Abstract
In this chapter I shall propose a form of descriptivism, Contextual descriptivism (CD, in brief). As explained in the introduction, CD relies on “contextualized properties ” of the form F@t, where “@” is the contextualization sign , whose meaning is elucidated in §§ 5.3-5.4. Very roughly speaking, and given the default option that we shall pursue (according to which t is a linguistic token), CD can be seen as a sort of generalization to proper names and incomplete determiner phrases (including definite descriptions) of Reichenbach ’s (1947) token-reflexive approach to indexicals. According to CD, indexicals and proper names are (in typical cases) incomplete (truncated) definite descriptions, and thus the descriptive contents which constitute their pragmatic meanings should be characterized by relying first on an auxiliary account of the pragmatic meaning of incomplete definite descriptions. Since, as we have seen, descriptions can be incomplete just like other members of the more general class of determiner phrases to which they belong, it follows that this auxiliary account should be extracted from a more sweeping theory about the pragmatic meaning of incomplete determiner phrases. The next few sections will then be devoted to this preliminary issue in a way that involves, for generality’s sake, a discussion of complete determiner phrases as well. After this, I shall try to characterize the pragmatic meaning of various kinds of singular terms from the point of view of CD.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
This worry was raised in correspondence by Garcia-Carpintero .
- 2.
As far as we are concerned here, the relation PRAG is a primitive notion, just as the meaning relation that links an expression to its semantic meaning(s). Although no complete analysis of PRAG is provided, it is implicitly, albeit partially, characterized by various principles, such as P1-P3, listed below. The fact that PRAG is taken to be primitive here does not rule out that it can be analyzed at a different deeper level, say, from the point of view of a naturalization project (although I am sceptical about it). The same goes of course for the meaning relation. (Thanks to Aldo Frigerio for prompting me to insert this note as a reply to one of his queries.)
- 3.
More precisely, we may then want to say that PRAG is a four-term relation, involving as an additional “grammatical” parameter G: PRAG(t, G, M, C) (similarly, we might want to introduce this parameter for the relation COMPL, to be discussed below). The idea is that G is a grammatical category that t somehow contributes to “realize” to the extent that it contributes to convey the contextualized meaning M. For instance, as regards the example based on (2), G would be the predicate category and as regards the example based on (1), it would be the determiner phrase category. However, for present purposes, we can neglect this further parameter.
- 4.
In case (c), the grammatical parameter discussed in the previous note would be the grammatical category tense. Specifying this additional parameter may be useful if we think that a statement as a whole may have a conversational impliciture other than the one it has in relation to its tense component.
- 5.
If we are to take into account the grammatical parameter discussed in note 3 above we should say something like “relevant with respect to t and G”, where G is the grammatical category in question.
- 6.
The assumption that a new predicate “M@t” has become part of the language can be seen as presupposing that t, understood as if it were a singular term directly referring to itself, has in turn become part of the language. By taking advantage of the lambda operator discussed in note 25 of § 1.7, a predicate of the form “M@t(x)” can be understood as short for “[λy @(M, t, y)]”. By the principle of lambda conversion , “[λy @(M, t, y)]” is equivalent to “@(M, t, x)”. This makes it clear that @ is a relation, more specifically a triadic relation that can be true of a property, M, a linguistic token, t, and an item of whatever nature, x. If we take into account the grammatical parameter discussed in note 3 above, however, we should rather speak of properties expressed by predicates of the form “[λy @(M, t, G, y)]” (or, in shorter form, “M@<t, G>”), thus viewing @ as a tetradic relation. Here “G” stands for a certain grammatical category, intuitively a category to which t, qua linguistic item, belongs. By the principle of lambda conversion, M@<t, G>(x) would be equivalent to “@(M, t, G, x)”. Moreover, instead of P4 above, we would need: M@<t, G>(x) ↔ ∃C(PRAG(t, G, M, C) & C(x)).
- 7.
Since I focus on English, I ignore for simplicity’s sake the fact that in some languages, e.g. ancient Greek, proper names and indexicals may be used with a definite article and thus should be regarded, at least in a descriptivist perspective, as integral (untruncated) descriptions at the level of contextualized linguistic meaning.
- 8.
This example was formulated when Romano Prodi was the prime minister of Italy. Other examples involving Prodi dating to that period will occur in the following.
- 9.
I regard an expression type that counts as a proper name predicate (as it expresses a nominal property) as possibly ambiguous in that it could express other properties as well. For example, “Miller” could express a nominal property but also the property of being a miller (more on this below). Alternatively, rather that saying that there is ambiguity we may say that there are distinct expression types. For example, there is a type “Miller” expressing a nominal property and a type “miller” expressing the property of being a miller. For the main purposes of this book nothing crucial hinges on this. As against the view that proper names are common nouns expressing properties that can be shared by different objects, Leonardi 1990 pointed out that one can always append ever new specifications (compare “Frederick I” and “Frederick II”) to a given proper name in an attempt to force it to have just one object in its extension. But such embellishments of a given proper name are best regarded as new proper names and, of course, there can be no absolute guarantee that these new proper names have a singleton as their extension. That is, they should also be viewed as common nouns (cf. Castañeda 1990a).
