Abstract
Kripke holds the thesis that identity statements containing natural kind terms are if true, necessarily true; these statements can be denominated theoretical identities. Kripke alleges that the necessity of theoretical identities grounds on the linguistic feature that natural kind terms are rigid designators. Nevertheless, I argue that the conception of natural kind terms as rigid designators, in one of their most natural views, hinders the establishment of the truth of theoretical identities and thus of their necessity. However, in Kripke’s works another proposal, not linguistic but metaphysical, is found to justify the presumed necessity of theoretical identities; it grounds on essentialism concerning natural kinds. In this regard, I question some of Kripke’s main claims, focusing on one of the main examples of theoretical identities put forward by Kripke, i.e., “Water is H\(_2\)O”. I challenge his a priori claims concerning what should be the essence of a natural kind like water. Furthermore, I adduce that the character of that theoretical identity is not that claimed by Kripke, since in the term flanking the right side of the identity sign it has to be resorted to the notion of similarity or it should have the form of a disjunction of a cluster of substances.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Another similarity that, according to Kripke, exists between proper names and natural kind terms is that the descriptivist theory conceived as a meaning theory is not adequate with respect to either sorts of terms—neither of them are synonymous with descriptions or clusters of descriptions. As we will see, this similarity is related to that consisting in the rigidity of both sorts of terms (see note 4).
Kripke asserts that “since names are rigid de jure [...] I say that a proper name rigidly designates its referent even when we speak of [...] [possible worlds] where that referent would not have existed” (1980: 21, n. 21).
Thus, if proper names were synonymous with descriptions or clusters of descriptions, most names would not be rigid designators; the same consideration applies to natural kind terms (see note 1).
This consideration would also be applicable to the monadic predicates built with those general terms and with which such descriptions have been formed.
In the bibliography, other types of kind terms are distinguished from natural kind terms; for instance, artifactual kind terms and sociolegal kind terms (see Devitt and Sterelny 1999: 93 ff.). All of these terms belong to the category of general terms that are common nouns. Color terms, in some of their uses, are also sometimes considered as kind terms.
For three different versions of essentialist predicates see Gómez-Torrente (2006); in this paper it is argued that the essentialist view can justify the claim that theoretical identities—understood in Soames’ way aforementioned—are necessary if true on the condition that one assumes possibilist quantification and that essentialist predicates are obstinately essentialist; a predicate is obstinately essentialist if in case it applies to an object at a possible world, it also applies to the object at all possible worlds, even with respect to worlds at which the object does not exist. In my view, this is the most suitable proposal to justify the claim that theoretical identities are necessary if true, under the assumptions that they have the form of universally quantified conditionals or biconditionals and that one accepts essentialism concerning natural kinds. I will question the first assumption, considered as Kripke’s view of the form of theoretical identities, in Sect. 4 and the second in Sect. 5.
According to some of Kripke’s assertions (see, e.g., 1980: 136), he has to sustain that the fact that a world contain instances of a natural kind is at least a necessary condition for the kind to exist in that world.
This condition can be reformulated as having the form of a universally quantified biconditional if we universally quantify over natural kinds, e.g., for all natural kinds A and B, they are identical if and only if the instances of those kinds are the same in all possible worlds.
Soames has proposed one of the conditionals constituting that biconditional, the one read from right to left, substituting “if and only if” by “if” and with the obvious reformulation, as providing sufficient conditions for the identity of natural kinds (see Soames 2004: 165 f., 2007: 338). It is noteworthy that in the formulation of those—necessary and sufficient or only sufficient—conditions for the identity of natural kinds the instances of the kind are being taken into account and hence also the extension of the corresponding general terms—or predicates in Soames’s view before (2006b). Thus, according to this type of proposals there is a close relation between the reference and the extension of natural kind terms. A different condition for the identity of natural kinds is that two natural kinds are identical if and only if one of them is the essence of the other. Kripke seems to have assumed this view; see Sect. 5.
The claim that natural kinds do not necessarily exist is grounded on Kripke’s view on possible worlds according to which possible worlds are not discovered, but stipulated, since “a possible world is given by the descriptive conditions we associate with it” (1980: 44). Thus, we can describe a possible world beginning with the following words: “Let us imagine a possible world that does not contain instances of water”. A similar stipulation could be made regarding the instances of any natural kind on the condition that the description of the possible world in question does not imply, explicitly or implicitly, the existence of such type of instances.
