The aim of this paper is to describe phrasal verbs in eighteenth and nineteenth century English on the basis of the ample data I have collected from my reading. Special attention has been paid to the variety of genre, such as in essays, fiction, or in letters and drama. The following particles: away, back, down, forth, off, out and up have been examined in collocation with verbs, and their semantic properties have been discussed in detail in relation to Late Modern English.

View full abstract

In the literature, it has often been argued that the change of the word order from Object-Verb to Verb-Object occurred around 1200 in English. But in the case of object pronouns, the former order still survived after that time and disappeared during the ENE period. The present paper is an attempt to study this history under Chomsky's (1993)Minimalist Approach.
I first show that ME object pronouns in the OV pattern were clitics. I claim that they overtly moved as DP to [Spec, AGR_oP] to have their Case feature checked against the Case feature of a verb adjoined to AGR_o, and then moved as D^o and left-adjoined to the complex VAGR_o-T-AGR_s. Assuming that their Case feature was strong, I argue that the Case-checking was possible because verbs overtly moved to AGR_o in ME. Then I claim that the loss of overt V-raising in ENE made such Case-checking impossible, and thus the order 'object pronoun+V' disappeared.

View full abstract

The English verb see means 'see something happening' in I saw John kick Mary, while it denotes an inference from direct observation in I saw that John was honest. Dik (1989) and Hengeveld (1989) argue that there are two complement types: predication and proposition. These different types can be seen in perceptional constructions. In this paper I show that the two complement types influence the meanings of verbs of perception like see so that these verbs have two different meanings.
Not all the verbs of perception have those two meanings: perceptional meaning and cognitive meaning. Verbs like watch cannot have cognitive meaning. I also show that there are some constraints on the semantic development of verbs of perception in English which create distinction between see type verbs and watch type verbs.

View full abstract

This paper deals with the historical development of two auxiliaries, modals and the aspectual auxiliary have. In Present-day English, their syntactic behaviors are different: (1) Modals take bare infinitives as their complements, whereas havee takes a past participle. (2) Modals must precede have, but have cannot precede modals. In Old and Middle English, they both could take NP complements just as main verbs did.
This paper historically examines why these differences emerged. In the first place, modals and have were grammaticalized as auxiliaries in a certain period, but the processes of their grammaticalization were not the same. Secondly, as for the assignment of θ-roles, which is a major factor that distinguishes verbs from auxiliaries, modals and have cannot θ-mark as a result of their grammaticalization. Thirdly, however, have still retains the ability to assign Case to INFL in their complements, though modals cannot assign Case. Fourthly, modals and have were classified as functional and lexical auxiliaries, respectively.
Furthermore, I will show the validity of our analysis by applying it to German. German modals and aspectual haben ‘have’ are lexical auxiliaries and both can assign Case to INFL. As a result, both word orders “modal+haben” and “haben+modal” are possible in German.

View full abstract

It is well known that that in object clauses can be dropped in Present-day English. But the phenomenon was not seen in OE and the early period of ME. In addition, it does not happen in modern languages like German and Dutch. I will point out that the languages that allow C⁰ to be empty allow I⁰ to be empty, and that languages that do not allow C⁰ to be empty do not allow I⁰ to be empty. This interaction can be explained if we adopt the idea of the Extended Projection (Grimshaw (1991)). This is because the two heads share the categorial feature [+V, -N], and that both of them can be the extended projections of V. Then, I will propose that the ellipsis of that does not depend upon the nature of C⁰ but the categorial feature of V. A diachronic study shows that this seems to be the case. In the earlier stage, English required the two heads to be lexicalized. As the result of the loss of verbal inflections in EModE, the categorial feature [+V] was weakened. Therefore, lexicalization of the heads became unnecessary.

View full abstract