Skip to main content
Log in

Quine on matters of fact

  • Published:
Synthese Aims and scope Submit manuscript


The idea of there being “no fact of the matter” (NFM) features centrally in Quine’s indeterminacy theses. Yet there has been little discussion of how exactly Quine understands this idea. In this paper I identify, develop and then critically evaluate Quine’s conception of NFM. In Sects. 34 I consider a handful of intuitive semantic and ontological conceptions of NFM and argue that none is workable from within Quine’s philosophy. I conclude that the failure of each of these proposals is due to the immanent status of truth and existence for Quine. In Sect. 5 I then present Quine’s official conception of NFM. Briefly, Quine’s idea is that there is NFM between two theories of (say) translation iff those theories are physically equivalent. I develop this idea in detail. Finally, I raise two independent problems for this conception of NFM. In Sect. 6 I argue that Quine’s definition is too strong: given what he means by NFM, his arguments for indeterminacy—even granting all their premises and internal reasoning—simply cannot support his claim that there is NFM regarding translation; instead they establish a strictly weaker conclusion. In Sect. 7 I argue that Quine’s conception of NFM is in significant tension with his thesis of physicalism, and that he must give up one or the other.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Subscribe and save

Springer+ Basic
EUR 32.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or Ebook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others


  1. A wealth of commentary surrounds Quine’s indeterminacy thesis. For a small but representative sample, see Chomsky (1969), Rorty (1972), Friedman (1975), Searle (1987), Føllesdal (1990), and Soames (1999).

  2. Our talk of “disciplines”, as well as what Quine means by natural science and translation, will be made precise in Sect. 2.

  3. My use of ‘NFM’ will be somewhat loose throughout the paper. Usually I’ll speak in terms of the general concept of NFM, though for stylistic reasons I will on occasion also speak of the phrase ‘no fact of the matter’. Similarly, I will speak of both Quine’s particular conception of NFM as well as what he means by NFM, taking these to amount to the same thing. I will also use ‘NFM’ both to mention the concept/phrase and also to use it. The context should make clear which sense is intended on any given occasion.

  4. You don’t have to agree that there is NFM in all or any of these cases in particular. The point is just that we can and frequently do make such claims, and that at least sometimes they are true.

  5. These conceptions will be made more precise in Sect. 3.

  6. Here are some nearby variants of the semantic conception: There is NFM regarding X iff our sentences about X express no propositions, or express multiple propositions differing in truth-value, or are systematically false. What we ultimately say about the conceptions as stated above will carry over to these as well (Sect. 3); see notes 21, 25 and 26. One might also take NFM as primitive, but I take this to be a last resort, and in any case it is clearly not what Quine himself has in mind (though see the suggestion at the very end of this paper). Finally, note that there is nothing stopping someone from combining the semantic and ontological conceptions above.

  7. We give further support to this claim in Sect. 7.

  8. One exception is Gaudet (2006). For further points of contact with Gaudet, Gibson, Friedman and other critics, see notes 49 and 57.

  9. Here I follow Hylton (2007, p. 197).

  10. Quine usually refers to the indeterminacy of reference as the inscrutability of reference. It is closely related to his thesis of ontological relativity. We will take the indeterminacy of reference to include any case in which two manuals \(\hbox {T}_{1}\) and \(\hbox {T}_{2}\) translate a single word w into two different words a and b, respectively, where a and b differ in extension, and in which there is indeterminacy between \(\hbox {T}_{1}\) and \(\hbox {T}_{2}\) (i.e. NFM). Hence the indeterminacy of reference, thus defined, does not preclude holophrastic indeterminacy in addition. Indeed, holophrastic indeterminacy would seem to entail the indeterminacy of reference (though not vice versa). For more on the connection between these ideas, see Quine (1968). Finally, notice that I have characterized the indeterminacy of reference in terms of translation rather than directly in terms of reference. In some contexts this may seem to get things back to front, but in Quine’s case it is translation that comes first. The reason that it is indeterminate whether my word w refers to x or to y is because it is indeterminate whether w should be interpreted/translated as ‘x’ or as ‘y’. See Quine (1968) for this way of reasoning. Of course Quine does often talk about the indeterminacy of reference in terms of reference itself, and there’s nothing wrong with that, but given the phenomenon’s source in the interpretation/translation of terms, I take it there is nothing wrong with my characterization either.

