The purpose of this paper is to analyze the base-generated position of the negative element not in English negation and the movement of verbs across not. The theoretical background of the study is based on Rizzi`s Relativized Minimality and the range of researches is spanned from Old English(OE) to Modern English(ModE). The results obtained from the analysis are summarized as follows: 1. In the surface structure of OE negation, the negative element preceded the finite verb in both main and subordinate clauses. The surface word order of main clauses in OE negation was derived from the movement of the negative element from the specifier position of VP to that of CP and the raising of the finite verb to the verb-second position. The surface word order of subordinate clauses in OE negation manifested the underlying word order itself : the finite verb projected a preverbal negative element as its specifier. 2. Owing to the levelling of morphological inflections, the underlying word order of Middle English(=SIVO) wa reanalyzed from that of OE(=SIVO) through syntactic operations such as verb-raising, extraposition, decliticization, the establishment of the subject position in IP and the loss of the verb-second phenomenon. The grammatical category of the negative element was still analyzed as the specifier of VP. 3. In the early ModE period, do and auxiliaries underwent a diachronic reanalysis of the grammatical function and the base-generated position of the negative element not was reanalyzed as the head position of NEGP. Consequently the raising of a lexical verb to I, the head position of IP or C, the head position of CP comes to be impossible in a ModE negation construction. In that case, an inserted-do or a base-generated auxiliary in the head position of ModP raises to I or C in order to satisfy the morphological conditions in PF and LF. But aspectual be and have still shows the syntactic properties of OE.

• 박상수 | Sang Soo Park