Abstract
We once thought of ‘water is H2O’ as being synthetic, knowable only a posteriori, and expressing a contingent truth. We once thought of ‘bachelors are unmarried’ as being analytic, knowable a priori, and expressing a necessary truth. We once thought of ‘water is clear’ as akin to the latter, that is, as being analytic, knowable a priori, and expressing a necessary truth. Then Putnam (1962a, 1975) introduced semantic externalism. We learned that ‘water is H2O’ expresses a necessary truth despite being synthetic. We learned that ‘water is clear’ expresses a contingent truth even if it is analytic. With some refinement from Kripke (1972/1980), we learned that ‘water is H2O’ expresses a necessary truth even though it is only knowable a posteriori, and that ‘water is clear’ expresses a contingent truth even if it is knowable a priori. We learned that this is all so because ‘water’, unlike ‘bachelor’, is a natural kind term, and thus, its extension is determined at least partly by the “external” (i.e., extra-psychological) world. More broadly, the discovery of semantic externalism taught us that the categories analytic, a priori, and necessary are not coextensive, even though they coincide for textbook analyticities such as ‘bachelors are unmarried’.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
Hume (1748/1961) inspires a version of modal conventionalism according to which relations among ideas, not matters of fact, determine whether any given statement expresses a truth that is necessary, and any justifiable belief about relations among ideas can arise from “inward sentiment”, rather than “outward sense”, and thus, is justifiable a priori (p. 360). On this view, the necessity of any necessary truth reduces to relations among ideas that are expressed by a statement that expresses that truth and is knowable a priori. Ayer (1946) offers a version of modal conventionalism according to which a conventionally determined aspect of the meanings of any given statement’s constituent terms, not the natures of the entities to which those terms refer, determines whether that statement expresses a truth that is necessary, and thus, any statement that expresses a necessary truth is analytic. On this view, necessity reduces to analyticity. Since differences between these views are not relevant for present purposes, I often focus on “Ayer-style” modal conventionalism.
- 2.
- 3.
For one who worries that Oscar and Twin Oscar would surely know relevant chemical facts if they were competent with ‘water’, Putnam asks us to consider Oscar1 and Twin Oscar1, who live before the chemical structures of respective samples of clear, potable liquid were discovered. For one who worries that Oscar1 and Twin Oscar1 could distinguish H2O from XYZ if their chemistries differ, Putnam asks us to suppose that neither has explored his clear, potable liquid outside normal temperatures and pressures, wherein H2O and XYZ behave the same insofar as Oscar and Twin Oscar can investigate them. More broadly, Putnam’s thought-experiment is coherent if each of Oscar and Twin Oscar can understand his ‘water’ without thereby differing psychologically (in the narrow sense).
- 4.
Others have offered similar analyses of the semantics of ‘water’ (e.g., Bealer 2002; Korman 2006), others are committed to such conditionals (e.g., McLaughlin and Tye 1998), and still others have suggested that identifying the extension of ‘water’ requires affirming the antecedent of such conditionals while denying that the conditionals are implied by the semantics (e.g., Block and Stalnaker 1999). As this suggests, these conditionals can be understood in two quite different ways. First, perhaps the conditionals capture the complex intension of ‘water’. Second, perhaps the conditionals provide metasemantic guidelines for choosing a single, simple intension for ‘water’. On this way of thinking, we should append the phrase “otherwise ‘water’ is empty” to each of (W1)–(WN-1), leaving (WN) as is, resulting in a series of distinct, simple intensions, only one of which ‘water’ has—where which it has depends on which of (W1)–(WN) has a true antecedent. For ease of presentation, and because I prefer it, I presume the former interpretation. The central points of this essay can be reframed to suit the latter interpretation (cf. Biggs in progress). Of course, some reject this kind of semantic externalism, regardless of how it is understood. Opponents include both those who reject semantic externalism outright (e.g., Wikforss 2005) and externalists who hold that any natural kind term is empty unless there is a relevant natural kind (e.g., Nuccetelli 2003). My assessment presumes semantic externalism.
- 5.