  11. The main exception here has to do with Quine’s view on the existence of meanings. See note 35. The difference between holophrastic and referential indeterminacy also comes up in our discussion of Quine’s arguments for the indeterminacy of translation (Sect. 6). See notes 42 and 46.

  12. Alternatively, if the reader would prefer to reserve ‘the indeterminacy of translation’ as a name for holophrastic indeterminacy alone, as is arguably more traditional, then she may, so long as it is kept in mind that everything we say regarding the holophrastic case is meant to carry over to the indeterminacy of reference as well.

  13. The nature of this acceptance and what makes a theory “best” are not issues that we’ll address here. Note also that “science” is construed rather broadly by Quine, and is contrasted primarily with “first philosophy”. See Quine (1995b).

  14. It may be that this shift is only appropriate given a relatively broad understanding of Quine’s notion of “science”. This is indeed how I read Quine (see note 13), but I admit that the matter is not without controversy. For complications, Haack (1993).

  15. This characterization is extracted primarily from Quine (1970). The “totality of evidence” is my own phrase, which I intend to be equivalent to Quine’s idea of “all possible observations”—i.e. the set of all true “place-timed” observation sentences (1970, p. 179). My change in terminology is only to avoid distracting talk of possibilities. We will not decide what Quine means by the two uses of ‘(in)compatible’ except to note that they need not have the same sense.

  16. Quine usually speaks of natural science as a whole being underdetermined, rather than of each area individually (though see 1970). But his reasons for the former carry over to the latter, and for the purposes of comparison with indeterminacy it will be helpful to focus on the individualized version. The underdetermination of each natural science, however, does not require that these areas be underdetermined independently of each other. Empirically-invariant changes in chemistry (say) may be required to accommodate an empirically-invariant change in biology.

  17. With Quine’s official conception of NFM it is somewhat inaccurate to talk of NFM “regarding” a given theory type, as we’ll see in Sect. 5. But until then we’ll stick to this loose way of speaking. Each conception we look at in this section will make precise, in its own way, what it means for there to be NFM “regarding” a theory type.

  18. For much of this paragraph I follow Hylton (2007, p. 318).

  19. Quine’s (1960, Chap. 2) remarks to the effect that the indeterminacy of translation does not undermine the actual linguist’s enterprise are also relevant here.

  20. The above formulation of S-NFM is actually a bit simplistic, and if the very idea of truth-value gaps were not such a nonstarter for Quine (as we’ll see below that it is), it would probably be worth refining this formulation to better match Quine’s thesis. For instance, there will always be theories of translation that simply contradict the evidence, and hence should probably be considered just plain false (e.g. one that translates ‘gavagai’ as ‘horse’). Accordingly, a more careful formulation of S-NFM might go: there is NFM regarding \(\mathcal{T}\) iff (a) certain theories of \(\mathcal{T}\) contain (some) sentences that are neither true nor false; and (b) no theory of \(\mathcal{T}\) contains only true sentences. But this formulation will be just as unworkable for Quine as the simple one above, insofar as both require that we view at least some sentences as neither true nor false. For simplicity, I stick to what is above.

  21. This also demonstrates the mistake in thinking, alternatively, of NFM as systematic falsity of a theory type.

  22. Yet another problem with S-NFM stems from Quine’s disquotationalism about truth (1960, p. 24). For there are well known difficulties with trying to combine disquotationalism (or any form of deflationism) with truth-value gaps. See Boghossian (1990) and Horwich (1999). For an attempt to get around these difficulties, see Burgess (2010).

  23. Facts are to be taken here as worldly, rather than linguistic, entities. Like our formulation of S-NFM, F-NFM, as stated, is also a bit simplistic (see note 20). Refinements analogous to those proposed for S-NFM could be put in place, but the resulting formulations would do no better than the simpler version above.