Many interpreters of Kripke propose the conception of natural kinds, and in general of kinds, as universals or abstract entities, especially with the aim of making it possible for natural kinds to be rigid designators. One of the main authors who consider them as universals is N. Salmon (see e.g., Salmon 1981: 43, 45 f., 53 ff. and 71 f., 2003: 481 and 487, 2005: 120 and 133 f., as well as 2012: 472). LaPorte in (2000, 2004) and (2006) denominates such abstract entities “kinds” or “abstract kinds” and in (2013) “properties”—in this last paper he conceives of kinds as properties; the last denomination is used by Linsky in (1984, 2006, 2011). Kaplan and Donnellan are other authors who hold that the best way to interpret the claim that natural kind terms are rigid designators is to conceive their designata as abstract entities; see Kaplan (1973: 518, n. 31) and Donnellan (1973: 712, 1983: 90). It is noteworthy that none of these authors commit themselves with a detailed view of the ontological nature of those abstract entities. In the following, I generally use Salmon’s terminology of universals as the abstract entities in question.
The first objection, put forward among others by Schwartz (2002) and Soames (2002: 250 f.), is that sameness of universals designated at all possible worlds turn all general terms into rigid designators, including those general terms that are apparently non-rigid, since they will also have a rigid reading. Thus, the general term “liquid that is colorless, transparent, tasteless, thirst-quenching, falls as rain and fills the lakes and rivers”, which has a non-rigid reading, could be considered, though, as (rigidly) designating with respect to all possible worlds the universal of being the liquid that is colorless, transparent, tasteless, thirst-quenching, falls as rain and fills the lakes and rivers. This assumed, the problem can be put in a general way as that of how to distinguish between the rigid and non-rigid readings of those general terms that are liable to a non-rigid reading. For convincing replies to such objection, see LaPorte (2013) as well as Martí/Martínez-Fernández (2010, 2011).
See, e.g., Salmon (1981: 71 f., 2003: 480 f., 2005: 119), Soames (2002: 260) as well as Martí/Martínez-Fernández (2010: 51 f.) Many other authors, among which are Kaplan (1973), Donnellan (1983), Laporte and Linsky, recognize that this is a natural consequence of the conception of the referents (or designata) of general terms mentioned in note 12.
Some authors do not distinguish between the trivialization and the overgeneralization problem. In fact, the latter can be considered partly as an aspect of the former; according to the overgeneralization problem many more terms than those considered by Kripke as rigid become rigid designators.
Here and in the following by “semantically simple” terms I mean terms that do not contain meaningful parts or, if they do, those whose meaning (and reference) is not determined by the meaning (and reference) of their component parts. I will use “semantically compound” as the adjective opposed to “semantically simple”.
In all these examples I would delete the definite article to obtain what is indisputably a general term, and not a definite description that when used predicatively can be regarded as a general term.
However, the latter assertion is susceptible to a caveat. There are some construals of definite descriptions, and also of semantically compound general terms, according to which they are rigid de jure and hence obstinate designators (see LaPorte in 2013: 139 f.; for his conception of them as rigid de facto see 2013: 137–139). Nevertheless, leaving aside the possible differences between the stipulations involved in the introduction of proper names and those that can be established for general terms, and specially for compound general terms, and even accepting that through some stipulations any rigid description could be regarded as rigid de jure, given the distinction of rigidity for singular terms put forward by Kripke, i.e., proper names are rigid de jure while rigid definite descriptions are rigid de facto, I prefer to apply that distinction to general terms, although as I will say in the next paragraph I will not put much weight on it in the debate concerning the status of theoretical identities.
In Salmon’s framework all designators of kinds are obstinate (see Salmon 1981: 72 f.). In Soames (2002: 259 ff.) it is taken into consideration another possible interpretation of the notion of rigidity for predicates—besides the one above-mentioned according to which rigid predicates are essentialist predicates, in which the rigidity of predicates derives from the rigidity of some corresponding singular term, and in his definition of such predicates he considers rigid predicates as persistent (Soames 2002: 260)—he rejects, though, this interpretation since, among other reasons, it overgeneralizes the notion of rigidity, which he considers as a decisive argument against that view. On the other hand, Soames’ essentialist predicates and Devitt’s rigid appliers in Devitt (2005)—both aforementioned—are persistent. As already said, Gómez-Torrente (in 2006), assuming possibilist quantification, claims that natural kind predicates are obstinately essentialist, although his arguments for that claim are far from conclusive. Moreover, he concedes that “our intuitions about obstinacy and persistence are somewhat flimsy” (Gómez-Torrente 2006: 252).
