Incommensurability

Introduction: The Problem of Comparison

Angela Bogdanova
Research Dossier · Philosophy of Science
Published by Aisentica Development
American English · November 7, 2026
Reserved canonical: https://philosophybybogdanova.com/dossiers/incommensurability/
Editorial identity: research-stage manuscript; first-publication date not assigned.

The central question of incommensurability is how scientists can compare accounts of the world when the accounts also alter the terms in which comparison is conducted. Scientific theories assign roles to quantities, specify natural kinds, authorize instruments, prescribe relevant problems, and establish explanatory expectations. A revolutionary revision can change several of these relationships at once. The historian confronting such a revision must reconstruct the older intellectual setting while assessing the later achievements. The philosopher must explain how rational criticism remains possible when a stable vocabulary cannot be assumed in advance.

Thomas Kuhn and Paul Feyerabend made the problem prominent in 1962 through very different lines of argument. The Stanford analysis of incommensurability distinguishes their early positions and later developments (Oberheim and Hoyningen-Huene, 2024). Kuhn situated the issue within changes in disciplinary practice. Feyerabend foregrounded incompatible theoretical meanings and questioned models of reduction that treat historical theories as if their central terms survived intact. Their shared concern was not simply that scientists disagree about facts. The dispute concerns what makes an alleged fact, a relevant test, or a successful explanation intelligible within a scientific tradition.

The strongest reading—that rival scientific communities inhabit sealed intellectual worlds and lack any basis for communication—would imply historical and practical consequences much stronger than the central sources demonstrate. Scientists argue across theories, reproduce apparatus, calculate correspondence limits, translate portions of technical discourse, and learn formerly alien classifications. These activities reveal the actual burden of the concept: an account of comparison must describe its constructive labor rather than presuppose complete, costless transfer. The analysis below therefore treats incommensurability as a variable relation between particular theoretical or practical structures, tested case by case, and asks what remains comparable after identifying the relevant discontinuity.

The Mathematical Analogy and Its Limits

In elementary geometry, two magnitudes are incommensurable when no common magnitude measures each by a whole-number multiple. The side and diagonal of a square supply the classical illustration: their ratio is the irrational number √2. Their mathematical incommensurability neither prevents their measurement by approximation nor makes comparison meaningless. Indeed, the precise theorem requires a shared geometrical context in which both magnitudes can be specified.

The transfer of the term into philosophy of science is therefore analogical. Kuhn’s paradigms do not possess a single measurable ratio whose irrationality establishes their relationship. The relevant difficulty concerns partially incompatible classificatory, semantic, and evaluative practices. The historical analysis of the term tracks the transition from mathematical to philosophical uses (Oberheim and Hoyningen-Huene, 2024). The analogy usefully suggests that a universal common measure may fail to exist, but it supplies no warrant for concluding that every comparative relation fails.

The distinction matters because philosophical discussion frequently trades on an ambiguity between lack of one complete measure and lack of any comparison. Two theories may share predictions but disagree about what underlying entities exist. They may employ similar terms but organize those terms differently. They may address overlapping experiments while ranking explanatory objectives in different orders. Each is a different problem requiring a different method of comparison. The mathematical heritage licenses the question of commensuration; it does not decide the epistemic answer.

The 1962 Historical Conjuncture

The modern debate emerged from a particular intellectual setting. Logical empiricist reconstructions often described theoretical progress through confirmation, explanation, and relations between theoretical languages. Some accounts of reduction asked whether one theory could be deduced from another once suitable bridge principles connected their terms. Against this background, historical episodes raised a difficulty: the terms requiring bridges could change their meaning through the very historical transition under study.

In The Structure of Scientific Revolutions, Kuhn (1962) connected these difficulties to scientific communities, exemplars, normal science, and revolutionary reorientation. In the same year, Feyerabend’s Explanation, Reduction, and Empiricism examined the presumed invariance of meanings across theories (Feyerabend, 1962). Their arguments arrived at related problems by different routes and should be attributed independently.

The coincidence of 1962 is historically important without requiring an artificial story of two identical doctrines. Kuhn examined the character of disciplinary transitions and scientific learning. Feyerabend interrogated methodological rules that would outlaw conceptually distant alternatives before they could become productive. The resulting debate spans historical description, semantics, and normative epistemology. A theory can be historically revolutionary without every term changing meaning; conversely, a local semantic replacement can occur without a comprehensive change of scientific institutions. The category must retain these distinctions to explain rather than merely label change.

Kuhn’s Paradigm-Centered Account

Kuhn’s original word “paradigm” covered exemplary achievements and a community’s shared resources for solving scientific problems. The term became contested partly because readers treated it as a single rigid worldview. Kuhn subsequently refined his account with the disciplinary matrix and exemplars. The Kuhn scholarly entry reconstructs these distinctions and the subsequent debate (Bird, 2025).

An exemplar is a worked solution through which a novice learns what counts as a properly framed scientific problem. Learning to identify a physical system, isolate a variable, and deploy an equation occurs through practice with cases, not only through mastery of explicit definitions. A disciplinary matrix includes symbolic generalizations, values, models, and shared examples; its components organize a scientist’s responses to novel cases. Incommensurability arises partly because successive arrangements teach different habits of recognition, classification, and assessment.