- 10.
There is a city in South-Tyrol, a German speaking part of northern Italy, called “Vipiteno” in Italian and “Sterzing” in German. Aldo Frigerio has pointed out in correspondence that there may be a difficutly for my view of nominal properties, since the Italian sentence “Vipiteno è in Italia” (Vipiteno is in Italy) and the German sentence “Stierzing ist in Italien” (Stierzing is in Italy) express precisely the same proposition (for one is the translation of the other) even though “Vipiteno” and “Stierzing” are not in a relation of transliteration (cf. Frigerio 2006). I object however to the idea that the two sentences express the same proposition. The practice of translation may for good reasons fail to preserve the proposition expressed in the original sentence, for the primary goal is to maximize mutual understanding and this may sometimes be in conflict with preserving the proposition. Sometime it may be better to choose an extensonally equivalent but different proposition. This is a case in point. Two different but extensionally equivalent propositions are chosen, because, we may assume, the same city was baptized independently and with two different names by the Italian-speaking and the German-speaking communities. According to Frigerio, another difficulty may be due to the fact that the view in question seems to imply that a speaker who hears a certain name N for the first time, while being told simply that precisely one individual bears this name, is just as competent about N as the speaker is competent about the definite description “the unique individual called N”. According to Frigerio, this is counterintuitive. But I think that it may seem counterintuitive only if we confuse competence about a name and knowing who (or which individual) the bearer of the name is (see § 7.6).
- 11.
It should be clear that, by accepting PNBN, my approach to proper names embeds elements of the so-called metalinguistic theory of proper names, proposed in Kneale 1972 (p. 630), fallen in disgrace after being criticized by Kripke (1980, pp. 68–70) and recently revived in various guises by different authors (see, e.g., Loar 1976, Katz 1979, 1990, Recanati 1993, Garcia-Carpintero 1996 and Bach 2002).
- 12.
See Kroon 1987 and the references in footnote 1 of that paper, where Lewis 1984 is taken to be the work where the term “causal descriptivism” is first used. Kroon there counts Loar 1976 as defending causal descriptivism , but it is not clear to me that this is right. It may also be worth noting that Lewis had in mind something like causal descriptivism at least since 1968; see his 1972, p. 215. See also Castañeda 1977, note 5 and Fumerton 1989.
- 13.
Evans 1973 also claims that a threat to the causal theory is posed by “deferential” uses of proper names in which one uses names “with the overriding intention to conform to the use made by them by some other person or persons” (p. 21; see his “Louis” example, at pp. 6 and 21). But it seems to me that, even if a subject has such an intention in using the token n of a proper name, this should not undermine the fact that n refers to the source of the nominal-causal chain that leads to n.
- 14.
Similarly, we can perhaps regard in this fashion the hypothetical story by Evans in which the name of a scribe, “Ibn Khan”, is associated by the mathematical community to the mathematician who proved certain theorems. The idea is that the member of the mathematical community who started using a token of “Ibn Khan” to refer to whoever proved the theorems realized a new baptism of the latter by means of a reference-fixing description of the kind “the mathematician who proved such and such”.
- 15.
The point made here has some analogy to Devitt ’s claim (1981) that a successful act of naming requires that the namer correctly conceptualizes the baptized item as belonging to a certain category (expressed by a predicate).
- 16.
For example, PNCD could be circumscribed to “inheritance contexts” as follows:
-
PNCDR. Proper Names: Causal Descriptivist PRAG principle, Restricted version. Suppose that n is a proper name token, with the nominal property |N| as linguistic meaning, such that (i) |the N| is the contextualized linguistic meaning of n; (ii) the contextual complex for n is primary; and (iii) n occurs in an inheritance context. Then, the following holds: PRAG(n, |N|, |source of the nominal-causal chain leading to n|).
An inheritance context typically involves names of historical figure or places or notable individuals with whom the speaker is not personally acquainted and with respect to whom he has very impoverished knowledge. A context in which people are discussing philosophy and the name “Thales ” is used provides an obvious example.
-
- 17.
Perhaps a sentence such as “John bought a donkey and a horse and Tom vaccinated the donkey” illustrates better than (1) the option of using a standard determiner phrase like “the donkey” as an anaphoric term that does essentially the same job of a pronoun (e.g., “it”). However, for simplicity’s sake, I shall deal with (1).
- 18.
Let us assume, for our illustrative purposes, that “Coralville Farm” is a predicate corresponding to a property that uniquely identifies a certain Farm, namely the one where the John and the Tom of our example work.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2010 Springer Science+Business Media B.V.
About this chapter
Cite this chapter
Orilia, F. (2010). Definite Descriptions and Proper Names. In: Singular Reference: A Descriptivist Perspective. Philosophical Studies Series, vol 113. Springer, Dordrecht. https://doi.org/10.1007/978-90-481-3312-3_5
Download citation
DOI: https://doi.org/10.1007/978-90-481-3312-3_5
Published:
Publisher Name: Springer, Dordrecht
Print ISBN: 978-90-481-3311-6
Online ISBN: 978-90-481-3312-3
eBook Packages: Humanities, Social Sciences and LawPhilosophy and Religion (R0)