Two Dogmas of Empiricism

Two Dogmas of Empiricism Willard Van Orman Quine

Tag layer of the DHSS Meta-Tagging Project. The article text is reproduced from the staged copy named in sourceDesc and is not altered; the tags are held separately in standOff so that overlapping and nested spans are preserved.

Two Dogmas of Empiricism 1951 The Philosophical Review 10.2307/2181906 quine1951-two-dogmas quine1951-two-dogmas.txt, the project's staged copy 1a2f7b1a20ee8676bb206d5e4584f288d090ab2d open-access philosophy

the paper's own words

the tag layer; a sentence written by the tagger about the paper, always anchored to a quotation but not itself a quotation

0 spans located by searching the stored text for the quote's exact words, allowing any whitespace between them The empiricist distinction between analytic and synthetic truths and the doctrine of reductionism. Quotation withheld because this record is not cleared for redistribution. Can the analytic/synthetic distinction and reductionism withstand scrutiny, and what follows for empiricism if they cannot? Quotation withheld because this record is not cleared for redistribution. Attacks the two dogmas - the analytic/synthetic distinction and reductionism - arguing for confirmation holism, reshaping twentieth-century analytic philosophy. Quotation withheld because this record is not cleared for redistribution. critique Quotation withheld because this record is not cleared for redistribution. the analytic-synthetic distinction Quotation withheld because this record is not cleared for redistribution. conceptual-analysis conceptual-analysis Quotation withheld because this record is not cleared for redistribution. Each proposed explanation of analyticity is shown to presuppose a notion that is itself no clearer than analyticity, so the explanations travel in a circle rather than terminating Quotation withheld because this record is not cleared for redistribution. Definition cannot ground synonymy because the lexicographer only records a synonymy that usage already established Quotation withheld because this record is not cleared for redistribution. Interchangeability salva veritate is too weak in an extensional language and question-begging in a language rich enough to contain 'necessarily' Quotation withheld because this record is not cleared for redistribution. Appeal to semantical rules fails because 'semantical rule' is itself undefined except by where the phrase is printed Quotation withheld because this record is not cleared for redistribution. Because statements meet experience only collectively, any statement can be preserved by revising others, so no statement is analytic in the sense of being immune Quotation withheld because this record is not cleared for redistribution. The two dogmas support each other: only if confirmation attaches to statements one at a time does a vacuously confirmed statement make sense Quotation withheld because this record is not cleared for redistribution. the notion of analyticity resists a clear boundary; borderline cases remain undecidable what this limits: the applicability of the analytic/synthetic distinction to real cases: I DO NOT KNOW WHETHER THE STATEMENT 'EVERYTHING GREEN IS EXTENDED' IS ANALYTIC Quotation withheld because this record is not cleared for redistribution. Technical problems of ambiguity and homonymy for the proposed formulation of cognitive synonymy are set aside rather than solved. what this limits: the proposed formulation of cognitive synonymy: CERTAIN TECHNICAL QUESTIONS ARISE ... OVER CASES OF AMBIGUITY OR HOMONYMY; LET US NOT PAUSE FOR THEM - set aside, not resolved Quotation withheld because this record is not cleared for redistribution. Not every explanation of analyticity available in the literature is explicitly addressed. what this limits: the completeness of the survey of rival accounts: NOT ALL THE EXPLANATIONS OF ANALYTICITY KNOWN TO CARNAP AND HIS READERS HAVE BEEN COVERED EXPLICITLY Quotation withheld because this record is not cleared for redistribution. The key notion of cognitive synonymy is left uncharacterised within the paper itself. what this limits: the argument's central term: JUST WHAT THIS IS CANNOT BE SAID WITHOUT SUCCESSFULLY FINISHING THE PRESENT STUDY - cognitive synonymy is left uncharacterised inside the paper that turns on it Quotation withheld because this record is not cleared for redistribution. The paper argues no boundary has been drawn, not that one provably cannot be - the case is the repeated failure of every proposed criterion what this limits: the strength of the paper's own conclusion: A BOUNDARY BETWEEN ANALYTIC AND SYNTHETIC STATEMENT SIMPLY HAS NOT BEEN