The historical claim has an institutional dimension. Laboratories and classrooms transmit ways of seeing instruments and phenomena; textbooks retrospectively arrange earlier research around concepts accepted by a later community. A historian who imports present-day classifications into earlier practice can conceal the significance of a transformation. Yet Kuhn’s view also requires continuity: otherwise the story of competing paradigms, scientific crises, and scientific advances would be unintelligible even to its author. The relevant phenomenon is altered organization of shared and partially shared problems, not historical amnesia.

Normal Science, Crisis, and Revolution

Normal science organizes research around accepted accomplishments and open puzzles. Researchers extend calculations, improve instruments, specify boundary conditions, and investigate anomalies. A recalcitrant result initially appears as a difficulty within the prevailing program rather than an immediate refutation. Over time, some anomalies acquire exceptional importance because they resist available conceptual and technical repairs. A revolutionary alternative may then redefine the relevant problem and the expected form of its solution.

Kuhn (1962) describes these patterns in his study of scientific revolutions. The account illuminates why transitions cannot be read simply as accumulating an inventory of neutral statements. A new framework may preserve many successful computations and observations while introducing different explanations, entities, or admissible questions. Describing a revolution as cumulative in one respect and disruptive in another is perfectly coherent.

A significant analytical restriction follows. Crisis is not a universally necessary historical trigger for every conceptual revision, and the episodic scheme is not a chronology that all sciences must obey. Fields also change through instrument innovation, transfers among disciplines, mixed theoretical practices, and gradual adaptation. The historical model helps locate incommensurability when changes in exemplary problems and classifications become substantial. It should not be imposed as a rigid template on every episode. The historian must determine independently which events constituted recognized crises, which changes were retrospective reconstructions, and which alternatives coexisted in actual work.

The Semantic Dimension

Semantic incommensurability concerns relations among the meanings of key theoretical expressions. An identical word may no longer occupy the same conceptual position in competing theories, while distinct words may overlap only partially. The identity of a sound or written label does not guarantee constancy of the term’s inferential, classificatory, or measurement role.

Feyerabend’s 1962 reduction critique argued against treating the relevant terms as invariant when a successor theory changes fundamental relations (Feyerabend, 1962). Later Kuhn developed the more focused claim that certain groups of kind terms cannot be translated individually without disturbing their structural relations. The semantic problem is especially visible when a later text appears to preserve an earlier noun while altering the conditions under which the noun applies.

One should resist two shortcuts. The first is to infer semantic identity from successful numerical approximation: a later theory can reproduce the values of an earlier one under special conditions while giving its quantities a different interpretation. The second is to infer complete semantic isolation from conceptual revision: historians can often identify the earlier term’s inferential role, reconstruct its use, and explain why it was replaced. Semantic history is therefore an investigation of structured correspondences and failures of correspondence, with neither universal equivalence nor universal unintelligibility as its starting point.

The Methodological Dimension

Methodological incommensurability arises when competing traditions differ in how they select relevant problems, rank scientific virtues, or interpret decisive tests. A research community may value precision on established phenomena while a rival emphasizes unification of formerly separate domains. The disagreement can persist even when both parties acknowledge the same instrument reading.

The philosophy of Kuhn’s scientific change describes methodological concerns accompanying his account of revolutions (Bird, 2025). These concerns should be distinguished from the semantic thesis. Scientists can understand one another’s statements yet continue to disagree about what constitutes sufficient explanatory achievement. Conversely, two communities might prefer the same empirical virtues while relying on concepts that resist exact translation.

Methodological comparison is possible whenever the standards themselves can be discussed. Researchers may enumerate desiderata, locate tensions among them, expose the consequences of different rankings, and construct new procedures for adjudication. The absence of a fixed algorithm does not eliminate the possibility of a reasoned verdict. Historical judgments about scientific progress involve such negotiated appraisal, even when the verdict is subsequently stabilized in textbooks. The philosophical challenge is to explain how standards can remain criticizable when their historical application is open to revision.

Perception, Exemplars, and Learned Recognition

The perceptual aspect of incommensurability concerns learned discrimination: trained observers can respond differently to a common situation because their practices direct attention to different patterns. Kuhn’s discussion of scientific perception drew on examples from psychology and on the training required to recognize scientific objects. A physicist and a novice can look at the same display while identifying different significant features.

The historical study of Kuhn explains why exemplars and perception belong to a theory of scientific education rather than a simple assertion that sensory organs change (Bird, 2025). Scientists acquire category-sensitive expectations. They learn which deviations matter, which traces correspond to signals, and which apparatus effects can be disregarded. This learning can be revised during conceptual change.

Perception here requires careful handling. Training can affect observation reports, selective attention, and practical recognition; it does not establish that material objects become unshareable or that every experience is exhausted by theory. The historian is often able to compare diagrams, surviving instruments, notebooks, and laboratory procedures across the change. These traces make perceptual reorganization investigable. What changed can be specified only through the retained material and social practices that support the comparison. The primary burden is to reconstruct those practices without collapsing a historical reclassification into either a change of pure sensation or a purely verbal renaming.

Feyerabend’s Critique of Reduction

Feyerabend opposed the expectation that a successful successor theory must preserve the meanings and laws of its predecessor in order to explain that predecessor’s successes. In standard models of intertheoretic reduction, bridge statements connect vocabularies and the reduced theory follows as a limiting or special case. Feyerabend argued that important historical replacements involve conceptual transformations that challenge such straightforward deduction.

His Explanation, Reduction, and Empiricism explicitly associates conceptual change with the assessment of apparently incompatible theories (Feyerabend, 1962). For him, requiring all new theories to employ the ontology and meanings of accepted theories would artificially constrain exploration. Distinct alternatives could reveal otherwise hidden weaknesses in current empirical and conceptual commitments.

