2022-08-10T13:22:11Zhttps://kitami-it.repo.nii.ac.jp/oaioai:kitami-it.repo.nii.ac.jp:000065312021-03-01T06:50:58ZOn the Logical Properties of universally-qnantified English SentencesHidekatsu, NIIZUMA32955application/pdfSynopsis: This paper investigates the logical properties of any-sentences (in contrast to those for every-sentences) within the framework of a logically-based grammar. (1) a. John spoke to anyone who came. b. John didn’t shoot any bird. Labov (1972) points out that (1a) expresses the meaning of “a lawful generalization.” This semantic characterization of (1a) can be extended to include (1b). Reichenbach (1947) would give (1a) and (1b) the following logical representations. (2) a. Ax ((X came)⊃(John spoke to X)) b. Ax((X is a bird)⊃?(John shot X)). In (2a) and (2b) any is analyzed as the universal quanlifier. In (2b) the apodosis “John didn’t shoot X” is the proposition schema which represents the prepositional part of (1b) and the protasis “X is a bird” defines the category name under which the x in the schema behaves as an argument, so the X stands for a free argument variable. This causes the‘analyzability’ of the X with respect to the schema. The same holds for (2a). The universal quantification results in the same type of conjunctions as shown below. (3) a. John spoke to X_1 and John spoke to X_2 and …∞. b. John didn’t shoot X_1 and John didn’t shoot X_2 and …∞. Assuming that Horn’s (1972) Factoring is a set-formative rule, (1a) and (1b) can be derived by applying the rule to (3a) and (3b). (4) a. John spoke to (X_1 or X_2 or …∞). b. John didn’t shoot (X_1 or X_2 or …∞). The sets so formed are lexicalized with any. Several arguments for the plausibility of this derivation of any are presented in this paper. After a brief survey, in Section 1, of some important contributions in the literature, Section 2 discusses the referential problem for any in contrast to every, clarifying that Factoring, in sharp contrast to Conjunction Reduction, depends for its applicability crucially on the referential nonspecificity condition that must be imposed on the X’s in (3). Section 3 concentrates on the logical properties of sentences like (1b) and on the set-formative function of any. There, under the proposed assumptions, it is naturally explained that the sets in (4) are obligatorily formed by Factoring on the referential condition that the individual X’s can only be included in the sets under the category names given by the protases. This condition will also explain the immunity of Factoring from the Cross-over Constraint in deriving (4) from (3) and, most importantly, that the existential interpretation for (1b), often asserted by linguists, results from the logical entailment that the truth of the proposition with respect to one arbitrarily chosen argument X_1 can vouch for the truth of it with respect to any other argument in the set (4b) (the unmarked condition). The marked condition is necessary for (1a). The stress on any signals reference to the totality of the set (4a). section 4, summarizes the points made in Section 2 and 3. Section 5 presents evidence for the set-formative function of any in contrast with Fauconnier’s (1975) Scale Principle, demonstrating that an analysis of any made otherwise than in quantificational terms will prove invalid.departmental bulletin paper北見工業大学1982-03application/pdf北見工業大学研究報告213149169https://kitami-it.repo.nii.ac.jp/record/6531/files/13-2-6.pdfeng