DRAWN - has not, not cannot, and it is cited for the stronger claim Quotation withheld because this record is not cleared for redistribution. Germaneness of a statement to particular experiences is left as a loose association rather than a defined relation what this limits: the notion of germaneness: I ENVISAGE NOTHING MORE THAN A LOOSE ASSOCIATION REFLECTING THE RELATIVE LIKELIHOOD, IN PRACTICE - a loose association, not a defined relation Quotation withheld because this record is not cleared for redistribution. The positive account of what does guide theory revision is left at the level of conservatism and simplicity, not a criterion what this limits: the positive account of theory revision: CONSERVATISM FIGURES IN SUCH CHOICES, AND SO DOES THE QUEST FOR SIMPLICITY - named, not specified Quotation withheld because this record is not cleared for redistribution. The author's own inability to classify a sample statement is offered as evidence about the term rather than about his grasp of it, which is asserted rather than shown what this limits: what the classification failure is evidence about: THE TROUBLE IS NOT WITH 'GREEN' OR 'EXTENDED,' BUT WITH 'ANALYTIC' - the fault is the category, not the case Quotation withheld because this record is not cleared for redistribution. using the physics image of a field of force (with boundary conditions) to model the holistic web of scientific knowledge and its contact with experience Quotation withheld because this record is not cleared for redistribution. cognitive synonymy cognitive synonymy Quotation withheld because this record is not cleared for redistribution. logical truth logical truth Quotation withheld because this record is not cleared for redistribution. state-description state-description Quotation withheld because this record is not cleared for redistribution. interchangeability salva veritate interchangeability salva veritate Quotation withheld because this record is not cleared for redistribution. extension extension Quotation withheld because this record is not cleared for redistribution. physical objects physical objects Quotation withheld because this record is not cleared for redistribution. the gods the gods Quotation withheld because this record is not cleared for redistribution. EVIDENCE CORRECTED (PARAPHRASE CORRECTED - THE KUMAR2016 HALF OF THIS ROW WAS A MISREADING, AND IT IS THE PROJECT'S OWN FAULT COMMITTED BY ME) · analytic → synthetic EVIDENCE CORRECTED: PARAPHRASE CORRECTED - THE KUMAR2016 HALF OF THIS ROW WAS A MISREADING, AND IT IS THE PROJECT'S OWN FAULT COMMITTED BY ME. The row paired quine1951 with Kumar2016 as two papers asserting THESE TWO CANNOT RELIABLY BE TOLD APART. The Quine half stands. The Kumar2016 half does not: I quoted 'the regressors for the maintenance and encoding phases may be correlated' and 'may thus be contaminated' and called it a concession. READING THE WHOLE SPAN, THOSE CLAUSES ARE THE SET-UP FOR A CONTROL THE AUTHORS THEN PASS: 'In our study, because the maintenance duration was much longer than the encoding phase, THE CORRELATION COEFFICIENT between the encoding and maintenance regressors in the design matrix WAS VERY SMALL (0.04).' They name a generic risk in order to show it does not apply to their design. So Kumar2016 asserts SEPARABILITY, with a number - and 'encoding --distinct-from--> maintenance' is SOUND, not a workaround for a missing negative type. Every quote in the original was verbatim and the interpretation was still wrong, which is the fault this project exists to expose: THE CONCERN QUOTED AND THE RESOLUTION DROPPED. Found by reading the full span for the sameness-no-contrast queue. — THE BOUNDARY BETWEEN THESE TWO IS NOT DRAWABLE - the NEGATION of 'distinct-from', and no type we have can express it. All eight types assert that a relation HOLDS; not one can say a relation FAILS, so a paper whose contribution is that a distinction cannot be drawn has no home, and the nearest available move records the distinction as holding. That is exactly what the retracted row of this record did: the paper's DENIAL of the boundary quoted as evidence FOR the boundary. SECOND ATTESTATION, in another field and with no shared vocabulary: Kumar2016 concedes in an fMRI methods paragraph that 'the regressors for the maintenance and encoding phases may be correlated' and 'the activity observed in the maintenance phase may thus be contaminated' - the authors warning they may not reliably tell the two apart, recorded in our corpus as 'encoding --distinct-from--> maintenance'. Strip the subject matter and the two papers assert the same relation: THESE TWO CANNOT RELIABLY BE TOLD APART, one argued in philosophy in 1951 