This paper deals with the historical change of the syntactic forms of coordinate constructions. In Old English the coordinate constructions of the form (ii) in which conjuncts are split were used, as well as those of the form ( i ): ( i ) X1 & X2 (X=a conjunct; &= a coordinate conjunction) ( ii ) X1... & X2 The form (ii) was becoming recessive in the late Old English period and disappeared almost entirely by early Middle English (Traugott 1972:148). The purpose of this paper is three fold; to offer detailed data on the frequency of the use of these forms, to examine the adequacy of the descriptions about these forms which are given by Reszkiewicz (1966), Traugott (1972), Lightfoot (1979) and Mitchell (1985), and to propose a hypothesis which explains the description of the syntactic forms of coordinate constructions.
The present article aims to clarify the process where the adverbial phrase (just) in case has been formed, through the intermediate stage as a complex preposition/conjunction. It has the meaning ‘apprehension’, which cannot be produced out of the simple summing up of all the lexemes involved in the phrase. It will be shown that the rise of ‘speech-act-related adjunct’or‘style disjunct’ usage has played an important role in the development of the adverbial usage. The result will then be reconsidered from the viewpoint of grammaticalization, with special reference to (a) the rise of pragmatic meaning, (b) syntactic scope increase and (c) the formal gain resuiting from the attachment of just. Some aspects of its development seem to go with the basic principles of grammaticalization, while others might rather pose some questions regarding the recent definition of the notion ‘grammaticalization’ itself. An interrelated understanding of grammaticalization and idiomatization will help place a case like this in the right context of the research field.
This paper attempts to account for the distribution of to-infinitives with lexical subjects in Middle and Modern English in terms of their Case licensing, within the framework of the minimalist program. After identifying two classes of to-infinitives whose subjects are accusative and nominative, respectively, the historical changes in the category and formal features of the infinitive marker to are examined on the basis of to-infinitives as verb complements. Then, it is argued that accusative subjects are licensed by to in to-infinitives as complements to adjectives and nouns, due to its prepositional nature that it retained until the sixteenth century. On the other hand, to-infinitives with nominative subjects, which were first attested in the fourteenth century, are shown to be categories of CP, with C responsible for nominative Case assignment under the system of Case/agreement based on the C-T configuration. Among them are the exclamatory and absolute infinitives, which are argued to have survived until quite recently because of their same sorts of illocutionary force as finite clauses.
From a diachronical point of view, there contact clauses are not derived by way of deleting subject relative pronouns of the relative clause constructions. But, rather, they are obtained, based on the apo koinou principle, by reducing the parataxis structure, i.e., two adjacent sentences, into one sentence. And these there contact clauses constitute part of the derivation of the relative structures. The basic assumption is that the relative structure is derived from the parataxis structure. And our new hypothesis is that the derivation is a two-way process: one way is from the parataxis, via pronominalization and conjunction insertion, to the relative structure; the other is from the parataxis, via there contact clause, to the relative structure. It is shown that this hypothesis accommodates all the data gathered from nursery rhymes. It is furthermore claimed that, in the transition from there contact clause to the relative structure in the above mentioned two-way process, emergence of the relative pronoun is motivated by a principle in Grammatical Dynamism which fills the gap in the syntactic paradigm, a step motivated by the existing process of relativization in the other route of this hypothesis.
Tracing the history of the expression “the + river name,” the present paper argues, among other things, that (a) the type without the or river (e.g., Euphrates) precedes the type with the or river (e.g., the (river of) Euphrates), (b) the type placing River after the river name (e.g., the Mississippi River) can be traced back to 17th century American English, and (c) river names came to take the regularly only in the latter half of the 19th century.
In my Syntax and Style in Early English-Finite and Non-finite Clauses c.900-600 (Tokyo: Kaibunsha,1979) (abbreviated to SSEE below), I have pointed out that both the proportion and number of non-finite clauses approximately equal those of finite clauses in the 15th century corpus (abbreviated to the SSEE corpus below). Only dependent finite clauses are considered, by which I mean clauses which are introduced by subordinators, and make up grammatical sentences only if subordinate to a further clause. Non-finite clauses are verbals which have close functional relations with finite clauses. The corpus there only consists of three prose and two verse texts, and the dates of the texts examined, are not well balanced, since the corpus contains only one text of the last quarter of the 15th century. Hence, I shall try to explore the clauses under consideration more closely for each quarter of the century, by comparing the findings in each quarter century corpus with those in the SSEE corpus. The syntactic description in the research is based on the following four corpora:
Since the influential work of Abney (1987), the nominal projection has been assumed to be headed by D taking NP s as its complement. This paper raises an objection to the claim that the NP/DP distinction in noun phrases corresponds to the predicate/argument distinction (Abney 1987, Stowell 1991, Longobardi 1994). I claim with reference to earlier English and Present-day Japanese that both lexical and functional expressions can denote argumenthood, and that only functionally determined referential expressions are “closed”, thus not allowing a further modification by a determiner.