  24. This is implicit in his discussion, and positive proposal, of what such things might be (1968, p. 196).

  25. Similarly, Quine does not recognize the existence of propositions in any traditional sense (i.e. as intensional objects). Hence any conception of NFM that would mark the difference between translation and natural science in terms of propositions—for instance one according to which a non-factual sentence expresses more than one proposition, or one according to which the sentence expresses no proposition—will clearly be unworkable for Quine.

  26. Similarly, one might suggest that the predicate ‘means the same as’ doesn’t describe any real “way” that things (i.e. words) could be related in the world, whereas presumably a particular spin predicate does describe a real “way” for particles to be. [Compare this with Lewis’s (2001) distinction between what things there are and how they are.] But what could this distinction amount to in the present context? For familiar reasons, it can’t just be that ‘mean the same as’ doesn’t describe a true way for the world to be. Instead, we might say that it doesn’t describe an objective way for the world to be, but this again seems to just reduce to a factual way, and hence presupposes the concept of NMF/factuality. Thanks to an anonymous referee for suggesting these lines of thought.

  27. It can be helpful to think about Quine’s immanence/transcendence distinction in comparison to Carnap (1950). The transcendental notions of truth and existence that Quine rejects are akin to the external ontological questions that Carnap rejects, and which both philosophers view as the preoccupation of traditional metaphysics. Similarly, the immanent notions that Quine allows for are akin to the framework-internal questions that Carnap allows for. The main differences between Quine are Carnap are two. First, Carnap believes in a sharp division between frameworks, and hence subscribes to a sort of ontological pluralism, whereas Quine thinks these frameworks all blend together to form one overall theory, and takes ontology to be monistic. Second, Carnap draws a sharp distinction between pragmatic questions of theory choice and internal questions of a framework, whereas Quine draws no such division, viewing all internal questions as, in principle, pragmatically tinged. See Quine (1951a). For more on this comparison see Price (2009).

  28. See also 1986b, p. 429.

  29. Our previous talk of NFM “regarding a theory type T” therefore turns out to be infelicitous. See note 17 Later on we will define a derivative sense of NFM that applies to theory types.

  30. I will speak loosely in terms of fundamental/elementary particles/entities. These needn’t be particles in any usual sense; they might for instance be space-time regions (1981a, pp. 16–17).

  31. We don’t need to determine the exact sense of incompatibility here (though see below for a suggestion).

  32. When I say “physical facts”, what I mean is the actual/true configuration of elementary physics, i.e. @.

  33. There is also a question as to whether these two senses of (in)compatibility match those involved in underdetermination. We won’t decide this here.

  34. See Hylton (2007, p. 54) for discussion.

  35. Quine’s denial of sentence meanings in particular would seem to rely specifically on NFM* regarding holophrastic translation, since only this would imply that there was not an objectively correct same-meaning relation between sentences. But in any case the general ideas of an objective property, clear identity criteria and a correct theory all gain significant clarity through the lens of Quine’s general conception of NFM.

  36. This presupposes that there is a uniquely correct theory of behavior, and hence that behavior itself isn’t indeterminate. Quine clearly believes this (1977, p. 162), and we will assume it too. We could drop this presupposition by requiring only compatibility with some correct theory of behavior.

  37. Thus, supposing our talk of behavioral dispositions can ultimately be reduced to, or replaced by, talk about (say) certain physiological mechanisms, then to this extent—and only to this extent—facts about the latter will play a role in determining the correctness of translation. [Quine is actually quite optimistic about the prospects of such a reduction (1973, pp. 12–15)]. What linguistic behaviorism implies is that the only way a type of theory can play a role in determining the correctness of translation is, as it were, via the behavioral facts.

  38. A more detailed defense of this claim would require specifying the exact senses of “compatible” and “incompatible” at work in NFM*. We won’t do this here, but see Taylor (2013).

  39. Note that, unlike any other theory type, the factuality of physics is arguably trivial, insofar as presumably only one theory of physics (up to incompatibility) will be compatible with @. This is an intriguing consequence of Quine’s definitions, but not something we can explore further here.

  40. For other arguments of Quine’s, see note 46 below.

  41. Recall that we’re assuming there is a single correct theory of behavior. See note 36. Like with physics, I will often refer to this theory as the behavioral “facts”.