There is a strong insight here that remains independent of Feyerabend’s wider methodological polemic. When an old theory is reconstructed from a later standpoint, the historian must distinguish genuine preservation from retrospective reinterpretation. The later account can explain why earlier calculations worked within a domain without accepting every earlier entity or meaning. This enables a precise question: which explanatory relations survive, under what mathematical transformation, and what has been altered in the interpretation of quantities? A reduction argument that addresses these questions may succeed in a qualified form. A deduction that conceals the changes by stipulating identical meanings merely assumes the disputed historical conclusion.

Theoretical Meaning and Empirical Testing

Theoretical meanings matter for experiment because the interpretation of an observational record depends on assumptions about apparatus, measurement, causal processes, and error. A stable reading of a gauge is compatible with controversy about what physical quantity the gauge measures and which theoretical description explains the outcome.

Feyerabend (1962) explored this problem in his account of empirical comparison. His argument is sometimes summarized too quickly as the impossibility of comparing predictions. The more productive point is that the mapping between raw events, observation reports, and theory-specific predictions may itself require reconstruction. Experimental confrontation remains possible, but a successful confrontation often has more conceptual stages than a neutral observation-language picture permits.

Consider a recorded trajectory in a chamber. Investigators may agree on the image, numerical pixel coordinates, and calibration history while disagreeing about the entities inferred from the track. They can compare detection rates, error models, and predicted patterns even as they debate classification. This example is analytical and generic, not a historical claim about a particular experiment. It displays how comparison proceeds across levels: a relatively shared record, contested interpretation, and theoretical conclusion. The primary analytical error is to mistake one shared level for semantic identity throughout or one disputed level for a failure of all empirical contact.

The Observational Vocabulary Question

Logical empiricist accounts of science treated observation and theoretical description as closely connected problems, with substantial internal disagreement about how the relation should be formalized. The idea of a single vocabulary suitable for translating all scientific claims provided one influential target for critics. Yet no historically accurate reconstruction should treat all empiricists as endorsing a simple, fixed catalog of theory-free impressions.

Carnap’s 1956 discussion of theoretical concepts analyzes theoretical language and observational interpretation in a technically nuanced way (Carnap, 1956). His work shows that the problem of theoretical reference and empirical significance was already being debated within empiricist philosophy. Against this background, Kuhn and Feyerabend sharpened the question of whether theoretical change can modify the interpretive conditions of observation itself.

The useful contrast is between an observation procedure and an observational statement. Procedures can be reproduced and documented while the statements interpreting their products vary. A sequence of instrumental operations can serve as evidence across theories without receiving one final, theory-independent description. Thus the failure of a universal translation language need not nullify the role of observation. It calls for richer records of how data were generated, how their relevance was justified, and how alternative interpretations arise. Methodological transparency becomes more important, not less, when the language of evidence is theoretically structured.

Theory-Ladenness and Its Degrees

Theory-ladenness means that prior concepts and expectations can influence the identification, reporting, or interpretation of observations. Its possible forms include selective attention, instrument calibration, classificatory judgment, background-model correction, and the language of a final published result. These forms have different degrees of revisability and intersubjective accessibility. They should not be compressed into the claim that observation simply repeats an accepted theory.

Kuhn’s treatment of perceptual training in The Structure of Scientific Revolutions and Feyerabend’s critique of theoretical invariance offer connected but distinguishable considerations (Kuhn, 1962; Feyerabend, 1962). Earlier analyses of scientific observation also belong to the historical background. Scientific records contain layered traces: instrument readouts, laboratory logs, calibrated values, theoretical classifications, and published explanations. Each layer may have a different degree of dependence on contested assumptions.

Such layering explains why theory-ladenness and empirical criticism coexist. A critic may challenge a data-processing model while accepting the recorded input, replace the classification scheme while retaining the observation record, or question the instrument’s design through independent calibration. Comparisons are strongest when they identify exactly which interpretive assumptions could affect which result. Incommensurability therefore places an evidentiary burden on both defenders and critics of a theory: neither can establish decisive contact with the world merely by invoking the word “observation.”

From Global Difference to Local Incommensurability

Kuhn’s later writing revised the scale and clarity of his account. The issue became less a comprehensive difference between entire ways of living and more a localized difficulty in translating certain interdependent scientific kind terms. Scientific languages can overlap extensively, and much ordinary communication continues throughout major changes. What resists direct translation is a restricted set of classifications embedded in different lexical arrangements.

In Commensurability, Comparability, Communicability, Kuhn (1983) distinguished successful comparison from strict translatability. The distinction is decisive. A bilingual historian may learn two technical languages, represent the theories’ assumptions, and evaluate their successes while lacking a perfect word-for-word substitution preserving every original relation. The later reconstruction of incommensurability explains how this local account differs from the more encompassing early thesis (Oberheim and Hoyningen-Huene, 2024).

This historical development discourages retrospective uniformity. It would distort Kuhn’s trajectory to read the 1962 text as if it had already formulated the entire later taxonomy. Equally, the local thesis does not erase the earlier concern with scientific practice and the evaluative changes accompanying revolutions. Each version asks a different question. The early work studies how a scientific community’s problem-world can transform; the later account identifies linguistic structures implicated in such transformation. Their relation is interpretive and developmental. It must be investigated as an evolution in Kuhn’s thought rather than converted into either a seamless continuity or an outright repudiation.

