SciELO - Scientific Electronic Library Online

Home Pagealphabetic serial listing  

Services on Demand




Related links



Print version ISSN 0100-6045On-line version ISSN 2317-630X

Manuscrito vol.43 no.4 Campinas Oct./Dec. 2020  Epub Nov 09, 2020 




1Instituto de Investigaciones Filosóficas. Universidad Nacional Autónoma de México (UNAM). Mexico City Mexico.


Luis Fernández Moreno has given a number of arguments that descriptive knowledge or stipulations have a greater role in the fixing of the reference of natural kind terms than I allow in my book Roads to Reference. In this note I criticize Fernández Moreno’s arguments.

Keywords: Natural Kind Terms; Descriptivism. Reference

Luis Fernández Moreno (2020) says that, although in Roads to Reference I note several differences between Kripke’s and Putnam’s theories of natural kind terms, I don’t mention some of the more important of those differences. That may be so, but I doubt that the example he focuses on is a good one. He says that

since Putnam assumes that the average or typical speakers are willing to defer to experts the reference fixing of natural kind terms as typical speakers use them, an entity falls into the extension of a natural kind term used by the average speaker if it falls into its extension as used by experts.

If by this he means that Putnam postulates that the fixing of the reference of ordinary natural kind terms is determined by the opinions of special experts (when they exist), I don’t think he is right. Although Putnam does think that some experts may be better than ordinary speakers at telling whether something falls in the extension of a certain natural kind term, and that there is “linguistic division of labor” in this limited sense, this does not mean that he thinks that (when they exist) the experts’ opinion has a special determinative role with regard to the fixing of the reference of the term: in the case of ordinary natural kind terms, intentions and beliefs of non-experts typically fix their reference independently of the availability of experts and of their opinions. Fernández Moreno does not refer to any specific passage by Putnam on this topic, though he refers us to the short note Kripke (1986), where, according to him, “Kripke does not accept the role played by the notion of expert in Putnam’s theory”. But I think he has misread the short note: Kripke (1986) emphasizes that Putnam’s theory is not different from his in this respect, even if Putnam’s use of the expression “linguistic division of labor” has often carried for others the suggestio falsi that it is. (Kripke does note that as a result of the Kripke-Putnam discoveries, descriptivists like Dummett did come to reserve a reference-fixing role for experts inside the descriptions that they took as reference-fixing.)

My own theory in Roads to Reference does not differ from Kripke’s or Putnam’s theories in not giving a special determinative role to the experts’ intentions and beliefs in the typical fixing of the reference of ordinary kind terms. However, the theory gives an important role to the beliefs and intentions of the community of speakers at large in the successful introduction, transmission and adoption of a term, and so it leaves room for the possibility that the beliefs and intentions of experts may influence the beliefs and intentions of the community of speakers at large in such a strong way that those beliefs and intentions become crucial for the determination of the reference of a kind term in a speaker’s mouth. To give and example, imagine that someone receives the term “water” from other members of the community and forms the implicit intention of using “water” as the community uses it, without forming other possibly conflicting or overriding referential intentions concerning “water”, and for some strange reason concurrently develops the belief that water has nothing to do with H2O (against the current belief of the community at large that water is H2O, originated in expert knowledge) and instead is an extraterrestrial metal alloy. It is at least dubious that this person thereby comes to refer to water by “water”, and the theory in Roads to Reference is compatible with this impression.

Fernández Moreno says that “Putnam, unlike Kripke, supports the necessity of descriptive requirements on the reference transmission of a term”, and that, like Putnam, “Gómez-Torrente includes some descriptive elements in the conventions for successful proper name and natural kind noun transmission”. When making the first claim he has in mind a passage from Putnam (1973) where Putnam says, in Fernández Moreno’s selection,

I do not feel that one should be quite as liberal as Kripke is with respect to the causal chains one allows. (...) unless one has some beliefs about the bearer of the name that are true or approximately true, then it is at best idle to consider that the name refers to that bearer in one’s idiolect. (Putnam (1973), 203)

(Putnam makes it clear that he has in mind non-trivial beliefs, e.g. he does not include among the required beliefs the belief that the bearer of a certain transmitted proper name is a person). And when making the second claim, Fernández Moreno has in mind the fact that, e.g., in my proposed convention for natural kind term reference transmission,2 the expression “a certain substance” appears (and similarly for “an object” in my corresponding convention for proper name reference transmission).

