11920. Crucially, Kuhn denied that theory choice is governed by a fixed algorithm: rival paradigms are often incommensurable, meaning their proponents disagree about what counts as a problem, a solution, and even the meaning of shared terms17. This was read by many as a threat to scientific rationality, and it provoked a long debate19. Imre Lakatos responded with the methodology of scientific research programmes (1970), which sought to preserve rational standards by distinguishing a "hard core" of protected assumptions from a "protective belt" of auxiliary hypotheses that can be revised, and by judging programmes by their progressive or degenerating problem-shifting. Paul Feyerabend went further, arguing in Against Method (1975) for epistemological anarchism: there is no universal method, and the history of science shows that great advances often violated the rules philosophers proposed12122.

Realism, Empiricism, and the Social Turn

The realism debate crystallized in the 1960s and after. Scientific realism holds that successful scientific theories are approximately true and that their theoretical terms refer to real entities—electrons, genes, black holes111. Its central argument is the "no miracles" argument: it would be a miracle if theories that predict novel phenomena so successfully were not at least approximately true1. Bas van Fraassen's constructive empiricism (1980) offered a rival: science aims at empirical adequacy, not truth1123. We should accept theories that save the phenomena, but we need not believe in the unobservable entities they posit111. The debate turned on what counts as observable and on the historical record of theories once thought true and later abandoned.

Meanwhile, a sociological challenge emerged1. The sociology of scientific knowledge (SSK), associated with the "strong programme" of David Bloor and Barry Barnes at Edinburgh in the 1970s, argued that the content of scientific knowledge—not just its institutional setting—could be explained by social causes242526. Bloor's symmetry principle held that true and false beliefs should be explained in the same way, without assuming that truth is its own explanation2526. This was radical and controversial, and it fed into the "science wars" of the 1990s127. Feminist philosophy of science, developing from the 1970s onward, raised related questions about whose knowledge counts and how values shape inquiry, arguing that the ideal of value-free science obscured the role of background assumptions and social position in theory choice.

Naturalism, Pluralism, and the Special Sciences

By the late twentieth century, philosophy of science had become increasingly naturalistic: rather than prescribing how science ought to proceed, philosophers increasingly described how it actually does, drawing on history, psychology, and sociology. This was reinforced by the growth of philosophy of the special sciences—biology, neuroscience, economics, and the social sciences—each raising problems that the physics-centered models did not address. Scientific pluralism, defended by philosophers such as Nancy Cartwright and Helen Longino, holds that there may be no single correct account of a domain, and that multiple models can be legitimately useful for different purposes28. Cartwright's How the Laws of Physics Lie (1983) argued that fundamental laws are idealizations that rarely hold in the messy world, while phenomenological models do the explanatory work.

Bayesian Confirmation and Inference to the Best Explanation

Two approaches to confirmation have dominated recent decades. Bayesian confirmation theory, developed from the eighteenth century onward and formalized in the twentieth by philosophers such as Rudolf Carnap and later Jon Dorling and Colin Howson, treats confirmation probabilistically: evidence confirms a hypothesis to the extent that it raises the hypothesis's probability, as calculated by Bayes' theorem. The approach is powerful and precise, but it faces the problem of where prior probabilities come from and how to handle the old evidence problem. Inference to the best explanation, articulated by Gilbert Harman (1965) and developed by Peter Lipton, holds that we infer the truth of a hypothesis because it provides the best explanation of the evidence—where "best" is judged by explanatory virtues such as scope, simplicity, and unification. This captures a form of reasoning that scientists clearly use, though critics ask whether "best explanation" can be defined without circularity1.

The Present

Philosophy of science today is a plural field. Formal epistemology and Bayesian methods continue to refine confirmation theory. The causal-mechanical approach dominates philosophy of biology and neuroscience, where mechanisms are the explanatory currency. Social epistemology, feminist philosophy of science, and the study of values in science have moved from the margins to the center, asking how scientific communities should be organized and how non-epistemic values legitimately enter inquiry. The realism debate persists in new forms, including structural realism and the question of how to interpret quantum mechanics. And the naturalistic turn has brought philosophy of science into closer contact with the sciences themselves, so that the boundary between philosophical and scientific reflection on method is often blurred2930. The old questions—what is a good explanation, when does evidence support a theory, is there a scientific method—remain open, but they are now asked with a richer sense of how science actually works.

References

  1. Philosophy of science (introduction, Modern, Philosophy of social science, Logical positivism, Thomas Kuhn, Coherentism, Scientific explanation, Anything goes methodology, The purpose of science, Sociology of scientific knowledge methodology, Justifying science), Wikipedia
  2. History and philosophy of science, Wikipedia
  3. Scientific method (History), Wikipedia
  4. Scientific Revolution, Wikipedia
  5. Works by Francis Bacon, Wikipedia
  6. Baconian method, Wikipedia
  7. Novum Organum, Wikipedia
  8. Positivism, Wikipedia
  9. Logical positivism, Wikipedia
  10. Vienna Circle, Wikipedia
  11. Scientific realism (History, introduction), Wikipedia
  12. Two Dogmas of Empiricism, Wikipedia
  13. Problem of induction, Wikipedia
  14. Falsifiability, Wikipedia
  15. Deductive-nomological model, Wikipedia
  16. Wesley C. Salmon, Wikipedia
  17. Thomas Kuhn, Wikipedia
  18. Paradigm shift, Wikipedia
  19. The Structure of Scientific Revolutions, Wikipedia
  20. Normal science, Wikipedia
  21. Paul Feyerabend (Anarchist phase), Wikipedia
  22. Against Method, Wikipedia
  23. Constructive empiricism, Wikipedia
  24. Sociology of scientific knowledge, Wikipedia
  25. Strong programme, Wikipedia
  26. Science studies (History of the field), Wikipedia
  27. Science wars, Wikipedia
  28. Scientific pluralism (introduction, History), Wikipedia
  29. Naturalized epistemology, Wikipedia
  30. Naturalism (philosophy), Wikipedia

The frameworks of Philosophy Of Science

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

  1. 350 BCE-1650
  2. 350 BCE-1600
  3. 1600-1750

    Mechanical philosophy

    What explains the behavior of natural bodies?