Taxonomy and the No-Overlap Principle

Kuhn’s later taxonomic discussion identifies a specific problem involving kind categories. A lexicon groups objects and processes into kinds, and those groupings constrain which generalizations can be formulated without contradiction. The no-overlap principle, as Kuhn described it, concerns the intersection of kind categories within a structured taxonomy; some relations among categories that are admissible in one scheme are inadmissible in another.

His The Road since Structure essay develops this vocabulary of taxonomy and kind terms (Kuhn, 1991). The argument gains significance because taxonomic revision can alter the referential possibilities of a theory even when empirical investigations remain interconnected. Suppose a prior vocabulary treats a collection of phenomena as members of one kind, while a later theory repartitions them into distinct kinds according to different explanatory relations. An attempt to replace each old term with a single new term can leave out the inferential organization that made the earlier statements meaningful.

The no-overlap language should be read in its technical context. It is not a universal claim that no scientific categories ever overlap or that all kinds exclude one another. Taxonomies ordinarily organize hierarchical and disjoint relations among appropriately specified categories. The contested point is whether two particular category systems can be merged without violating those internal constraints. This makes semantic incommensurability a tractable research topic: the historian can identify relevant kinds, reconstruct the rules of application, and test whether a proposed mapping preserves those rules. The outcome may be a partial mapping, a many-to-one translation, an explanatory paraphrase, or a documented failure of exact translation.

Translation, Interpretation, and Language Learning

Translation aims to carry statements between languages with enough preservation to sustain relevant uses. Interpretation explains what an expression does in a practice, and language learning enables participation in that practice. Kuhn increasingly emphasized the differences among these activities. An interpreter may learn a historical scientific lexicon and successfully understand its statements without finding identical counterparts for all the lexicon’s categories in present-day science.

Kuhn’s account of comparability argues that comparison need not depend on sentence-by-sentence translation into one neutral scientific language (Kuhn, 1983). His later reflections on lexicons develop the role of acquisition and interpretation (Kuhn, 1991). These arguments illuminate a concrete historiographic task. When describing an older term, the scholar can reproduce the circumstances of use, explanatory connections, instructional examples, and canonical disputes. Such contextual interpretation may be more faithful than importing a contemporary English synonym.

The distinction also clarifies the relation between scientific understanding and semantic skepticism. Learning an alternative language is itself a way of understanding it. Accordingly, a claim that a complete translational equivalence is unavailable cannot simultaneously be taken as evidence that the historical language is entirely unintelligible. The point is about the losses involved in reducing the unfamiliar conceptual system to the familiar one. When a historian explicitly preserves those losses in commentary, comparisons become better specified, and their limits become visible. Genuine historical comprehension requires labor; it is not delivered by a universal dictionary.

Davidson’s Challenge

Donald Davidson questioned the philosophical coherence of the radical conceptual-scheme idea when that idea relies on the possibility of a wholly untranslatable but nonetheless recognized alternative system of thought. The problem is methodological as well as semantic: by what evidence could an observer establish that another speaker has meaningful beliefs while denying the interpretability of all of the speaker’s discourse?

In On the Very Idea of a Conceptual Scheme, Davidson (1974) attacked the dualism between an organizing scheme and a neutral uninterpreted content. This challenge presses directly on expansive forms of incommensurability. If there is sufficient basis to identify a community as holding beliefs about a shared world, the evidence for this identification may already provide substantial grounds for interpretation. The assertion of complete conceptual separation risks dissolving its own evidence.

Davidson’s criticism has considerable force against total unintelligibility. It need not defeat Kuhn’s more local thesis about particular kind terms. A historian can interpret most of an earlier theory while finding no structure-preserving translation for some expressions. Davidson’s argument therefore helps delimit, rather than simply abolish, a coherent thesis. It requires supporters of incommensurability to identify specific obstacles and to avoid inferring global isolation from local linguistic failure. The two philosophers may still disagree over whether the scheme-content distinction is indispensable to the taxonomic account. The strongest exchange asks what interpretive successes and failures would substantiate each claim in actual scientific histories.

Reference, Continuity, and Kitcher’s Response

The debate about incommensurability includes questions of reference: whether terms used in earlier and later theories pick out the same things. A term’s role in a theory can change substantially while scientists continue to investigate many of the same material structures. Conversely, the apparent continuity of a label can obscure a historical change in what its users considered an instance of the kind.

Philip Kitcher challenged oversimplified accounts of reference change in Theories, Theorists and Theoretical Change (Kitcher, 1978). Historical uses of a term may refer through different mechanisms and intentions, producing a mixed picture of continuity and discontinuity. The intellectual advantage of this approach is that reference becomes an empirical-historical issue to investigate rather than a consequence imposed by a general philosophy of language.

Consider a term associated with an obsolete substance theory. A contemporary historian may reconstruct its role in the older explanatory network and also identify certain processes later explained under a different ontology. Some earlier uses can track real phenomena while other uses fail to refer as intended. Such findings do not automatically validate the earlier theory, yet they prevent a blanket verdict that the older community spoke about nothing. The relevant analysis separates what scientists attempted to refer to, the causal circumstances of their observations, their characteristic descriptions, and the later mapping proposed by historians. This distinction preserves the intelligibility of learning from false theories while acknowledging genuinely transformed ontologies.