and one conceded in a methods note. Neither cites the other and they share no words. Kumar2016 is fenced and is cited here, not touched. Two attestations START A COUNT, they do not license a ninth type. Quotation withheld because this record is not cleared for redistribution. analytic → synthetic THE BOUNDARY BETWEEN THESE TWO IS NOT DRAWABLE - the NEGATION of 'distinct-from', and no type we have can express it. All eight types assert that a relation HOLDS; not one can say a relation FAILS, so a paper whose contribution is that a distinction CANNOT BE DRAWN has no home, and the nearest available move records the distinction as holding. That is exactly what the retracted row of this record did: the paper's DENIAL of the boundary quoted as evidence FOR the boundary. REPLACES unhoused_relations[0], WHICH WAS RETRACTED FOR A CLAIM I ADDED AND GOT WRONG. That row paired this with Kumar2016 as a cross-field attestation - two papers, no shared vocabulary, both asserting that two things cannot reliably be told apart. THE KUMAR2016 HALF WAS A MISREADING: I quoted 'the regressors ... may be correlated' and 'may thus be contaminated' as a concession, when reading the full span shows those clauses SET UP A CONTROL THE AUTHORS THEN PASS - 'the correlation coefficient ... was very small (0.04)'. Kumar2016 asserts SEPARABILITY, with a number. The Quine claim here is unaffected and stands on its own span; only the cross-reference was wrong, and it is not repeated here. Quotation withheld because this record is not cleared for redistribution. analytic - first class, logical truth: true and staying true under every reinterpretation of everything except the logical particles Quotation withheld because this record is not cleared for redistribution. analytic - second class, synonymy-dependent: reducible to a logical truth only by putting synonyms for synonyms, and this is the class the paper says cannot be explained Quotation withheld because this record is not cleared for redistribution. analytic - Kant's sense: the subject already conceptually contains the predicate, a formulation the paper faults for assuming subject-predicate form Quotation withheld because this record is not cleared for redistribution. analytic - Carnap's sense: true under every state-description, which the paper argues reconstructs only logical truth Quotation withheld because this record is not cleared for redistribution. analytic - Leibniz's sense: true in all possible worlds, i.e. incapable of being false Quotation withheld because this record is not cleared for redistribution. analytic - the self-contradiction sense: a statement whose denial is self-contradictory, rejected as no clearer than what it explains Quotation withheld because this record is not cleared for redistribution. synonymy - cognitive: the only sense the argument needs, explicitly not identity of psychological association or poetic quality Quotation withheld because this record is not cleared for redistribution. definition - lexicographic report: records a synonymy already implicit in usage, so it cannot be the ground of that synonymy Quotation withheld because this record is not cleared for redistribution. definition - explication: improves on the definiendum instead of reporting a synonymy, yet still rests on prior synonymies Quotation withheld because this record is not cleared for redistribution. definition - conventional abbreviation: the sole case where synonymy is genuinely created rather than reported Quotation withheld because this record is not cleared for redistribution. reductionism - radical: every meaningful statement translatable into a statement about immediate experience Quotation withheld because this record is not cleared for redistribution. reductionism - attenuated: the surviving supposition that a statement in isolation can be confirmed at all, which the paper says is the form that still does damage Quotation withheld because this record is not cleared for redistribution. meaning - as against reference: what essence becomes once it is detached from the object and attached to the word Quotation withheld because this record is not cleared for redistribution. economy - two antagonistic kinds in a formal system: brevity of expression versus poverty of primitive grammar and vocabulary Quotation withheld because this record is not cleared for redistribution. claim_type: critique a second reading reaching the same value. Recorded: critique — note: a second reading reaching the same value from a different sentence. Not applied - a person decides. Quotation withheld because this record is not cleared for redistribution. method: conceptual-analysis a second reading reaching