  42. This argument is sometimes thought of as an argument for holophrastic indeterminacy in particular (Hylton 2007, pp. 215–216). But Quine (1968) also seems to use it to support the indeterminacy of reference directly. Which way we view it will depend on whether we think (P1) is plausible with respect to manuals that disagree “holophrastically” or only with respect to those that disagree merely “extensionally” (if either). We don’t need to consider this here, however, since we are granting (P1) to Quine for the sake of argument (in whichever of these strengths he’d like). Quine does have another, more direct argument, in terms of proxy functions, for the indeterminacy of reference that doesn’t apply to holophrastic indeterminacy. For discussion of this and other arguments, see note 46.

  43. In this example we focus on the translation of individual words, rather than sentences. As mentioned in note 10, however, holophrastic indeterminacy would seem to bring along indeterminacy in the translation of sub-sentential expressions as well, i.e. what we’re calling the indeterminacy of reference. So focusing on individual words here let’s the example and argument apply to both holophrastic and referential indeterminacy.

  44. The relativization to \(\hbox {T}_{1}\) is important here. What we’d like to be able to say is that ‘part’ and X have the exact same use in the two languages. But that is not quite right, since we use ‘part’ in conjunction with a whole slew of English words, whereas the foreigner uses X with a whole slew of her words. But these uses will be analogous with respect to \(\hbox {T}_{1}\) just so long as they are exactly the same up to translation according to \(T_{1}\).

  45. I’ve occasionally met the following sort of response to the preceding example: “The differences between B and B* show that the pair of languages considered in B is a different pair of languages from that considered in B*. But Quine is only concerned with multiple manuals between the same pair of languages. So any appeal to B* is beside the point.” Now, I agree that Quine is primarily concerned with multiple manuals between the same pair of languages. And this concern is reflected in our example: with respect to the actual behavioral facts (B), \(\hbox {T}_{1}\) and \(\hbox {T}_{2}\) do translate the same pair of languages. Where B* comes in is only in evaluating whether there is NFM between \(\hbox {T}_{1}\) and \(\hbox {T}_{2}\). Given Quine’s definition of NFM, whether there is NFM between \(\hbox {T}_{1}\) and \(\hbox {T}_{2}\) will depend on how \(\hbox {T}_{1}\) and \(\hbox {T}_{2}\) relate to alternative behavioral realities. If this means that we must consider how \(\hbox {T}_{1}\) and \(\hbox {T}_{2}\) relate to alternative languages, as the response suggests, then so be it. But that does nothing to undermine the idea that \(\hbox {T}_{1}\) and \(\hbox {T}_{2}\) in fact translate between the same two languages. In any case, suffice it to say that if consideration of any alternative behavioral (and thus physical) reality is truly irrelevant to Quine’s indeterminacy thesis, then this immediately secures our main claim in this section, i.e. that Quine’s official definition of NFM is too strong.

  46. This isn’t Quine’s only argument for his indeterminacy thesis. In 1970 he argues that holophrastic indeterminacy follows directly from underdetermination by evidence, and in 1968 he argues that it follows from a holistic picture of evidence. See Hylton (2007) for details. However, it is clear that, like the argument above, neither of these arguments can support anything stronger than UP. As for the indeterminacy of reference, Quine often appeals to the existence of proxy functions. Indeed he thinks these provide a “trivial proof” of the indeterminacy of reference (1986c, p. 728). Be that as it may, this sort of “proof” again has the form only of underdetermination, not equivalence. What considerations of proxy functions show, ultimately, is that our choice of how to translate ‘x’ and ‘P’ is underdetermined by our overall use of the sentence ‘x is P’—that is, quite consistent with the role that ‘x is P’ plays in our overall theory and language, we are free to translate it either homophonically or as ‘f(x) is the f of a P’, where f is an appropriate proxy function (1981a, p. 19). But none of this shows that were our use of either ‘x’ or ‘P’ drastically different—were it to play a different role in our overall theory and language—still either both or neither of these translations would be compatible with that use. It doesn’t show that these translations are equivalent in this sense. So there is no reason to think they will be physically equivalent, and so again we don’t get NFM*; at most we get UP.