There are several things that ought to be clarified here. First, the fact that “a certain substance” appears in my proposed convention for natural kind term reference transmission cannot be called “a descriptive element” of my theory, insofar as “a certain substance” is only employed in the statement of the impersonal condition “if a natural kind noun N as used by a community of speakers refers to a certain substance”, not in any attribution to speakers of descriptive knowledge associated with a natural kind noun. Second, although I do think that speakers who inherit successfully a natural kind term or a proper name from other members of their linguistic community must have intentions and beliefs necessarily containing concepts of various sorts, I reject any idea that in a successful transmission a true description of any kind must perforce be associated with the inherited noun by the inheriting speaker. Third, my proposed conventions for successful transmission leave it as a possibility that the transmission may not be accomplished successfully if the inheriting speaker forms at the transmission stage a set of beliefs involving the inherited noun that on the whole conflict with the beliefs of the community of users of the term at large. Thus, to use an example (involving a proper name) of Putnam (1973) from the page from which Fernández Moreno extracts the text he quotes, I may receive the name “Quine” together with the absurd belief, widely rejected by the community, that Quine is a Roman emperor, and in such a case it is dubious that I from then on will be referring to Quine with my uses of “Quine”. My theory is compatible with this impression.

Fourth, Putnam’s claim that “unless one has some [non-trivial] beliefs about the bearer of the name that are true or approximately true, then it is at best idle to consider that the name refers to that bearer in one’s idiolect”, inspired precisely by his “Quine” example, seems clearly false, as shown by Kripke’s “Newton” example (Kripke (1972), 95), where the community transmits the name “Newton” despite having only one widespread non-trivial but false belief about Newton. Note that, by constrast, my convention gives as a roughly sufficient condition for successful transmission (in the absence of other conflicting beliefs) the absence of beliefs involving the inherited noun that on the whole conflict with the beliefs of the community of users of the term at large, independently of whether these beliefs are true or false. Here it is worth noting that Putnam’s “Quine” example is similar to Kripke’s “George Smith” example in Naming and Necessity, and that Putnam, as he himself says in Putnam (1973), gave it and extracted consequences from it before having access to the text of Naming and Necessity and thus to the Kripkean examples, including the “Newton” example. Kripke was certainly fully aware of examples like Putnam’s “Quine” example, but in this particular case refrained from postulating “descriptive requirements on the reference transmission of a term”, in view of examples like the more relevant “Newton” case.

Fernández Moreno says that

we could ask Gómez-Torrente for a characterization of natural kinds applicable to both scientific kinds and ordinary natural kinds - in (2019) he mainly deals with their differences - as well as for more precision in his assertion that ‘on the view I am inclined to adopt, ordinary kinds are sui generis universals’.

My conception of natural kinds is simply the intuitive one of a natural kind as a naturally occurring stuff, genus of living beings, or physical phenomenon. Among natural kinds I speak of the ordinary ones as those that have apparently been singled out by ordinary speakers of natural languages, previously to scientific investigation; and of scientific kinds as those that have been singled out through procedures involving scientific investigation. I should add that some scientific kinds could perhaps have been singled out by ordinary speakers, but also that, as I see things in Roads to Reference, paradigmatic scientific kinds like H2O (which I postulate to be different from the ordinary kind water) could not have been singled out without the intervention of scientific investigation. Now when I say that I’m inclined to see ordinary kinds as sui generis universals, what I have in mind is, first, that, unlike metaphysicians like E. J. Lowe, I am inclined to see ordinary kinds as universals and not as entirely sui generis entities not identical either to properties or individuals; and second, that, unlike some philosophers of science like R. Boyd and K. Hawley & A. Bird, I don’t think natural kinds in general, and ordinary natural kinds in particular, are to be identified with clusters of individually non-necessary but jointly sufficient properties (for membership in the kind). The positive ideas in Roads to Reference suggest, by contrast, that ordinary kinds are not to be identified with clusters of properties, and have their own mode of being, so to speak, part of which is often constituted by facts of composition in terms of kinds discoverable only by scientific investigation.

Fernández Moreno suggests that a limited but in a way strong form of descriptivist thesis may hold for proper names and natural kind terms, namely that, properly viewed, their introduction proceeds always by description. In order to dispose of two apparent obstacles to this suggestion, he claims that apparently non-descriptive introductions by ostension could be subsumed under introductions by description; and that cases of introduction via a description that is false of the eventually named object are in fact (as suggested apparently by Kripke himself) cases where some implicit requirement is in operation that there be an introducing description and that it be at least “approximately true” of the named object. When thinking of these latter cases, Fernández Moreno has in mind especially Kripke’s example of a hypothetical introduction of “Phosphorus” via the (untrue) description “the morning star” (Venus is not a star), in which nevertheless Venus becomes intuitively the referent of “Phosphorus”. As for the first obstacle, my own view is that cases of introduction of a name by ostension, accompanied by a perceptual referential intention, need not always be reducible to cases of introduction via a referential intention based on a description, untrue or not; there may simply be cases where a speaker successfully fixes the reference of a name by pointing to an object and forming the referential intention of referring to that-a thing that she is perceiving, but that she is not representing to herself via any description. As for the second obstacle, my own diagnosis of the hypothetical “Phosphorus” example is that, as is often the case, there are several referential intentions involved, including one formed in terms of a (false) description and one perceptually or demonstratively based, and that there is a convention in operation that has the effect that perceptually based intentions generally override non-perceptually based intentions at the introduction stage, with the result that the reference of “Phosphorus” is successfully fixed as Venus because Venus is the target of the perceptually based referential intention (see Gómez-Torrente (2019), 93-4, n.16).