the same value. Recorded: conceptual-analysis — note: a second reading reaching the same value from a different sentence. Not applied - a person decides. Quotation withheld because this record is not cleared for redistribution. contradicts → the analytic-synthetic distinction (first dogma of empiricism) Quotation withheld because this record is not cleared for redistribution. contradicts → reductionism / the verification theory of meaning (second dogma) differs in: the UNIT of confirmation: Quotation withheld - reductionism assigns a range of confirming experience to each statement singly Quotation withheld because this record is not cleared for redistribution. builds-on → Carnap, Der logische Aufbau der Welt Quotation withheld because this record is not cleared for redistribution. contradicts → Carnap, "Empiricism, Semantics, and Ontology" (1950) differs in: an assumption BOTH parties name: CARNAP HAS RECOGNIZED THAT HE IS ABLE TO PRESERVE A DOUBLE STANDARD FOR ONTOLOGICAL QUESTIONS AND SCIENTIFIC HYPOTHESES ONLY BY ASSUMING AN ABSOLUTE DISTINCTION BETWEEN THE ANALYTIC AND THE SYNTHETIC; AND I NEED NOT SAY AGAIN THAT THIS IS A DISTINCTION WHICH I REJECT Quotation withheld because this record is not cleared for redistribution. extends → the pragmatism of Carnap, Lewis, and others Quotation withheld because this record is not cleared for redistribution. critiques → Carnap's explanation of analyticity by state-descriptions and by semantical rules differs in: which notion the criterion captures: THE CRITERION IN TERMS OF STATE-DESCRIPTIONS IS A RECONSTRUCTION AT BEST OF LOGICAL TRUTH - it succeeds for logical truth and is being asked to do the work of analyticity Quotation withheld because this record is not cleared for redistribution. critiques → Kant's formulation of the analytic statement differs in: the SCOPE of the formulation and the status of its key term: IT LIMITS ITSELF TO STATEMENTS OF SUBJECT-PREDICATE FORM, AND IT APPEALS TO A NOTION OF CONTAINMENT WHICH IS LEFT AT A METAPHORICAL LEVEL Quotation withheld because this record is not cleared for redistribution. critiques → the term-by-term empiricism of Locke and Hume differs in: term-by-term against holistic: THE DOCTRINE IS UNNECESSARILY AND INTOLERABLY RESTRICTIVE IN THE TERM-BY-TERM CRITIQUE WHICH IT IMPOSES Quotation withheld because this record is not cleared for redistribution. extends → Carnap's own treatment of the physical world in the Aufbau Quotation withheld because this record is not cleared for redistribution. supports → White's essay on the analytic and the synthetic as an untenable dualism Quotation withheld because this record is not cleared for redistribution. Evening Star —same-as→ Morning Star Evening Star Morning Star strength of the relation: 1 Quotation withheld because this record is not cleared for redistribution. Evening Star —distinct-from→ Morning Star Evening Star Morning Star strength of the relation: 1 Quotation withheld because this record is not cleared for redistribution. 9 —same-as→ the number of the planets 9 the number of the planets differs in: MEANING. The sentence asserts sameness of ENTITY and denies sameness of MEANING in one breath - which is the distinction the whole paper turns on. 'same-as' unqualified reads as identity of sense, and Quine's point is that identity of reference does not give it. · strength of the relation: 1 Quotation withheld because this record is not cleared for redistribution. creature with a heart —distinct-from→ creature with a kidney creature with a heart creature with a kidney differs in: THE AXIS IS STATED IN THE SENTENCE, AND IT IS THE WHOLE POINT OF THE EXAMPLE. 'The general terms creature with a heart and creature with a kidney, e.g., are perhaps alike in extension but unlike in meaning.' The two terms are not distinct simpliciter - they are IDENTICAL on one dimension and different on another, and a bare 'distinct-from' records only the second half. Note also that 'extension' and 'intension' differ by one character, which is why these must never be de-duplicated by edit distance. · strength of the relation: 1 Quotation withheld because this record is not cleared for redistribution. intension —same-as→ meaning intension meaning strength of the relation: 1 Quotation withheld because this record is not cleared for redistribution. intension —renamed-across-fields→ connotation intension connotation strength of the relation: 1 Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the span calls connotation a 'variant vocabulary', not another field's name; renamed-across-fields is loose. Quotation withheld because this record is not cleared for redistribution. extension —renamed-across-fields→ denotation extension denotation strength of the relation: 1 Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the span calls denotation a 'variant vocabulary', not another field's name; renamed-across-fields is loose. Quotation withheld because this record is not cleared for redistribution. EVIDENCE CORRECTED (RE-GROUNDED) · intension —contrasted-with→ extension intension extension EVIDENCE CORRECTED: RE-GROUNDED. The old span named only one endpoint. This sentence names both and states the contrast in the field's own words - to OPPOSE intension to extension. — strength of the relation: 1 Quotation withheld because this record is not cleared for redistribution. radical reductionism —is-a→ reductionism radical reductionism reductionism strength of the relation: 1 Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but 'reductionism' occurs only inside 'radical reductionism'; is-a rests on the name. Quotation withheld because this record is not cleared for redistribution. EVIDENCE CORRECTED (RE-GROUNDED) · logical truth —part-of→ analytic logical truth analytic EVIDENCE CORRECTED: RE-GROUNDED. The old span listed analytic statements without naming logical truth. This sentence names both IN APPOSITION - 'the first class of analytic statements, THE LOGICAL TRUTHS' - which states the containment as directly as Quine states anything. — strength of the relation: 1 Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the span writes 'logically true' for the first class of analytic statements; endpoint variant. Quotation withheld because this record is not cleared for redistribution. EVIDENCE CORRECTED (RE-GROUNDED) · cognitive synonymy —operationalised-as→ interchangeability salva veritate cognitive synonymy interchangeability salva veritate EVIDENCE CORRECTED: RE-GROUNDED. The old span - 'a natural suggestion, deserving close examination…' - states the relation but names NEITHER endpoint as the edge writes them. This span names both and states the same operationalisation. — status: A SUGGESTION UNDER EXAMINATION, NOT A DEFINITION THE ESSAY ENDORSES. Quine offers interchangeability salva veritate as a 'natural suggestion, deserving close examination' and the rest of the paper is the examination that finds it wanting. Without this field the corpus reads as if Two Dogmas defined synonymy this way. · strength of the relation: 0.4 — the record marks this weak Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the span offers this as 'a natural suggestion, deserving close examination', which the paper goes on to reject; hedged. Quotation withheld because this record is not cleared for redistribution. analyticity —operationalised-as→ state-description analyticity state-description strength of the relation: 0.4 — the record marks this weak Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but attributed: the span reports how Carnap tends to explain analyticity. Quotation withheld because this record is not cleared for redistribution. EVIDENCE CORRECTED (RE-GROUNDED) · synonymy —operationalised-as→ method of empirical confirmation synonymy method of empirical confirmation EVIDENCE CORRECTED: RE-GROUNDED. The old span named only one endpoint. This sentence names both and gives the operationalisation directly. — strength of the relation: 0.5 — the record marks this weak Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but attributed: the span says statement synonymy 'is said to be' likeness of confirmation method. Quotation withheld because this record is not cleared for redistribution. reductionism —same-as→ the analytic and the synthetic reductionism the analytic and the synthetic strength of the relation: 0.8 Audit 2026-09-28: DOUBTFUL - read against its span and kept, but the span is cut before it states how reductionism relates to the analytic-synthetic distinction; same-as may overstate it. Quotation withheld because this record is not cleared for redistribution. bachelor —same-as→ unmarried man bachelor unmarried man strength of the relation: 0.6 — the record marks this weak Quotation withheld because this record is not cleared for redistribution. RETRACTED (A NEAR-DUPLICATE OF [17] THAT THE FOLD CANNOT SEE) · physical objects —distinct-from→ the gods physical objects the gods RETRACTED: A NEAR-DUPLICATE OF [17] THAT THE FOLD CANNOT SEE. Same record, same type, same span to the byte - '…objects and the gods differ…' - and the only difference is the endpoint name: 'the gods' here, 'the gods of Homer' at [17]. The fold keys on (from, to, type, evidence), so a looser spelling of one endpoint produces a second edge asserting the identical thing. Keeping [17], which uses the paper's own fuller