This is the synopsis of the report of the symposium held in 1968 at the 40th national convention of the English Literary Society of Japan. The organization of the symposium and hence of this synopsis is as follows : 1. Tow papers read by two speakers, Kajita and Ikegami respectively 2. Prepared comments given by three discussants, Hasegawa, Inoue, and Yasui 3. Speakers' answers to the comments 1. PAPERS 1.1. Masaru Kajita, "The mechanism of embedding" The refinement of syntactic studies often leads to clarifying or confirming our vague, undefined intuition about meaning. As an example, let me cite the problem of sentence embedding. Chomsky says that what is recursively introduced in the base component is S(entence) only, but I assume that there are four sentence-like elements (call them S_1, S_2, S_3, S_4) which can be recursively introduced. They are introduced by the following rules: (1) i) S →(SAdv_1) S_2 ii) S_2 →(SAdV_2) S_3 iii) S_3 →Tense⌒S_4 SAdv_1 are such sentence-modifying adverbs as frankly, briefly; incidentally, however; etc. SAdv_2 are probably, certainly; fortunately, wisely; reportedly; therefore; etc. Tense is expanded by the following rules: (2) i) Tense→Tns (Tense-Adverbial) ii) Tns→{Past Present Future}(Perfect) SAdv_1, SAdv_2, and Tense are in hierarchical relation with each other in respect to the internal restriction imposed upon embedded sentences; that is to say, in an embedded sentence where SAdv_1 can occur, the other two can occur, but not conversely; in an embedded sentence where SAdv_2 can occur, Tense can also occur, but not conversely. See, for example, sentences listed in (4), (6) (p. 241), and (13) (p. 243). Accordingly, the following revision will be necessary in Chomsky's base component: (3) i) VP-V ({S_1 S_3 S_4}) ii) NP-(Det)N ({S_2 S_3 S_4}) As a result of this revision, such verbs as believe, think, etc. will be given a syntactic feature [+_ it S_2], request, demand, etc. [+_ it S_3], and begin, continue, cease, etc. [+it S_4_]. This classification of verbs based purely on syntactic evidence meshes well with our intuitive semantic classification of them. Also different statuses assigned to SAdv_1, SAdv_2, and Tense reinforce our vague intuition about the semantics of these constituents. 1.2. Yoshihiko Ikegami, "Syntax and semantics" I assume i) that 'semantics' defines the combinability of elements, ii) that 'syntax' defines the linear order of the elements, and iii) that there is some correlation between semantic elements and syntactic elements. Essentially in line with the stratificational grammar, i) will be called 'semotactics,' ii) 'lexotactics,' and iii) 'semo-lexemic realizational structure.' There is no one-one correspondence between the sememic and the lexemic strata; hence the phenomena explained in stratificational terms as 'diversification,' 'neutralization,' 'zero realization,' 'portmanteau realization,' The diagram given on p. 256 shows the (a?) structure in the sememic stratum that will be realized as 'sentence' in the lexemic stratum. For a sample analysis, see illustrative sentences (1) to (8) (pp. 254-255). Notice that there is a hierarchical relation between the constituents in the diagram; S^<verb> is combined first with Aspect, and then Agent is combined with the resultant constituent, Cause being combined with the constituent resulting from it, etc. There are also selectional restriction between the types of elements. For example, if 'certain' is selected in Mode, there is no restriction in Tense (Both 'The door must be open' and 'The door must have been open' are all right), but if 'necessary' is selected in Mode, the selection of Tense is limited to 'present,' 'future,' or 'zero' ('The door had to be
(View PDF for the rest of the abstract.)
抄録全体を表示