Criteria of Scientific Appraisal

Scientific appraisal uses several standards: empirical accuracy, internal consistency, fit with established successful practices, explanatory scope, simplicity, fertility, and prospects for new investigation. These standards can conflict. A theory may be more accurate in one domain while proving less economical in another; a new approach may explain previously unrelated phenomena while initially performing poorly on some familiar benchmarks.

The question is whether these differences produce irrational theory choice. The Kuhn literature on theory choice provides a contrary picture: scientists can invoke shared values and reach divergent reasonable judgments (Kuhn, 1977). Disagreement about values and their weights does not erase the public argument that makes a decision accountable.

No single vector of scores automatically represents every historically relevant consideration. Assigning numerical rankings may clarify a particular choice, but the legitimacy of the weights remains itself a subject of scientific and philosophical argument. Rational appraisal requires the specification of consequences: what an investigator sacrifices by retaining an older framework, what a new framework makes possible, and how those judgments depend on the present state of evidence and research priorities. This reasoning permits comparative verdicts while treating algorithmic completeness as an additional claim needing defense. Incommensurability sharpens the question of appraisal by demanding a record of exactly which standards change and which remain active across the contested transition.

Kuhn’s Values and Rational Disagreement

Kuhn’s essay “Objectivity, Value Judgment, and Theory Choice” identifies familiar shared scientific values, including accuracy, consistency, scope, simplicity, and fruitfulness. He argues that values guide judgment while allowing differing applications and priorities. A scientist’s personal and communal commitments consequently influence which promising solution deserves work, but they do so in relation to publicly discussable considerations.

The collection The Essential Tension gathers these arguments and locates them within Kuhn’s broader account of tradition and innovation (Kuhn, 1977). Their significance for incommensurability is direct. Rival scientists can understand the same words “accurate” and “simple” while estimating different domains of performance or ranking benefits differently. A change in conceptual organization may thus coincide with rational disagreement even where major evaluative ideals remain shared.

The resulting view has a distinct epistemic discipline. An appraisal should specify the standards used, document the evidential basis of their application, and acknowledge losses alongside gains. It should also permit retrospective reconsideration as evidence changes. Such discipline does not make disagreement disappear and cannot guarantee convergence. It explains how disagreement can remain answerable to reasons. The historian’s task is to recover actual evaluative practices rather than retroactively attribute one unanimous rule to a community. Science advances through judgments carried out in situated settings, assessed through subsequent inquiry and the durable scrutiny of other scientists.

Lakatos and Research Programs

Imre Lakatos sought to describe scientific rationality through the comparative development of research programs. Rather than demand that a theory be rejected immediately upon encountering an anomaly, he examined sequences of theories associated with a relatively stable core of commitments and changing auxiliary hypotheses. A progressive program yields novel, corroborated developments; a degenerating one relies increasingly on retrospective accommodations.

In Falsification and the Methodology of Scientific Research Programmes, Lakatos (1970) offers a rival normative perspective on historical scientific change. His analysis supplies longer time horizons than a single theory–observation confrontation. A program that initially looks weaker may develop resources that transform its competitive position.

Against incommensurability, this view suggests a way to compare sequences of research achievements without requiring that rival programs have identical languages or accept all the same problems. Historical assessment can reconstruct the novel results produced, the adjustments undertaken, and their evidential fortunes. Lakatos’s criteria are themselves open to debate, especially concerning what counts as genuinely novel and when a verdict becomes warranted. Nonetheless, they show that conceptual difference and rational comparative inquiry are compatible. The analyst should distinguish the question of semantic mapping from the question of whether a sustained research tradition delivers explanatory and predictive advances over time.

Laudan and Changing Aims

Larry Laudan treated the assessment of scientific theories and traditions in relation to the problems they solve and the aims they pursue. Methodological standards and epistemic objectives can themselves become subjects of rational criticism. This approach places theory change within a broader pattern of reciprocal adjustment among theories, methods, and goals.

Laudan’s Science and Values examines disagreements about scientific aims and their relation to methodological judgment (Laudan, 1984). The work is important because it prevents a false dilemma: either all scientists employ timeless standards or changes in standards are immune to appraisal. A proposed research aim can be criticized for its feasibility, its relation to available methods, and its consequences for inquiry.

Incommensurability therefore must be tested at more than one level. Different theories may use different classifications; different researchers may pursue different problems; methodological commitments may vary as well. The historian can ask which changes were explicit and contested, which were tacit, and which supplied productive criticism of the older practice. Laudan’s analysis provides tools for studying rational changes in the objectives of inquiry without presuming one universal measure. It also reminds readers that scientific goals have plural histories. A historically persuasive comparison cannot substitute the observer’s favorite present-day goals for all the goals operative in the earlier transition.

Local Comparison and Historical Reconstruction

A local comparison begins by specifying the exact object of disagreement. Is the conflict about an equation, a category of natural phenomena, a measurement protocol, the acceptability of an explanation, or the values guiding a research agenda? The inquiry then determines what records and competencies make a comparison possible. It may require learning the older vocabulary, reconstructing an experiment, and distinguishing the later theory’s retrospective interpretation from the earlier scientist’s own description.

The historical reconstruction of scientific revolutions by Hoyningen-Huene (1993) supplies an especially important guide to the changing structure of Kuhn’s account. The Stanford discussion of scientific revolutions surveys the broader historiographic and philosophical challenges. These sources support an approach that locates continuity and discontinuity in specific scientific practices rather than in a single unqualified assertion of total change.