phrase. NOTE ALSO, and left standing on [17]: the span scopes its own difference - 'ONLY IN DEGREE…' - and 'distinct-from' has no slot for that, so the edge reads as a difference in kind, which is what the sentence denies. — the relation itself is separately evidenced · strength of the relation: 0.15 — the record marks this weak · note: RE-GROUNDS another row of this record: the span stated the relation, the endpoint names did not match its words. · the relation stated: the two kinds of posit differ in DEGREE, not in KIND - both are posits Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the span says the two differ 'only in degree and not in kind'; distinct-from is loose. Quotation withheld because this record is not cleared for redistribution. Quine argues both central dogmas of empiricism — the analytic/synthetic distinction and reductionism — are ill-founded; abandoning them blurs the boundary between metaphysics and natural science and shifts toward pragmatism. Quotation withheld because this record is not cleared for redistribution. The supposed analytic/synthetic boundary has never actually been drawn, despite its a priori plausibility. Quotation withheld because this record is not cleared for redistribution. self-contradictoriness, in the sense needed to define analyticity, needs clarification as much as analyticity itself Quotation withheld because this record is not cleared for redistribution. definition rests on synonymy rather than explaining it Quotation withheld because this record is not cleared for redistribution. interchangeability salva veritate in an extensional language does not establish cognitive synonymy Quotation withheld because this record is not cleared for redistribution. appeal to semantical rules of an artificial language does not clarify analyticity Quotation withheld because this record is not cleared for redistribution. the unit of empirical significance is the whole of science, not the individual statement Quotation withheld because this record is not cleared for redistribution. no statement is immune to revision Quotation withheld because this record is not cleared for redistribution. ontological questions differ from scientific ones only in degree Quotation withheld because this record is not cleared for redistribution. abandoning the dogmas blurs the boundary between metaphysics and natural science and shifts philosophy toward pragmatism Quotation withheld because this record is not cleared for redistribution. synthetic truths synthetic truths Quotation withheld because this record is not cleared for redistribution. reductionism reductionism Quotation withheld because this record is not cleared for redistribution. logically true statement logically true statement about this quote: QUOTE RECOVERED FROM THE PAPER'S OWN CHARACTERS. The recorded quote was LETTERS-ONLY IDENTICAL to this stretch of the staged text but not an exact substring of it - spaces, hyphens, ligature glyphs or mathematical operators differed, so it could never anchor. This is the file's own text between the same two letters: not produced by a rule about the damage, copied out of the file. Round-trip checked (the recovered span is letters-only equal to the original) and length-checked. The text as originally recorded is kept in evidence as extracted. Quotation withheld because this record is not cleared for redistribution. logical truth logical truth Quotation withheld because this record is not cleared for redistribution. extension extension Quotation withheld because this record is not cleared for redistribution. state-description state-description Quotation withheld because this record is not cleared for redistribution. reductionism reductionism Quotation withheld because this record is not cleared for redistribution. verification theory of meaning verification theory of meaning Quotation withheld because this record is not cleared for redistribution. radical reductionism radical reductionism Quotation withheld because this record is not cleared for redistribution. explication explication Quotation withheld because this record is not cleared for redistribution. primitive notation primitive notation Quotation withheld because this record is not cleared for redistribution. extensional language extensional language Quotation withheld because this record is not cleared for redistribution. analytic analytic Quotation withheld because this record is not cleared for redistribution.

The full text of this article is not redistributable, so it is not included here. The evidence sentences are withheld with it; the tag layer and its grounding status remain. The article itself is at https://doi.org/10.2307/2181906.