  47. Over time Quine grew less confident in holophrastic indeterminacy, ultimately viewing it as a mere “conjecture” later in his career (1986c, p. 728). And so perhaps, one might think, it shouldn’t be too surprising that his arguments cannot establish NFM*. Again, the thought would be that the problem here doesn’t have anything to do with Quine’s conception of NFM; rather his conjecture was just too ambitious, and that is no surprise. But again I think this reaction would be a mistake; his conception of NFM* is too strong. Four reasons for this. First, as we are about to see, there are good independent reasons for thinking that NFM* is too strong; instead the weaker principle UP better matches our intuitive understanding of indeterminacy. Second, Quine’s apparent shift in attitude doesn’t explain why, in 1960, when he confidently presented the indeterminacy of (holophrastic) translation as a proper thesis, his argument there still did not match the structure of NFM*, but rather seems only to have (at best) established UP. Third, as indicated in notes 42 and 46, I don’t think that Quine’s arguments for the indeterminacy of reference do any better at establishing NFM* than do his arguments for holophrastic indeterminacy; again they only get as far as UP. Finally, even if we do view Quine’s thesis of holophrastic indeterminacy as a mere conjecture, presumably the conjectural part lies with (P1)—the idea that there could be two distinct theories of translation that disagreed not only extensionally but also holophrastically, but which nevertheless were each compatible with the behavioral facts. [See Hylton (2007, p. 217)]. It’s not that Quine is confident in (P1) but can’t see exactly how to get from there to NFM*; it’s rather that, if anything, he is not so confident about (P1) itself (in the holophrastic case).

  48. See Kripke (1982) and Parfit (1986). See also van Cleeve (1992).

  49. Indeed, given these considerations it would be natural to simply identify Quine’s conception of indeterminacy with UP (and thus multi-correctness), and most commentators interpret Quine in just this way (Friedman 1975; Gibson 1988; Soames 1999). I don’t disagree with this way of viewing things. My point is just that this conception of indeterminacy cannot support Quine’s official definition of NFM.

  50. The obvious “fix” here would be to move away from physical equivalence and instead say that there is NFM between two theories iff either both or neither is compatible with @, and then identify NFM* with UP. But we will not consider this proposal further here.

  51. Our criticism in this section is independent of the one raised in Sect. 6. In particular, the current difficulty would be just as problematic for the revised versions of NFM and NFM* suggested in note 50.

  52. In particular, this is why, at least in principle, it is always an option, against arguments purporting to show that there is NFM regarding some domain, to maintain that that domain itself is irreducible or sui generis. For instance, see Boghossian’s (1989) anti-reductionist response to Kripke (1982).

  53. 1960, Chap. 7; 1977; and 1995a, Chap. 2.

  54. Here I follow Hylton (2007, Chap. 12).

  55. 1960, pp. 1–4, 234–236 and 264-266.

  56. Notice that, even if we were wrong in the first step of our objection—i.e. wrong to think that Quine’s conception of NFM requires a thesis of physicalism—the basic tension being pointed to here would remain. For it wouldn’t change the fact that Quine in fact subscribes to a physicalist thesis that is identical to his conception of NFM. The only thing that would be different would be that Quine would now have the option, without doing harm to his conception of NFM, of giving up his physicalism full stop. But that doesn’t seem like an attractive option in any case, given how central Quine’s physicalism is to his philosophy overall.

  57. As a purely interpretive matter, I don’t think there is any good evidence to suggest that Quine really intends (*) in one way rather than the other. As the quotations from this section and Sect. 6 demonstrate, he explicitly uses (*) for both physicalism and NFM, sometimes even within the same paper (e.g. 1977). However uncharitable it may be, I think Quine meant (*) in both ways, and just didn’t realize the dangers of his dual intentions.