That is, what Putnam (1975) says suggests that these are on all fours. Putnam (1962b) delivers a different verdict. While considering a case in which all bachelors share a particular neurosis (p. 58), Putnam (1962b) suggests that words only lose their conventional character when their referents figure into scientific explanations and laws, and that “one is simply not going to find any laws … about [bachelors]” (p. 57), which suggests that he thinks that ‘bachelor’ retains its conventional character, and consequently, its semantics doesn’t include conditionals such as (B1). Of course, one is not going to find any laws about pediatricians either. Perhaps, then, Putnam changed his mind. This possibility fits other contrasts between these papers—e.g., whereas Putnam (1962b) emphasizes that words introduced by stipulation to have a certain meaning don’t invariably retain their “conventional character” (p. 38), Putnam (1975) emphasizes, as I note above, that “there is a strong tendency for words which are introduced as “one criterion” words to develop a “natural kind” sense” (my italics, p. 163). Although this isn’t the place for an extended discussion of Putnam’s assessment of these cases, I think he should agree that the semantics of ‘water’, ‘pediatrician’, and ‘bachelor’ are on all fours, and whatever Putnam’s ultimate verdict, his semantic theory suggests that they are on all fours—see Biggs (in progress) for further discussion.
- 6.
Whether Kripke’s would embrace the extension is unclear because it is unclear whether he accepts that the semantics of natural kind terms includes “fallback conditions”; in discussing the empirical supposition that two distinct underlying kinds are prominent among samples that we have called ‘gold’, Kripke (1972/1980) says, “reactions can vary: sometimes we may declare that there are two kinds of gold, sometimes we may drop the term ‘gold’” (p. 136). Biggs (in progress) explores what Kripke should say about such cases and the proposed extension of semantic externalism, given his understanding of semantic externalism, finding that he should accept this extension.
- 7.
Putnam thinks that stereotypes are critical for fixing the reference of natural kind terms. This suggests that actual people whom we have correctly called ‘bachelor’ must satisfy the stereotype for ‘bachelor’, i.e., they must be unmarried males, which suggests that each of (B1)–(BN) implies that bachelors are unmarried males, which suggests, given the supposition semantic competence allows us to identify (B1)–(BN), that ‘bachelors are unmarred males is analytic (despite being contingent). Kripke, by contrast, does not accept that stereotypes fix the reference of natural kind terms. He notes that the stuff we have been calling gold may be blue, although we have seen it as yellow because of a pervasive optical illusion (1972/1980, p. 118). He notes that the animals we have been calling ‘tiger’ may in fact have three legs, although we have seen them as four-legged because of a pervasive optical illusion (p. 119–120). Similarly, one might maintain, it is consistent with the truth of (B1) that the people we have called bachelors are married or female, although we have seen them as unmarried males because of pervasive illusions. What matters for successful reference, on this way of thinking, is just that the referents share something sufficiently important (perhaps B, perhaps being unmarried males), not that they share being unmarried and being male. For more on this dialectic, including reasons to think that Kripke’s view is more similar to Putnam’s than this brief note suggests, see the discussion of primary intensions in Biggs (in progress).
- 8.
Biggs (in progress) develops these considerations.
- 9.
One might object that Putnam never uses his applications of ‘water’ to motivate his theory; rather, he deduces those applications from his theory. After all, the objector would emphasize, Putnam provides independent motivation for his theory, claiming that if it were false, then scientific realism would be hopeless. But Putnam does not argue as follows: “since realism implies my theory of natural kind terms, we should accept my theory even though it delivers counterintuitive results for ‘water’.” Rather, he uses his discussion of ‘water’ not only to elucidate his theory but also to motivate it—especially against traditional alternatives. At any rate, many people reject any theory that implies (T1)–(T2) because they take Putnam’s applications of ‘water’ to scenario-world pairs to match their own, rather than because they think they must accept his theory, despite its counterintuitive results, on pain of losing realism. That being said, one might think of Putnam’s method as providing a heuristic, maintaining that the ultimate arbiter of disputes about modal claims is something else entirely (see Biggs 2011; Biggs and Wilson 2016).
- 10.