Consider a hypothetical comparison of two classification schemes for the same collection of laboratory samples. A historian can preserve original labels and laboratory protocols, then document how each scheme partitions the specimens. The resulting mapping may show that one old kind crosses the boundaries of two newer kinds. It is possible to calculate agreement on individual specimens without pretending that the category systems are semantically interchangeable. This is a constructed illustration rather than a report of historical research. Its lesson is methodological: overlapping empirical access provides a basis for analysis, while the limits of translation reveal what the competing classifications organize differently.

Case Study: Copernican Astronomy

The Copernican transformation is a central example in Kuhn’s historical writing, but the shift should not be described as a moment when previously known planets abruptly vanished from observation. It altered the organization of astronomical kinds, the relation of Earth to the heavens, and the explanatory significance of celestial motions. In a geocentric organization Earth occupied a privileged stationary position and the Sun could be classified among moving heavenly bodies. In a heliocentric account Earth became a planet, while the Sun occupied a different classificatory role.

Kuhn (1991) explicitly discusses shifts in the use of “planet” in The Road since Structure. The historical reassessment of Copernican change by Barker (2021) helps prevent a simplistic reading of a long, technically complex revolution. The reclassification of Earth and Sun mattered because it reorganized accepted relations among bodies and the explanations of their appearances. It was not merely a new list of labels.

At the same time, astronomical data, calculation techniques, and observable positional relations gave rivals grounds to argue across their systems. Early Copernican astronomy did not instantly outperform every competitor on every predictive criterion. The historical evidence must be traced through specific astronomical works, instruments, and changing standards rather than inferred from the ultimate victory of heliocentrism. The case illustrates the distinctive logical structure of incommensurability: a taxonomic discontinuity can occur together with overlapping observational constraints and sustained mathematical comparison. The older practice remains historically intelligible precisely because its conceptual distinctions can be reconstructed.

Case Study: Oxygen and Phlogiston

The chemical revolution involved competing interpretations of combustion, calcination, gases, and material composition. Phlogiston-based approaches and Lavoisier’s oxygen chemistry differed in the entities and reactions through which they explained observed changes. The older vocabulary addressed real experimental problems, including changes observed when metals were heated or substances burned. The emerging chemistry reorganized the understanding of substances, elements, and the role of gases in chemical processes.

The Stanford survey of philosophy of chemistry discusses the conceptual and historiographic significance of chemical categories (Weisberg, Needham, and Hendry, 2026). Chang’s historically informed study of water also shows why the history of chemical concepts warrants sustained scrutiny beyond a one-step narrative of replacement (Chang, 2012). One must avoid presenting Lavoisier’s reinterpretations as if every contemporary chemist immediately accepted the same set of results or as if the old community lacked empirical rigor.

Mass measurements, experimental procedures, apparatus, and gas-related phenomena provided material points of connection even as the theoretical interpretation of those phenomena shifted. An early analyst might explain an observation with reference to phlogiston, while a later analyst redescribed the process through oxygen and combination. The concept of “element” itself underwent historical development. Incommensurability in this case is best located in the organization of explanatory kinds and the interpretation of reactions, with shared experimental practices as part of the history. The case supports semantic discontinuity and rational comparison simultaneously; the evidence does not require an absolute gulf between old and new laboratories.

Case Study: Newtonian Mechanics and Relativity

The passage from Newtonian mechanics to special relativity illustrates conceptual discontinuity because the later theory changes the structural roles of simultaneity, time, and spatial measurement. Under Newtonian assumptions, a universal temporal parameter organizes motions throughout space. In special relativity, the assignment of temporal and spatial intervals depends on relations among reference frames; invariant spacetime relationships replace expectations about absolute simultaneity.

Einstein’s 1905 electrodynamics paper presents foundational reasoning about simultaneity and the electrodynamics of moving bodies (Einstein, 1905). Later general-relativistic ideas should not be projected wholesale into that argument. Many calculations conducted with Newtonian mechanics remain extraordinarily accurate in appropriate low-speed domains. This mathematical relationship does not establish complete semantic identity between the two theories.

Two conclusions coexist. Relativistic mechanics recovers Newtonian predictions under suitable limiting conditions, supplying a mathematical bridge for evaluating empirical performance. Yet the interpretation of time and simultaneity changes, so the limit does not mean Newtonian time is literally identical to the relativistic theory’s frame-dependent quantities. Successful correspondence is established through mathematical and physical argument. It should be documented alongside conceptual differences. Incommensurability describes part of the change; the limiting relation explains part of the comparison.

Instrumentation and Shared Work

Scientific practice contains durable material features that can outlast changes in theoretical interpretation. Instruments, photographs, specimens, calibration protocols, and laboratory notebooks support comparison across theoretical traditions, although their use remains dependent on assumptions. Historians can reconstruct the chain from material trace to recorded quantity and from quantity to theoretical claim.

Peter Galison’s Image and Logic examines the material cultures of microphysics and the coordination of heterogeneous practices (Galison, 1997). His account of trading zones shows how groups coordinate work without complete theoretical agreement or one unified language. Shared practice can therefore precede full semantic commensurability.

A shared instrument is not automatically a universal common measure. Different investigators may calibrate it differently, reject different signals as noise, or disagree about what entity it detects. Yet these differences can be explored by examining operational histories: Were procedures equivalent? Which assumptions governed conversion of signals into quantities? Which records could be reprocessed under a rival model? Such questions transform a declaration of conceptual barriers into an analysis of how barriers are traversed or exposed. Material continuity often provides a disciplined beginning for historical comparison.