    Many commentators take (*), or something close to it, to be a statement of Quine’s (mature) physicalism. See Friedman (1975), Gibson (1988), Soames (1999), and Hylton (2007). None of these authors consider it as a definition of NFM. [Though this is a little less clear in the case of Gibson, who seems to relate (*) to both physicalism and NFM, but who never explicitly lays out either.] Perhaps this should give us pause in thinking of (*) otherwise. On the other hand, no such author seems to have explicitly considered how, exactly, we should interpret Quine’s use of NFM, and most seem to be working with either an unanalyzed notion of NFM, or else some sort of generic conception—e.g. truth-value gaps—that we now know is decidedly not open to Quine. Gaudet (2006), on the other hand, does consider Quine’s conception of NFM, and draws a close connection between it and (*). But she also takes (*) to be a statement of his physicalism (p. 183), and thus is subject to the very circularity we are here concerned with.

  58. For instance, if we took correctness to be truth, then multi-correctness would entail the truth of two incompatible theories. But this doesn’t comport with Quine’s view that to call a sentence true is just to accept it as part of our overall theory; for we do not accept multiple incompatible manuals as part of our overall theory (or at least we ought not to). Nor can we take unique correctness to be truth; for then multi-correctness would entail truth-value gaps.

  59. Alternatively, we could take (**) as a definition of physicalism, and correctness as primitive. What we say here would apply mutatis mutandis to that proposal as well.

  60. Rest assured, the irony of proposing a closed circle of primitive concepts to the author of “Two Dogmas” is not lost on me.


  • Boghossian, P. (1989). The rule-following considerations. Mind, 98, 507–594.

    Article  Google Scholar 

  • Boghossian, P. (1990). The status of content. The Philosophical Review, 99, 157–184.

    Article  Google Scholar 

  • Burgess, A. (2010). How to reconcile deflationism and nonfactualism. Noûs, 44, 433–450.

    Article  Google Scholar 

  • Carnap, R. (1950). Empiricism, semantics and ontology. Revue Internationale de Philosophie, 4, 208–228.

    Google Scholar 

  • Chomsky, N. (1969). Quine’s empirical assumptions. Synthese, 19, 53–68.

    Article  Google Scholar 

  • Field, H. (1973). Theory change and the indeterminacy of reference. Journal of Philosophy, 70, 462–481.

    Article  Google Scholar 

  • Føllesdal, D. (1990). Indeterminacy and mental states. In R. Barrett & R. Gibson (Eds.), Perspectives on Quine. Cambridge: Wiley-Blackwell.

    Google Scholar 

  • Friedman, M. (1975). Physicalism and the indeterminacy of translation. Noûs, 9, 353–374.

    Article  Google Scholar 

  • Gaudet, E. (2006). Quine’s notion of matter of fact. Dialectica, 60(2), 181–193.

    Article  Google Scholar 

  • Gibson, R. (1988). Enlightened empiricism. Tampa: University Press of Florida.

    Google Scholar 

  • Haack, S. (1993). The two faces of Quine’s naturalism. Synthese, 94(3), 335–356.

    Article  Google Scholar 

  • Horwich, P. (1999). Truth (2nd ed.). Oxford: Oxford University Press.

    Google Scholar 

  • Hylton, P. (2007). Quine. New York: Routledge.

    Google Scholar 

  • Kripke, S. A. (1982). Wittgenstein on rules and private language. Cambridge, MA: Harvard University Press.

    Google Scholar 

  • Lewis, D. (2001). Truthmaking and difference-making. Noûs, 35(4), 602–615.

    Article  Google Scholar 

  • Parfit, D. (1986). Reasons and persons. Oxford: Oxford University Press.

    Book  Google Scholar 

  • Price, H. (2009). Metaphysics after Carnap: The ghost who walks? In D. Chalmers, et al. (Eds.), Metametaphysics. Oxford: Oxford University Press.

    Google Scholar 

  • Quine, W. V. (1948). On what there is. The Review of Metaphysics, 48, 21–38.

    Google Scholar 

  • Quine, W. V. (1951a). On Carnap’s views on ontology. Philosophical Studies, 2(5), 65–72.

    Article  Google Scholar 

  • Quine, W. V. (1951b). Two dogmas of empiricism. The Philosophical Review, 60, 20–43.