If these statements are analytic and knowable a priori, then conditionals such as (W1)–(WN) can be altered to yield, for example: (W1*) If there is a unique hidden structure such that enough samples of the watery liquid around here have that hidden structure, then in all worlds ‘water’ refers to whatever and only whatever has that hidden structure, even in worlds where the most common watery liquid lacks that hidden structure. Similar alterations can be made to (P1)–(PN) and (B1)–(BN). I use the un-starred conditionals because they are less controversial (since they do not presuppose Putnam’s account of stereotypes), and are all that I require. But Putnam’s discussion of stereotypes suggests that he would entertain endorsing such “starred” alternatives, making his conditional analyticities even richer. Note that if Putnam were to endorse the starred conditionals, then ‘bachelors are unmarried’ would be analytic and knowable a priori since it would be true regardless of which of (B1)–(BN) were true. This analyticity would be little condolence for modal conventionalists since it would be disconnected from the relevant necessity—i.e., ‘bachelors are unmarried’ would express a necessary truth because of the truth of (BN) but would be analytic because of the conjunction of (B1)–(BN), which suggests that its necessity would not be reducible to its analyticity.
- 11.
Of course, Kripke would not think of such statements as analytic because he defines ‘analytic’ such that a statement is analytic only if it is necessary, but he may think of them as a priori. While Chalmers (2006) does not aim foremost to interpret Kripke, his epistemic interpretation of the two-dimensional semantic framework, which at least aims to be consistent with Kripke’s discoveries, endorses this kind of view. Others argue against this kind of interpretation of Kripke, e.g., Byrne and Pryor (2006).
- 12.
Specifically, I think that inference to the best explanation can yield justified beliefs about mind-independent necessities (cf. Biggs 2011) even though it is an “a priori mode of inference” (cf. Biggs and Wilson 2016, forthcoming) that consists of “appropriately a priori” theoretical virtues (cf. Biggs and Wilson forthcoming, 2016). One might not think of this approach as rationalist since it implies neither that we have a dedicated faculty for attaining modal knowledge nor that knowledge of real necessities stems from semantic competence (though it is consistent with the latter claim, cf. Biggs and Wilson 2016). This abductive approach is broadly rationalist, nonetheless, since it implies that a priori reasoning can amplify justified beliefs in non-modal statements (e.g., ‘all and only samples of water are samples of H2O’) into justified beliefs in modal statements (e.g., ‘water is necessarily H2O’).
- 13.
One who is inclined to move toward modal rationalism should note that the parity of ‘water’ and ‘bachelor’ might problematize some versions of modal rationalism as well. Chalmers (e.g., 2002) grants that (cf. Yablo 2002) his rationalism “has something in common with some versions of logical empiricism” (p. 184); his rationalism presumes, e.g., that every statement that expresses a necessary truth is either knowable a priori on the basis of semantic competence alone or corresponds to an underlying statement that is knowable a priori on the basis of semantic competence alone. One might predict, then, that Chalmers’s rationalism would struggle to deal with the parity of ‘water’ and ‘bachelor’ no less than modal conventionalism does. Considering the accuracy of this prediction is beyond the scope of the present essay.
- 14.
Van Fraassen’s invocation of theories to explain modality should not be confused with Fischer’s (Chap. 14, this volume, and 2016). Fischer maintains a broadly realist metaphysics of modality and suggests that we can justifiably believe modal statements by believing them on the basis of (e.g., inferring them from, intuiting them because of) ordinary theories that we justifiably believe. Van Fraassen, contrariwise, rejects any realist metaphysics of modality, holding (though he might reject this formulation of his view) that modal entities reduce to our theories.
References
Ayer, A. J. (1946). Language, truth, and logic. Mineola: Dover Publications.
Bealer, G. (2002). Modal epistemology and the rationalist renaissance. In T. Gendler & J. Hawthorne (Eds.), Conceivability and possibility. Oxford: Oxford University Press.
Biggs, S. (2011). Abduction and modality. Philosophy and Phenomenological Research, 83, 283–326.
Biggs, S. In progress. Pervasive Externalism.
Biggs, S., & Wilson, J. (Forthcoming). The a priority of abduction. Philosophical Studies.
Biggs, S., & Wilson, J. (2016). Carnap, the necessary a priori, and metaphysical anti-realism. In S. Blatti & S. Lapointe (eds.), Ontology after Carnap. Oxford: Oxford University Press.
Block, N., & Stalnaker, R. (1999). Conceptual analysis, dualism, and the explanatory gap. Philosophical Review, 108, 1–46.