Problem Loss, Problem Gain, and Progress

Scientific revolutions sometimes resolve difficult problems and also displace or temporarily lose solutions central to an earlier tradition. This history complicates a picture of progress as the simple addition of propositions to a growing inventory. Conceptual replacements rearrange which tasks count as important, and a problem once central may cease to occupy the same position.

Kuhn’s historical account of progress emphasizes scientific advances in problem-solving capacities (Kuhn, 1962). Laudan’s study of scientific values provides a complementary way of evaluating aims and traditions (Laudan, 1984). A credible assessment must describe concrete gains, identify losses, and explain changes in the importance assigned to problems.

The presence of losses does not invalidate progress. It means that comparative judgments require a specified domain and rationale. A new theory may unify observations, permit new experimental interventions, and explain why older calculations succeeded under restricted conditions. These achievements warrant a qualified judgment of progress when supported by documented evidence. A retrospective account should also preserve costs and unresolved questions rather than impose a story of inevitability. Incommensurability strengthens historical assessment by making the purposes of comparison explicit.

Realism, Reference, and Historical Worlds

Scientific realism asks whether successful theories approximately describe a mind-independent world, including entities beyond ordinary observation. Incommensurability challenges realism because scientific kinds and inferential relations change over time. If theories transform their classifications, the question becomes how reference and truth can persist through that history.

The posthumous Kuhn writings contain unfinished work on incommensurability, progress, and realism (Kuhn, 2022). Their editorial status matters: an unfinished manuscript documents continuing inquiry rather than a finalized solution. Hoyningen-Huene’s Reconstructing Scientific Revolutions places the realist problem within Kuhn’s changing conception of scientific worlds (Hoyningen-Huene, 1993).

There are several independent realist tasks. One must explain how the world constrains competing theories, how inquiry can retain contact with causal structures through changing descriptions, and what type of continuity the historical record actually establishes. A realist may grant local semantic change while defending robust experimental or causal continuity. A critic may insist that such continuity becomes identifiable through the successor theory’s categories. Neither verdict follows automatically from the label “incommensurable.” A substantive argument demands actual cases, referential practices, and an account of how instruments and interventions bear on existence claims.

Rationality and Relativism

If scientific standards and meanings change, what warrants a claim that one theory is better than another? Three propositions must be distinguished: there is no universal algorithm ranking all theories; some scientific expressions resist exact translation; every evaluative outcome is equally defensible. The third proposition does not follow from the first two.

Sankey’s The Incommensurability Thesis surveys major interpretive difficulties (Sankey, 1994). Friedman’s Dynamics of Reason examines transformations in conditions of scientific rationality (Friedman, 2001). Their arguments invite scrutiny of how reasons operate when scientific concepts and methodological commitments change.

A rational verdict is supported by considerations open to examination, challenge, and revision. It does not require that the criteria be timeless or weighted through a mechanical rule. The pertinent issue is whether investigators articulate their evidence, represent alternatives fairly, and explain the consequences of methodological choices. Historically formed norms are open to criticism for their explanatory and empirical effects. Radical relativism would need to establish the stronger claim that no rational basis for cross-tradition appraisal exists. Documented scientific comparisons provide substantial grounds against such universal skepticism.

Pluralism and the Continued Life of Alternatives

Feyerabend connected criticism of universal methodological rules with the defense of theoretical pluralism. If accepted theories influence what counts as evidence, competing approaches may expose assumptions otherwise hidden. Preserving alternatives can therefore be productive even when those alternatives initially seem far from established conceptual practices.

His Against Method develops methodological pluralism (Feyerabend, 1975), while Science in a Free Society extends the argument to scientific authority and public life (Feyerabend, 1978). These are related but distinct interventions.

Pluralism entails obligations. Alternative theories must undergo empirical elaboration, conceptual criticism, and scrutiny of evidential results. The usefulness of preserving a minority approach does not establish its truth. Nor does criticism of one universal methodological formula entail that every procedure succeeds in every scientific setting. The defensible insight concerns the productive possibility of conceptual variety and the intellectual hazard of excluding it in advance. Incommensurability can impede rapid appraisal and simultaneously uncover new questions. Responsible plurality maintains room for exploration alongside demanding standards of argument.

Incommensurability and Scientific Education

Scientific instruction transmits practices as well as propositions. It teaches students to recognize exemplary problems, operate instruments, separate signals from noise, and assess whether a solution is sufficiently precise or promising. Kuhn placed training near the center of his account of disciplinary stability because shared examples coordinate research without explicit agreement about every assumption.

The analysis of Kuhn’s philosophy explains why exemplars and normal science matter for education (Bird, 2025). A textbook may depict an older theory as a rough version of a present one. This approach can be pedagogically convenient while concealing the earlier tradition’s distinctive explanatory aims.

Historical instruction attentive to incommensurability reconstructs earlier problems and terms before evaluating their solutions. Students can investigate the resources available to historical actors, the experimental settings guiding them, and the reasons that made particular alternatives credible. Such exercises cultivate comparative competence: understanding a rejected system well enough to criticize it accurately. That competence is independent of endorsement. It improves current scientific self-criticism by revealing that accepted classifications emerged through experimental, conceptual, and institutional decisions.

Historical Method and Source Discipline

Philosophical history can be distorted when a later category is imposed on earlier actors without examining their evidence. Calling an episode “incommensurable” requires documentation of differences in language, classification, procedure, or standards. An episode should not receive the label simply because later textbooks describe it as revolutionary.

