Indeterminacy of Translation

Overview and Origins

The indeterminacy of translation is a thesis advanced by W. V. O. Quine, especially in Word and Object (1960). It concerns the possibility and nature of translating one natural language into another. According to Quine, even in principle there may be no uniquely correct translation from one language to another, because multiple, systematically different translation schemes can fit all possible observable data about speakers’ linguistic behavior equally well.

Quine’s view emerges from a combination of empiricism, behaviorism about meaning, and a holistic view of language. He argues that the only evidence we have about what speakers mean by their words is how they use those words in response to stimuli and in communication with others. If two rival translation manuals predict the same linguistic behavior in all possible circumstances, there will be no empirical basis for declaring one “right” and the other “wrong.”

The thesis is distinct but related to Quine’s inscrutability of reference: the idea that there is no fact of the matter whether a term like “gavagai” refers to rabbits, undetached rabbit parts, or stages of rabbits, so long as the whole pattern of language use is preserved. Both ideas challenge traditional assumptions about determinate meanings and reference.

The Argument from Radical Translation

Quine presents his case through a thought experiment about radical translation: a field linguist encountering an entirely unknown language with no bilingual informants and no prior lexicon.

The linguist observes that native speakers utter “Gavagai!” whenever rabbits appear. It is natural to hypothesize that “gavagai” means “rabbit.” But Quine claims that many alternative hypotheses are equally compatible with all present and future observable data, if appropriately adjusted elsewhere in the translation scheme.

For example:

• “Gavagai” might be translated as “undetached rabbit part.”
• Or as “rabbit-stage” (a temporal slice of a rabbit).
• Or as “there goes rabbithood” (a sortal or universal).

By adjusting auxiliary translations—of terms for parts, wholes, times, events, identity, and so on—the linguist can construct a systematic translation manual where “gavagai” maps to any one of these alternatives while still preserving all correct predictions about when speakers will assent to or dissent from sentences, how they will infer one statement from another, and how they will respond to environmental stimuli.

Quine’s behaviorist constraint on evidence is crucial here. The linguist has access only to:

• Patterns of assent and dissent (when speakers agree or disagree with utterances),
• Stimulus situations (what is going on in the environment),
• And interlinguistic interactions (how utterances co-occur, inferential roles, etc.).

Within this evidential base, Quine argues, numerous incompatible translation manuals remain empirically equivalent: they make the same predictions about usage in all conceivable circumstances.

From this he infers the indeterminacy thesis:

• There is no fact-of-the-matter beyond these patterns of use that could determine that “gavagai” really means “rabbit” rather than “undetached rabbit part.”
• Translation is thus not fully objective in the way traditional semantic theories assume. It is constrained but underdetermined by behavioral data.

Quine stresses that this is not mere practical uncertainty. It is not just that we cannot find the correct translation; rather, there is no uniquely correct translation to be found, given his assumptions about what meaning could be.

Philosophical Implications

The indeterminacy of translation has wide-ranging consequences for philosophy of language, epistemology, and metaphysics.

1. Rejection of a robust notion of meaning

   The thesis supports Quine’s skepticism about intensional entities such as meanings, propositions, or senses. If there is no unique way to assign meanings across languages, then the idea of determinate, language-independent meanings becomes suspect. Quine instead treats meaning as derivative from patterns of use, and is suspicious of semantic explanations that appeal to abstract entities.

2. Holism about language and theory

   Quine holds that sentences face the tribunal of experience not individually, but only in clusters or theories. This holism helps generate the indeterminacy: changes in how we translate one expression can be compensated for by changes elsewhere. Similarly, in scientific theory choice, multiple empirically equivalent theories can fit the same data; the indeterminacy of translation is a linguistic analogue of underdetermination of theory by evidence.

3. Inscrutability of reference

   Indeterminacy of translation suggests that reference is not fixed by any fact over and above total linguistic practice. Whether words refer to rabbits, undetached rabbit parts, or “rabbithood” may be a matter of how we choose to regiment our language rather than a deep metaphysical fact. This challenges realist views that treat reference as tightly tied to objective structure in the world.

4. Relativity of ontology

   Quine links translation and ontology: different translation manuals may go along with different ontological commitments. One manual posits rabbits; another, rabbit stages; another, only events or sense-data. If all are equally compatible with all linguistic evidence, this supports his view that what there is (our ontology) is partly a matter of our conceptual scheme and choices of theory, not uniquely forced by experience.

5. Challenges to analyticity and synonymy

   If translation is indeterminate, then the idea of precise synonymy—perfect sameness of meaning—comes under pressure. This connects to Quine’s earlier critique of the analytic–synthetic distinction: if we cannot give a robust account of meaning and synonymy, we cannot neatly distinguish truths by meaning alone from truths by fact.

Criticisms and Responses

The indeterminacy thesis has been highly influential but also widely challenged. Criticisms target both its evidential assumptions and its philosophical conclusions.

1. Rejection of strict behaviorism

   Many philosophers argue that Quine’s behaviorist restriction on evidence is too narrow. Actual translators and language learners use not only overt behavior but also:

   • Shared intentions and communicative purposes,
   • Knowledge of cognitive capacities and perception,
   • Rich pragmatic and contextual cues,
   • And introspective or first-person access to their own meanings.

   If these are allowed as legitimate data, the space of acceptable translation manuals may shrink considerably, undermining indeterminacy.

2. Appeals to intention and charity

   Donald Davidson, while influenced by Quine, objects to strong indeterminacy. His principle of charity holds that, in interpreting others, we must systematically maximize truth and rationality in their beliefs. Combined with assumptions about shared objective reality and communicative intentions, charity may pick out a unique best interpretation or narrow the range drastically, weakening Quine’s claim that there is no fact of the matter.

3. Internalist and cognitive theories of meaning

   Critics from cognitive science, psycholinguistics, and internalist semantics argue that meaning is grounded in mental representations, not just public behavior. If we allow facts about internal cognitive states to constrain translation, then rival manuals that treat “gavagai” as “rabbit” versus “undetached rabbit part” may no longer be equally acceptable: they may posit implausible or overly complex mental structures for ordinary speakers.

4. Ordinary translation practice

   Some object that Quine’s scenario ignores the success of real-world translation. Human translators routinely converge on stable, highly reliable translations between natural languages. Proponents of indeterminacy often reply that Quine’s claim is in-principle and metaphysical, not about practical success; but critics contend that any adequate theory of translation must respect how translators actually work.

5. Indeterminacy vs. underdetermination

   Some philosophers accept a weaker thesis: underdetermination of translation by certain kinds of evidence, without concluding that there is never any fact of the matter. They argue that Quine overgeneralizes from logical possibility to metaphysical possibility, and that additional theoretical, pragmatic, or normative considerations can justifiably select a preferred translation.

Despite these criticisms, Quine’s indeterminacy thesis remains a central reference point in debates about meaning, interpretation, and the objectivity of language. It continues to shape discussions of how far empirical data can fix semantics, how we should understand cross-linguistic equivalence, and whether the notion of a uniquely correct translation is philosophically sustainable.