Boghossian, P. (1996). Analyticity reconsidered. Noûs, 30, 360–391.
Byrne, A., & Pryor, J. (2006). Bad intensions. In M. Garcia-Carpintero & J. Macià (Eds.), Two-dimensional semantics: Foundations and applications (pp. 38–54). New York: Oxford University Press.
Chalmers, D. (2002). Does conceivability entail possibility. In T. Gendler & J. Hawthorne (Eds.), Conceivability and possibility. Oxford: Oxford University Press.
Chalmers, D. (2006). Two-dimensional semantics. In E. Lepore & B. Smith (Eds.), Oxford handbook of the philosophy of language (pp. 574–606). New York: Oxford University Press.
Fischer, R. W. (2015). Theory selection in modal epistemology. American Philosophical Quarterly, 52(2): 381–395.
Fischer, B. (2016). A theory-based epistemology of modality. Canadian Journal of Philosophy, 46, 228–247.
Hume, D. (1748/1961). An enquiry concerning human understanding. In The empricists (pp. 307–430). New York: Doubleday.
Korman, D. (2006). What externalists should say about dry earth. The Journal of Philosophy, 103, 503–520.
Kornblith, H. (1980). Referring to artifacts. The Philosophical Review, 89, 109–114.
Kripke, S. (1972/1980). Naming and necessity. Princeton: Princeton University Press.
McLaughlin, B., & Tye, M. (1998). Is content-externalism compatible with privileged access? The Philosophical Review, 107, 349–380.
Nuccetelli, S. (2003). Knowing that one knows what one is talking about. In S. Nuccetelli (Ed.), New essays on semantic externalism and self-knowledge. Cambridge, MA: MIT Press.
Putnam, H. (1962a). It ain’t necessarily so. Journal of Philosophy, 59, 658–671.
Putnam, H. (1962b). The analytic and the synthetic. In H. Putnam (Ed.), Mind, language, and reality: Philosophical papers (Vol. 2, 1975, pp. 33–69). Cambridge: Cambridge University Press.
Putnam, H. (1975). The meaning of meaning. In K. Gunderson (Ed.), Language, mind, and knowledge (Minnesota studies in philosophy of science, Vol. VII, pp. 131–193). Minneapolis: University of Minnesota Press.
Quine, W. V. O. (1951). Two dogmas of empiricism. Philosophical Review, 60, 20–43.
Quine, W. V. O. (1968). Ontological relativity. Journal of Philosophy, 65, 185–212.
Russel, B. (1927/2001). The analysis of matter. New York: Routledge.
Russell, G. (2010). A new problem for the linguistic doctrine of necessary truth. In C. Wright & N. Pederson (Eds.), New waves in truth (pp. 267–281). New York: Plagrave McMillan.
Schwartz, S. (1978). Putnam on artifacts. The Philosophical Review, 87, 566–574.
Sidelle, A. (1989). Necessity, essence, and individuation. Ithaca: Cornell University Press.
Van Fraassen, B. C. (1980). The scientific image. New York: Oxford University Press.
Van Fraassen, B. C. (1989). Laws and symmetry. New York: Oxford University Press.
Wikforss, A. (2005). Naming natural kinds. Synthese, 145, 65–87.
Williamson, T. (2007). The philosophy of philosophy. Malden: Wiley-Blackwell.
Yablo, S. (1992). Review of necessity, essence, and individuation: A defense of conventionalism by Allen sidelle. The Philosophical Review, 101, 878–881.
Yablo, S. (2002). Coulda, woulda, shoulda. In T. Gendler & J. Hawthorne (Eds.), Conceivability and possibility. Oxford: Oxford University Press.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing Switzerland
About this chapter
Cite this chapter
Biggs, S. (2017). Modal Conventionalism and Textbook Analyticities. In: Fischer, B., Leon, F. (eds) Modal Epistemology After Rationalism. Synthese Library, vol 378. Springer, Cham. https://doi.org/10.1007/978-3-319-44309-6_9
Download citation
DOI: https://doi.org/10.1007/978-3-319-44309-6_9
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-44307-2
Online ISBN: 978-3-319-44309-6
eBook Packages: Religion and PhilosophyPhilosophy and Religion (R0)