    Article  Google Scholar 

  • Quine, W. V. (Ed.). (1954). The scope and language of science. The ways of paradox and other essays (pp. 228–245). Cambridge, MA: Harvard University Press.

  • Quine, W. V. (1958). Speaking of objects. Proceedings and Addresses of the American Philosophical Association, 31, 5–22.

    Article  Google Scholar 

  • Quine, W. V. (1960). Word and object. Cambridge, MA: MIT Press.

    Google Scholar 

  • Quine, W. V. (1968). Ontological relativity. The Journal of Philosophy, 65(7), 185–212.

    Article  Google Scholar 

  • Quine, W. V. (1970). On the reasons for indeterminacy of translation. The Journal of Philosophy, 67, 178–183.

    Article  Google Scholar 

  • Quine, W. V. (1973). The roots of reference. La Salle, Illinois: Open Court.

    Google Scholar 

  • Quine, W. V. (1975). Mind and verbal dispositions. In S. Guttenplan (Ed.), Mind and language. Oxford: Clarendon.

    Google Scholar 

  • Quine, W. V. (1977). Facts of the matter. Southwestern Journal of Philosophy, 9(2), 155–169.

    Article  Google Scholar 

  • Quine, W. V. (Ed.). (1978). Goodman’s ways of worldmaking. Theories and things. Cambridge, MA: Harvard University Press.

  • Quine, W. V. (Ed.). (1981a). Things and their place in theories. Theories and things. Cambridge, MA: Harvard University Press.

  • Quine, W. V. (1981b). What price bivalence? The Journal of Philosophy, 78, 90–95.

  • Quine, W. V. (1986a). Philosophy of logic (2nd ed.). Cambridge, MA: Harvard University Press.

    Google Scholar 

  • Quine, W. V. (1986b). Reply to Hilary Putnam. In L. E. Hahn & P. A. Schilpp (Eds.), Philosophy of W.V. Quine (2nd ed.). La Salle, Illinois: Open Court.

  • Quine, W. V. (1986c). Reply to John Woods. In L. E. Hahn & P. A. Schilpp (Eds.), Philosophy of W.V. Quine (2nd ed.). La Salle, Illinois: Open Court.

  • Quine, W. V. (1987). Indeterminacy of translation again. The Journal of Philosophy, 84, 5–10.

    Article  Google Scholar 

  • Quine, W. V. (1992). Pursuit of truth. Cambridge, MA: Harvard University Press. (revised edition).

    Google Scholar 

  • Quine, W. V. (1995a). From stimulus to science. Cambridge, MA: Harvard University Press.

  • Quine, W. V. (1995b). Naturalism; or, living within one’s means. Dialectica, 49, 251–261.

  • Rorty, R. (1972). Indeterminacy of translation and truth. Synthese, 23, 443–462.

    Article  Google Scholar 

  • Searle, J. (1987). Indeterminacy, empiricism, and the first person. The Journal of Philosophy, 84, 123–146.

    Article  Google Scholar 

  • Soames, S. (1999). The indeterminacy of translation and the inscrutability of reference. Canadian Journal of Philosophy, 29, 321–370.

    Google Scholar 

  • Taylor, D. E. (2013). Essays on indeterminacy. Ph.D. dissertation, Stanford University.

  • van Cleeve, J. (1992). Semantic supervenience and referential indeterminacy. The Journal of Philosophy, 89, 344–361.

    Article  Google Scholar 

Download references


I’d like to thank the following people for helpful feedback on earlier drafts of this paper: Alexi Burgess, Rahul Chaudhri, Marcello Di Bello, Dagfinn Føllesdal, Peter Hanks, David Hills, Pedro Jimenez, Ken Taylor and three anonymous referees. Thanks also to the participants of the Philosophy Department Weekly Meetings at the University of Minnesota, where some of this material was presented in the fall of 2012. Finally, I owe a special debt to Mark Crimmins, who provided me with invaluable comments and advice on a number of earlier drafts.

Conflict of interest

The author declares that he has no conflict of interest.

Author information

Authors and Affiliations


Corresponding author

Correspondence to David E. Taylor.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Taylor, D.E. Quine on matters of fact. Synthese 193, 605–636 (2016).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: