Barbara Abbott

Last year (2005) marked the 100th anniversary of the publication of Russell’s classic ‘On denoting’. It should not cast any shadow on that great work to note that the problems it provided solutions to are still the subject of controversy. Two of those problems involved noun phrases (NPs) which fail to denote. Russell’s examples (1a) and (1b) (1) a. The king of France is bald. b. The king of France is not bald. are puzzling because they have the form of simple contradictories, and yet we are not inclined to say either one is true. Example (2) (2) Pegasus does not exist. is even more problematic; the lack of denotation for Pegasus, which makes the sentence true, also seems to rob it of a meaningful constituent. Once the king of France is unpacked according to Russell’s analysis, (1b) is revealed to be ambiguous. It’s logical forms are given in (3). (3) a. ∃x[Kx ∧ ∀y[Ky ↔ y=x] ∧ ¬Bx] b. ¬∃x[Kx ∧ ∀y[Ky ↔ y=x] ∧ Bx] (3a) says that there is a unique (French) king who is not bald (obviously false), but (3b), the logical contradictory of (1a) says that it is not the case that there is a unique king who is bald (which is true). We can apply the analysis to sentence (2) once we recognize Pegasus as a concealed definite description, e.g. the winged horse of Greek mythology. (2) can then be unpacked as (4) (4) ¬∃x[Wx ∧ ∀y[Wy ↔ y=x]] which seems both meaningful and true, as required. Problems solved. Well, not quite. Strawson (1950) challenged the first solution above, arguing that neither (1a) nor (1b) could be used to assert the existence of a king of France. Rather, use of such sentences presupposes the existence of a king of France, and failing that existence, neither of (1a) or (1b) could be used to make either a true or a false statement – in Strawson’s words, “the question of whether it’s true or false simply doesn’t arise” (Strawson 1950, 330).1 In an extended series of essays, and one book, Jay Atlas (1977, 1978, 1979, 1989, 2004) has taken issue with the work of both Russell and Strawson..

This would have been a better book if Sampson had argued his main point, the usefulness of the Simonian principle as an explanation of the evolution, structure, and acquisition of language, on its own merits, instead of making it subsidiary to his attack on ‘limited-minders’ (e.g., Noam Chomsky). The energy he has spent on the attack he might then have been willing and able to employ in developing his argument at reasonable length and detail. He might then have found that argument faulty, or even wrong. Underdeveloped as it is, however, his point about the Simonian principle does seem to have thepossibility of some significance in understanding the empirical issues of this book (and the dichotomy of ‘limited’ and ‘creative’ minds is certainly not one of these).What Sampson has done inMaking Sense has the appearance of science, and theoretical linguistics, and therefore it is somewhat puzzling to find the conclusions that the human mind cannot be scientifically studied (p. 10), and that theoretical linguistics is a “mirage” (p. 210). His statement that It may still seem worthwhile to some to work out the exact implications for present-day linguistic structure of the fact that languages are evolved systems, but I cannot see that there is much glamour or, indeed, much substance in this task (p. 210) would surely be surprising to a geneticist or ethologist. We hope his apparent misunderstanding on this point will not lead him to give up his profession, in which he seems willing and able to make a significant contribution.What in particular is required is a clear indication of how the Simonian principle imposes itself in the ontogenesis of language. Consider Chomsky's example of the rule for forming questions in English. For sentence (1) below, children able to deal with such examples would give (2) but not (3) as the corresponding question (Chomsky, 1975, p. 31).The man who is tall is in the room.Is the man who is tall in the room?*Is the man who tall is in the room?Now compare Sampson's evolutionary explanation for the structure-dependence of rules which is illustrated in (1)-(3): “once a behavior-pattern has become thoroughly established, it will tend to be treated as a fixed given when used as a constituent of a more complex, later-learned behavior-pattern” (p. 183). In order for this to account for Chomsky's example it would have to be the case that children learn relative clauses before they learn questions, which simply isn't true.In conclusion, we wish not to seem unaware of the problems with the position Sampson is attacking, or unaware of the sometimes dogmatic way it too has been presented. Chomsky's hypothesis of innate linguistic ideas seems, fundamentally, to derive from the claimed rapidity and uniformity of language acquisition in the face of limited and degenerate data. Chomsky appears to have been unwilling to argue these points directly or to show their relation to others, for which he has argued, for example, the implausibility of some abstract constraint's having been learned (cf. 1971, pp. 22–44; 1975, pp. 30–33; 1980, pp. 35–52). And almost from his first invoking of rationalism in support of his linguistic method, Chomsky has himself resorted to an unnecessarily polemical attack on empiricism (cf. 1965, p. 58, n. 33). The innateness issue is just not clear or simple enough to justify so much animus, or even conviction.

Grice introduced generalized conversational implicatures with the following example: "Anyone who uses a sentence of the formX is meeting tz woman this evening would normally implicate that the person to be met was someone other than X’s wife, mother, sister, or perhaps even close platonic friend" (1975 : 37). Concerning this example, he suggested the following account: When someone, by using the form of expression an JQ implicates that the X does not belong to or is not otherwise closely connected with some identifiable person, the implicature is present because the speaker has failed to be specific in a way in which he might have been expected to be specific. (Grice 1975: 38.) Grice went on to sketch an explanation as to why such an implicature should arise, involving the different ways in which we behave towards things that are closely related to..

In the 1960’s, both Montague (e.g. 1970, 222) and Grice (1975, 24) famously declared that natural languages were not so different from the formal languages of logic as people had thought. Montague sought to comprehend the grammars of both within a single theory, and Grice sought to explain away apparent divergences as due to the fact that the former, but not the latter, were used for conversation. But, if we confine our concept of logic to first order predicate logic (or FOPL) with identity (that is, omitting everything which is not required for the pursuit of mathematical truth), then there are of course many other aspects, in addition to its use in conversation, which distinguish natural language from logic. Conventional implicature, information structure (including presupposition), tense and time reference, and the expression of causation and inference are several of these, which combine as well with syntactic complexities which are unnecessary in first order predicate logic. In this paper I will argue that such distinguishing aspects should be more fully exploited to explain the differences between the material conditional of logic and the indicative conditional of one natural language (English).

The phrase "natural kind term" has come into the linguistic and philosophical literature in connection with well-known work of Kripke (1972) and Putnam (1970, 1975a). I use that phrase here in the sense it has acquired from those and subseqnent works on related topics. This is not the transparent sense of the phrase. That is, if I am right in what follows there are words for kinds of things existing in nature which are not natural kind terms in the current sense - e.g. planet, woman, and weed. I use the term "nondescriptional", following Salmon (1981), to mean "lacking a Fregean sense", where a Fregean sense is taken to be a property or group of properties semantically associated with an expression which determines its denotation. I hope that the term "nondescriptional" will be clarified in subsequent discussion, but perhaps a few words about it and its opposite, "descriptional", would be useful here. These terms (as Salmon notes (p.15, n. 11)) should be taken as technical terms. The idea is that descriptional words, of which bachelor is the overwhelmingly favorite example, express a semantic content, or sense, which is what mediates their denotation. Entities fall in their denotation in virtue of possessing the property or properties expressed as their sense. (It is not necessary, however, that this sense be as easily paraphrasable with descriptive synonyms as it is typically held to be in the case of bachelor, or even that it be paraphrasable at all.) In the case of nondescriptional words there is no Fregean sense to mediate the link between word and object. Hence such terms are described by what Salmon, following Kaplan, calls "the theory of direct reference". What it is that does associate such terms with their denotations is another question which will be touched on briefly at the end. I have several aims in this paper. The first is to defend a conservative position on nondescriptionality. I will argue that relatively few natural language words have this property. My other aims are primarily clarificatory. Among the issues I would like to clarify are those connected with the determination of this property and flae explanation of why it exists, its relationship with other properties such as rigid designation and essentiality, and its significance for lexical semantics. The organization of the paper is as follows. The first section identifies the position taken here on the question of which expressions are nondescriptional and provides a general defense of it. The second section is a critical examination of criteria associated with nondescriptionality. In the third section provide further defense for the conservative position, and in the fourth section we turn to related issues in lexical semantics.

Partee (1973) discussed quotation from the perspective of the then relatively new theory of transformational grammar.2 As she pointed out, the phenomenon presents many curious puzzles. In some ways quotes seem quite separate from their surrounding text; they may be in a different dialect, as in her example in (1), (1) ‘I talk better English than the both of youse!’ shouted Charles, thereby convincing me that he didn’t. [Partee (1973):ex. 20] or even in a different language, as in (2): (2) Louise said ‘Je voudrais un auto-da-f´e’, but I didn’t know what she meant.

Donkey pronouns (e.g., it in Every farmer who owns a donkey beats it) are argued to have an interpretation more similar to a demonstrative phrase (e.g., . . . beats that donkey) than to any of the other alternatives generally considered (e.g., . . . the donkey(s) he owns, . . . a donkey he owns). Like the demonstrative phrase, the pronoun is not equivalent to Evans' E-type paraphrase, nor to either the weak or the strong reading sometimes claimed for donkey sentences. A consequence is to narrow the range of formal analyses