A DIACHRONIC SURVEY OF PERIPHRASTIC 'Go'

Abstract

In the history of the English language there is a tendency that most fundamental verbs that express motion, change, process, or static condition have weakened their original sense and turned into functional form-words as mere components of periphrastic verb-groups. One of the most striking verbs that have undergone this process is 'go.' Here I have observed various periphrastic or pleonastic usages with the verb that etymologically or semantically corresponds to ModE go, and how they have maintained diachronic currents through the periods of Old and Middle English. I have also noticed that the process is suggestive of some features in the development of auxiliaries in general. As to the phenomena of Old English, I shall first deal with uton gan (W.-S. Gosp., John xiv. 31). (W)uton in this verb-group was originally a subjunctive form in the plural first person, the corresponding infinitive being (ge)witan (to go). So the literary meaning of (W)uton gan was 'let us go to go.' Here (W)uton had been reduced to a genuine auxiliary for a periphrastic exhortative idiom. A more conspicuously periphrastic instance is offered by gewat faran (Beowulf 124). The simple infinitive faran (to go), which was originally adverbial and denoted purpose, has appeared as the chief element of predication, while the preceding finite form gewat merely functions as a kind of auxiliary which formally denotes tense and aspect. Such function of the uninflected infinitive resembles that of a present participle as it accompanies a verb of motion, hence the parallel use of both forms as in "com yrnan... com yrnende" (Eefric, Lives of Saints XXXI. 1038-43). Later, however, the latter comes to supersede the former as a more distinct means of expression in this syntactic position. Besides such subordinate constructions, there is the co-ordinate construction gap and leornigeap (W.-S. Gosp,. Matt. ix. 13), where gap and has lost its organic sense and turned into a formal device to invigorate the imperative tone in virtue of the ingressive force inherent in gap. In the Middle English period, up to the end of the thirteenth century, E (W)uton gan was inherited in the form of ute we fare (The Owl and Nightingale 1780), where the plural second person we was usually expressed as subject. But in and after the fourteenth century, the form was superseded by the more explicit and logical construction let us go. As a ME phenomenon corresponding to OE gewat faran, we may mention go roule aboute (Chaucer, C. T., D 653), which means something like 'gad about.' Here go, through structurally a main verb, is merely used to form a periphrastic construction, so as just to add some humorous or ironic touch to the expression. The periphrastic 'go' is also accompanied by a present participle. In such an instance as gooth ymagining (Chaucer, C.T., E 598), gooth has lost the original sense of motion, and the whole group means something like 'begins to imagine.' When 'go' is combined with a past participle, it naturally comes to act as a copula and hardly means anything more than 'become' or 'be,' as illustrated by "this was hire manere, To gon ytressed with hire heres clere" (Chaucer, Tr. & Cr. v. 809-810). The extreme extention of this periphrastic usage is observed in was go walked (Chaucer, C.T., D 1778). In this peculiar phenomenon, walked can be interpreted, as by Skeat, as a corrupted form of a-walketh, where -eth is a nominal suffix derived from OE -op, -ap. But the more justifiable explanation is that walked is a past participle, demoting a resultant state, and the intermediate go, though it may be morphologically a past participle weakened from goon, has here been inserted before the virtual past participle walked, so that go-walked can be considered a past participle of the compound go-walk. On the other hand,

(View PDF for the rest of the abstract.)