In (2006b: 713–774) Soames admits that the statement “Water is \(\hbox {H}_{2}\hbox {O}\)” can be considered as an identity statement where “water” and “\(\hbox {H}_{2}\hbox {O}\)” are general terms, but he claims that this view is not as a rule applicable to other theoretical identities mentioned by Kripke; according to my assertions below I disagree with that claim.
Kripke asserts that “science attempts, by investigating basic structural traits, to find the nature, and thus the essence (in the philosophical sense) of the kind” (1980: 138), and he speaks of the “scientific discoveries of species essence” (ibid), where by “species” has to be understood natural kind(s).
Here we find the basis for the identity condition of natural kinds mentioned in note 10: two natural kinds are identical if and only if one of them is the essence of the other. Nonetheless, given my skepticism about the justifiability of essentialism and about the—a priori—character of our judgements about the sort of essential properties of an entity I proposed another identity condition which does not involve such claims.
In order to justify Kripke’s thesis that the statement “Water is \(\hbox {H}_{2}\hbox {O}\)” is necessary given its truth, other authors, like J. LaPorte and S. Soames, link the notions of rigidity and essentialism. See, e.g., LaPorte (2013) as well as Soames (2002, 2006a, 2007, 2011). Salmon has convincingly argued that Kripke’s theses of the essentiality of origin for material objects and of the essentiality of chemical composition for substances are not consequences of Kripke’s reference theory but involve hidden nontrivial essentialist premises (see Salmon 1981: 193 ff. and 176 ff.).
However, this claim can be questioned (see below). Nevertheless, for the sake of the argument I assume, as I have done so far, that in the actual world instances of water are instances of \(\hbox {H}_{2}\hbox {O}\).
I claim that the atomic structure of water (in the actual world) might not be accepted as essential to water because there are philosophers of chemistry, like Needham and van Brakel—see, e.g. Needham (2011) and van Brakel (2000), who maintain that the macroscopic, physical properties are the essential properties of chemical substances, thus microstructuralism is not the only view in the field. Precisely one of the most important debates at present in Philosophy of chemistry is the one between that macroscopic view of chemical substances and microstructuralism. And this debate cannot be solved alluding to two senses of necessity, natural or metaphysical—the latter, the one involved in Kripke’s remarks, resorts to possible worlds whose laws can be different from the natural laws in the actual world, since in Philosophy of chemistry the relevant sense of necessity is natural necessity. Kripke’s claim that the essence of a substance is its atomic structure is based on (his) intuitions—probably this is at least in part the source of the mentioned a priority, but he does not take into consideration views on Philosophy of chemistry that disagree with this essentialist claim.
Tritium, unlike protium and deuterium, is a radioactive isotope. There are other isotopes of hydrogen that have been synthesized, but never occur in nature.
Rather, three stable isotopes, since oxygen has also other, though radioactive and short-lived, isotopes.
I leave aside considerations about the quantum-mechanical superpositions of such compounds involved in the structure of water.
Gold (Au) has, besides several radioactive isotopes, one stable isotope, \(^{197}\)Au, which is also its only naturally occurring isotope; thus samples of gold in nature can also be a disjunction of substances in which the element and its stable isotope appear.
A referee has alleged that speaking of a disjunction of substances is speaking of a complex formula and that unless one postulates complex properties, one should not be speaking here of an identity statement but rather of a biconditional. However, one can form an identity statement with the help of the Lambda operator. The identity statement in question would have the following form: Water is (identical to) (Lambda x (x is H\(_{2}^{16}\)O or x is H\(_{2}^{17}\)O or x is H\(_{2}^{18}\)O or...)). In any case, theoretical identities understood as identity statements would have a much more complex form than the one proposed by Kripke.
References
Devitt, M. (2005). Rigid application. Philosophical Studies, 123, 139–165.
Devitt, M., & Sterelny, K. (1999). Language and reality. In An introduction to the philosophy of language. Cambridge, MA: MIT Press (2nd revised and enlarged edition; 1st edition 1987).
Donnellan, K. (1973). Substances as individuals. Journal of Philosophy, 70, 711–712.
Donnellan, K. (1983). Kripke and Putnam on natural kind terms. In S. Ginet & S. Shoemaker (Eds.), Knowledge and mind: Philosophical essays (pp. 84–104). Oxford: Oxford University Press.
Fernández Moreno, L. (2016). The reference of natural kinds. Frankfurt am Main: Peter Lang.