Hoyningen-Huene’s historical reconstruction of Kuhn clarifies many interpretive difficulties (Hoyningen-Huene, 1993). Chang’s history of chemical knowledge demonstrates the importance of investigating earlier practices on their own terms (Chang, 2012). Historical inquiry needs to consider disagreements about an episode, not merely repeat one familiar anecdote.

Three evidential levels must be separated. Primary documents establish how scientists formulated claims or described operations. Secondary scholarship contextualizes documents and assesses competing historical interpretations. Contemporary philosophical analysis draws arguments about conceptual relationships. A publisher catalog establishes a work’s identity and declared scope; it does not independently verify every assertion in the work. A digitized primary essay can support close reading, while specialized claims about particular experiments may require laboratory records or additional archival scholarship. The companion evidence ledger therefore specifies whether a source was consulted directly as text, through a scholarly entry, or through bibliographic metadata. The force of a conclusion must remain proportionate to the evidence actually examined.

A Constructed Comparison Exercise

Imagine two fictional scientific groups investigating recurrent oscillations in a complex material. The first classifies the oscillations by visible modes and predicts the recurrence of established kinds. The second classifies them through an inferred state-space topology and predicts some of the same instrument records through transitions between differently defined states. Their categories overlap at the level of recorded traces, but one old kind maps onto several new ones.

Investigators could compare recurrence estimates on shared recordings, preserve distinct category labels, and document where direct translation fails. They may discover that the second account predicts an intervention the first cannot articulate. They might also find that the first system addresses an engineering task more efficiently. Both claims can be rationally tested even if a one-to-one dictionary between the categories is unavailable.

This constructed example reports no historical experiment and introduces no original theory. It illustrates four conclusions drawn from the preceding research: shared records can anchor criticism without fully determining their theoretical meaning; categorical divergence permits task-specific comparison; a verdict depends on stated evaluative purposes; and a scholar must avoid silently reconstructing an earlier vocabulary in later terms. The exercise makes the practical significance of local incommensurability visible without turning the observation into a new conceptual system.

The Scope of the Thesis

A defensible claim of incommensurability identifies the relevant theories, the relation between them, and the precise transfer that fails. The problem may involve lexical translation, classificatory organization, reference, perceptual training, standards of evidence, or explanatory priorities. Its strength increases when the same analysis specifies what remains comparable.

The scholarly survey of incommensurability shows how broad early formulations developed into more local semantic questions (Oberheim and Hoyningen-Huene, 2024). Kuhn’s later account is especially suitable for a precise linguistic claim, while the early discussion of scientific practice remains important for historical research. These levels should not be shifted interchangeably merely to produce an exaggerated conclusion.

Failure of exact translation does not prove failure of interpretation. Absence of an automatic theory-choice rule does not prove absence of reasons. Changes in scientific values do not prove arbitrariness. Mathematical reduction under limiting conditions does not prove complete semantic identity. These distinctions are the logical core of a disciplined inquiry. Incommensurability has explanatory force when its object and extent are identified, and when the record shows why the particular relation resists exact preservation.

Original Analytical Synthesis

The evidence supports a unified interpretive judgment: incommensurability is most informative as a specified, localized difficulty in preserving a relation through scientific change. The relation may concern semantic translation, taxonomy, evaluation, or the interpretation of measurements. These different objects demand different forms of evidence. The question is which comparative operations are supported by historical research, rather than whether scientists possess one eternal neutral language.

Copernican astronomy illustrates changed celestial categories amid overlapping astronomical constraints. The chemical revolution illustrates a reorganization of explanatory kinds alongside experimental continuities. Relativity illustrates transformed concepts of time and simultaneity together with precise limiting relationships. Their distinctive structures matter more than the convenience of describing all three as paradigm shifts. A responsible account identifies what changed, which correspondences survived, and how particular scientific judgments were warranted.

Comparison improves when original terms, procedures, standards, and records are preserved. Investigators can map agreements and differences, distinguish exact equivalences from approximations, and disclose assumptions introduced by retrospective interpretation. Such documentation permits a philosophical account of scientific rationality that recognizes discontinuity while maintaining empirical criticism.

This synthesis is original in the scholarly sense of independently assembling and evaluating a sustained argument from identified sources. It asserts neither priority for discovering partial comparability nor a newly invented doctrine of scientific change. Its contribution is to distinguish semantic, methodological, and historical claims consistently, then apply those distinctions to several differently structured scientific transformations.

Conclusion

Incommensurability matters because scientific change can transform categories, evidential practices, standards of explanation, and trained habits of recognition. Kuhn’s historical and taxonomic arguments and Feyerabend’s critique of reduction illuminate different dimensions of this transformation.

The decisive distinction is between lacking one universal metric for ranking theories and lacking every rational basis for their comparison. These are different claims. Researchers can compare predictions, reconstruct operations, learn earlier categories, examine standards, and assess problem-solving achievements while acknowledging that some technical expressions resist exact translation. Their conclusions remain open to argument and revision.

The Copernican, chemical, and relativistic transitions exhibit different combinations of continuity and rupture. Conceptual discontinuity must be established through specific texts and practices; comparison must likewise be justified with specified aims and evidence. Neither complete identity nor total separation should be presumed.

Incommensurability is consequently a definite research question about the limits of carrying meanings, classifications, and standards across scientific history. Its continuing philosophical value consists in clarifying those limits and explaining how inquiry can compare its alternatives while changing the concepts through which it understands the world.