    Fly to its tower in Nousfarer

  4. 1620-1800

    Baconian method

    How should we gather knowledge from experience?

    Fly to its tower in Nousfarer

  5. 1620-1900

    Inductivism

    How do particular observations support general claims?

    Fly to its tower in Nousfarer

  6. 1687-1800

    Newtonian methodology

    How should we reason from phenomena to laws?

    Fly to its tower in Nousfarer

  7. 1830-Present

    Hypothetico-Deductivism

    How are hypotheses tested against evidence?

    Fly to its tower in Nousfarer

  8. 1830-1950

    Positivism

    What counts as genuine knowledge of the world?

    Fly to its tower in Nousfarer

  9. 1840-1900

    Whewellian colligation and consilience

    How does the mind unify observations into a theory?

    Fly to its tower in Nousfarer

  10. 1843-1900

    Mill's methods

    How can causes be inferred from observed correlations?

    Fly to its tower in Nousfarer

  11. 1902-Present

    Conventionalism

    Which parts of a theory are chosen rather than discovered?

    Fly to its tower in Nousfarer

  12. 1920-1960

    Logical positivism

    How can scientific language be made logically clear?

    Fly to its tower in Nousfarer

  13. 1934-1980

    Falsificationism

    How do we tell science from pseudoscience?

    Fly to its tower in Nousfarer

  14. 1948-1980

    Deductive-nomological model

    What makes an explanation a scientific one?

    Fly to its tower in Nousfarer

  15. 1950-Present

    Bayesian confirmation theory

    How much should evidence raise our confidence?

    Fly to its tower in Nousfarer

  16. 1950-Present

    Bayesian Confirmation Theory

    How should belief change with new evidence?

    Fly to its tower in Nousfarer

  17. 1960-Present

    Scientific realism

    Are successful theories true of unobservable things?

    Fly to its tower in Nousfarer

  18. 1962-Present

    Kuhnian paradigms and revolutions

    How does scientific knowledge change over time?

    Fly to its tower in Nousfarer

  19. 1965-Present

    Abductive reasoning

    Which explanation of the evidence is best?

    Fly to its tower in Nousfarer

  20. 1969-Present

    Naturalism

    Should philosophy of science be continuous with science?

    Fly to its tower in Nousfarer

  21. 1970-Present

    Feminist philosophy of science

    How do social values shape scientific knowledge?

    Fly to its tower in Nousfarer

  22. 1970-1990
  23. 1970-Present

    Philosophy of the Special Sciences

    How does explanation work in biology and psychology?

    Fly to its tower in Nousfarer

  24. 1975-1995

    Epistemological anarchism

    Is there any universal scientific method?

    Fly to its tower in Nousfarer

  25. 1976-2000

    Sociology of scientific knowledge

    How is scientific consensus actually produced?

    Fly to its tower in Nousfarer

  26. 1980-Present

    Causal-Mechanical Explanation

    What mechanism produces the phenomenon?

    Fly to its tower in Nousfarer

  27. 1980-Present

    Constructive empiricism

    Should science aim at truth or only adequacy?

    Fly to its tower in Nousfarer

  28. 1980-Present

    Social Epistemology of Science

    How does testimony shape scientific knowledge?

    Fly to its tower in Nousfarer

  29. 1983-Present

    Scientific pluralism

    Should science tolerate many incompatible models?

    Fly to its tower in Nousfarer

  30. 1984-Present

    Causal-mechanical explanation

    How do parts and activities bring about an outcome?

    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.

Aristotelian Demonstration and Natura…Aristotelian demonstrative scienceMechanical philosophyBaconian methodInductivismNewtonian methodologyHypothetico-DeductivismPositivismWhewellian colligation and consilienceMill's methodsConventionalismLogical positivismFalsificationismDeductive-nomological modelBayesian confirmation theoryBayesian Confirmation TheoryScientific realismKuhnian paradigms and revolutionsAbductive reasoningNaturalismFeminist philosophy of scienceMethodology of scientific research pr…Philosophy of the Special SciencesEpistemological anarchismSociology of scientific knowledgeCausal-Mechanical ExplanationConstructive empiricismSocial Epistemology of ScienceScientific pluralismCausal-mechanical explanation

Beyond the field

Where Philosophy Of Science's ideas came from in other fields, and what they led to there: the strongest 12 of 188 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.

Philosophy Of Science is a planet in Nousfarer

Its city has a tower for each of these 30 frameworks. Land, walk in, and climb one concept at a time.

Land on Philosophy Of Science

Related fields

Fields its ideas are linked to

More in Philosophy Of Science

See where Philosophy Of Science sits on the Map of Human Knowledge