Gómez-Torrente, M. (2006). Rigidity and essentiality. Mind, 115, 227–259.
Kaplan, D. (1973). Bob and Karol and Ted and Alice. In J. Hintikka (Ed.), Approaches to natural language (pp. 490–518). Dordrecht: Reidel.
Kaplan, D. (1989). Afterthoughts. In J. Almog, et al. (Eds.), Themes from Kaplan (pp. 565–614). Oxford: Oxford University Press.
Kripke, S. (1971). Identity and necessity. In M. Munitz (Ed.), Identity and individuation (pp. 135–164). New York: New York University Press.
Kripke, S. (1980). Naming and necessity. Oxford: Blackwell (Revised and enlarged edition, first published In D. Davidson & G. Harman (Eds.), Semantics of natural language. Dordrecht: Reidel, 1972).
LaPorte, J. (2000). Rigidity and kind. Philosophical Studies, 97, 293–316.
LaPorte, J. (2004). Natural kinds and conceptual change. Cambridge: Cambridge University Press.
LaPorte, J. (2006). Rigid designators for properties. Philosophical Studies, 130, 321–336.
LaPorte, J. (2013). Rigid designation and theoretical identities. Oxford: Oxford University Press.
Linsky, B. (1984). General terms as designators. Pacific Philosophical Quarterly, 65, 259–276.
Linsky, B. (2006). General terms as rigid designators. Philosophical Studies, 128, 655–667.
Linsky, B. (2011). Kripke on proper and general terms. In A. Berger (Ed.), Saul Kripke (pp. 17–48). Cambridge: Cambridge University Press.
Martí, G., & Martínez-Fernández, J. (2010). General terms as designators: A defence of the view. In H. Beebee & N. Saabarton-Leary (Eds.), The semantics and metaphysics of natural kinds (pp. 46–63). London: Routledge.
Martí, G., & Martínez-Fernández, J. (2011). General terms, rigidity and the trivialization problem. Synthese, 181, 277–293.
Needham, P. (2011). Microessentialism: What is the argument? Noûs, 45, 1–21.
Salmon, N. (1981). Reference and essence. Princeton: Princeton University Press (Reprinted in N. Salmon, Reference and essence, 2nd edn. Amherst: Prometheus Books, enlarged with 7 Appendixes, 2005).
Salmon, N. (2003). Naming, necessity, and beyond. Mind, 112, 475–492.
Salmon, N. (2005). Are general terms rigid? Linguistics and Philosophy, 28, 117–134.
Salmon, N. (2012). Generality. Philosophical Studies, 161, 471–481.
Schwartz, S. P. (2002). Kinds, general terms and rigidity: A reply to LaPorte. Philosophical Studies, 109, 265–277.
Soames, S. (2002). Beyond rigidity. The unfinished semantic agenda of naming and necessity. Oxford: Oxford University Press.
Soames, S. (2004). Knowledge of manifest natural kinds. Facta Philosophica, 6, 159–181.
Soames, S. (2006a). The philosophical significance of the Kripkean necessary a posteriori. In E. Sosa & E. Villanueva (Eds.), Philosophy of language (Philosophical Issues, 16) (pp. 288–309). Oxford: Blackwell.
Soames, S. (2006b). Reply to critics. Philosophical Studies, 128, 711–718.
Soames, S. (2007). What are natural kinds? Philosophical Topics, 35, 329–342.
Soames, S. (2011). Kripke on epistemic and metaphysical possibility: Two routes to the necessary a posteriori. In A. Berger (Ed.), Saul Kripke (pp. 78–99). Cambridge: Cambridge University Press.
Van Brakel, J. (2000). Philosophy of chemistry. Between the manifest and the scientific image. Leuven: Leuven University Press.
Weisberg, M. (2006). Water is not \(\text{ H }_{2}\text{ O }\). In D. Baird, et al. (Eds.), Philosophy of chemistry: Synthesis of a new discipline (pp. 337–345). Dordrecht: Springer.
Acknowledgements
This paper modifies and develops ideas put forward in Fernández Moreno (2016). The author is grateful to the comments made by an anonymous referee. This paper has been supported by the Spanish Ministry of Economy and Competitiveness in the framework of the research Project FFI2014-52244-P.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Fernández Moreno, L. Language and metaphysics: the case of theoretical identities. Synthese 198 (Suppl 3), 831–848 (2021). https://doi.org/10.1007/s11229-017-1498-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11229-017-1498-5