Perhaps Fernández Moreno would object to this diagnosis with the help of what I think is his claim, that perceptually based referential intentions must include a categorization of the target object in terms of some sortal concept. He says, for example, that

in the introduction of a proper name for an object we only perceive a space-temporal fragment of it, but in order that the name be introduced for the object qua whole object it is required that the introducer of the name conceptualize (implicitly or explicitly) the object by means of some sortal term, in such a way that if the introducer were very wrong about the application of the term to the object, the name would lack reference.

In my view, such claims depend on dubious theories of perception. I don’t see compelling reasons to think that I only perceive a spatio-temporal fragment of Venus when I see it, and not the object Venus itself, independently of whether I categorize it appropriately (I may certainly wrongly categorize it as a star, for example). I also doubt that my cat does not perceive me, as opposed to a spatio-temporal fragment of me, despite her more than likely lack of concepts. Perceiving some thing o is probably just a matter of my brain or mind being in an appropriate causal/dispositional relation to o that need not involve concepts.

Fernández Moreno’s last remarks concern the ordinary speakers’ knowledge of referential conventions and the theoretician’s knowledge about the ordinary speakers’ knowledge, matters on which he wishes I could be more precise than I am in the book. This is a justified wish, and unfortunately I’m afraid I don’t have much more to offer on this topic for the moment. (But see my reply to Orlando in this issue for a bit more on the topic.) My guiding picture in the book is a pretty standard one of reference-fixing conventions as constituted, at least in part, by the existence of widespread agreements in how individual speakers’ and communities’ referential intentions give rise to referents for expressions or expression uses. These agreements I see as manifested in concordances and regularities in speakers’ linguistic behaviors, including reflective linguistic judgments by ordinary and, especially, by linguistically trained speakers, judgments which I see also typically as constitutive of their implicit knowledge of referential conventions. (Fernández Moreno asks specifically “whether Gómez-Torrente would consider the possibility of resorting to the procedures of experimental semantics to back up” claims of this sort; and I certainly would, at least if some precautions are taken, of the kind alluded to in Gómez-Torrente (2019), 171-2, esp. 172, n. 23.). Philosophical and linguistic theoreticians like myself, on the basis of the judgments of this kind that they are able to gather, develop hypotheses as to approximate forms for the relevant conventions, which may involve non-ordinary concepts. Such hypotheses will or will not constitute instances of knowledge or of partial knowledge, depending on how they fare against new data on linguistic behaviors, including again linguistic judgments by competent speakers. A critical discussion of this picture will have to be left for another occasion.


Fernández Moreno, L. (2020), “Reference Theories and the Relevance of Descriptions”, Manuscrito, this issue. [ Links ]

Gómez-Torrente, M. (2019), Roads to Reference. An Essay on Reference Fixing in Natural Language, Oxford University Press, Oxford. [ Links ]

Kripke, S. A. (1972), “Naming and Necessity”, in D. Davidson and G. Harman (eds.), Semantics of Natural Language, Reidel, Dordrecht, 1972, 253-355, 763-9. Book edition with an added preface: Naming and Necessity, Harvard University Press, Cambridge (Mass.), 1980. References to the book edition. [ Links ]

Kripke, S. A. (1986), “A Problem in the Theory of Reference: the Linguistic Division of Labor and the Social Character of Naming”, in V. Cauchy (ed.), Philosophie et Culture (Actes du XVIIe Congrès Mondial de Philosophie), vol, 1, Éditions du Beffroi - Éditions Montmorency, Montreal, 241-247. [ Links ]

Putnam, H. (1973), “Explanation and Reference”, in G. Pearce and P. Maynard (eds.), Conceptual Change, Reidel, Dordrecht, 1973, 199-221. Reprinted in Putnam, Philosophical Papers, Volume 2: Mind, Language and Reality, Cambridge University Press, Cambridge, 1975, 196-214. References to the reprint. [ Links ]

1Support from the research project no. PIDPID-107667GB-I00 of the Spanish Ministry of Science and Innovation is gratefully acknowledged.


Here is my statement of that convention, for the reader’s convenience:

Successful natural kind noun transmission. If a natural kind noun N as used by a community of speakers refers to a certain substance (/species/phenomenon) A and a speaker of the community, S, clearly perceives some use of N by another speaker of the community and understands that N is used as a noun for a natural kind, then N as used by S will refer to A, if S forms the (explicit or implicit) intention of using N as is used by the community of users of N and forms no intention conflicting with this intention, and if S doesn’t form at the transmission stage a set of beliefs involving N that on the whole conflict with the beliefs of the community of users of N at large” (Gómez-Torrente (2019), 162, my emphasis).

Article info CDD: 401

Received: October 01, 2020; Accepted: October 17, 2020

Creative Commons License This is an open-access article distributed under the terms of the Creative Commons Attribution License