It is impossible here to discuss the question how far this habit of spelling may have been introduced by analogy from French words with a similar sound of the vowel, and how far it may have grown out the Kentish use of ie, ye for O. E. ēo, ē, īe. Reference may be made to the remarks in the Introduction to the volume of Gower’s French Works, p. xxi, where it is suggested that ie, having lost its value as a diphthong in later Anglo-Norman, came to be regarded as a traditional symbol in many cases for close ē, and hence such forms as clier, clief, pier, prophiete, &c., and as regards ie in the Kentish dialect there is a useful statement in the paper by W. Heuser, Zum Kentischen Dialekt im Mittelenglischen, published in Anglia, xvii, 78 ff.
In any case the fact is pretty clear that this variation was confined by Gower to words in which he gave to the vowel a close sound, and it is therefore useful as a distinguishing note, though there are few words in which this is the only form of spelling.
Both in stems of words and in their terminations i is on the whole preferred to y, so that we have crie, hide, lif, like, mile, ride, &c. more usually than crye, hyde, &c. (but perhaps y more often after m, n, as knyht, myhte, nyht), and also arrai, mai, dai, hardi, ladi, worþi, mi, thi, more often on the whole than array, may, &c., but -ly in adverbs more often than -li.
In some few cases it seems that a distinction is pretty consistently made, as between wryte (inf.) and write (past participle), and perhaps between wite (know) and wyte (blame).
Before gh followed by t we find a, o almost regularly in place of au, ou. Thus we have aghte, straghte, taghte, boghte, broghte, doghter, noght, oghte, oght, soghte, wroghte, &c., but occasionally broughte, doughter, ought, &c. Beside some of these there are forms in which au (aw), ou (ow) are written, but followed by simple h, as strawhte, tawhte, douhter (dowhter).
There is no difference between -oun and -on as terminations of such French words as divisioun, complexioun, &c., but -oun is much the more usual form[P]. Where they occur in rhyme, the rhyme-words are usually assimilated to one another in form of spelling, but sometimes -oun, -on rhyme together, as division: doun, ii. 1743 f., toun: condicion, v. 2551, constellacioun: relacion, vi. 2253 f.
In the case of an followed by a consonant in a tone-syllable the variation to aun seems to be merely a question of spelling, and we have such rhymes as chaunce: remembrance, ii. 893 f., demande: comaunde, iv. 2794, supplanted: enchaunted, ii. 2491, covenant: supplaunt, ii. 2367. In the French terminations -ance, -ant, the simple form is decidedly preferred (but governaunce: porveaunce, Prol. 187 f., graunt: amblaunt, ii. 1505 f.), and so also in many other words, as change, strange, comande, demande, supplante (also comaunde, supplaunte). In other cases au is either the usual or the only form, as daunce, daunte, enchaunte, haunte, sclaundre.
With regard to the consonants, it should be observed that Gower consistently wrote sch for sh initially, so that we have regularly schal, schape, sche, schewe, schip, schrifte, and also lordschipe, worschipe, &c.[Q], in other places usually ssh, as bisshop (also bisschop), buissh, fissh, fleissh (also fleisch), freissh, reisshe, wisshe.
The almost regular use of h for gh in such words as hih, nyh, sih, kniht, liht, miht, niht, heihte, sleihte, &c. will be spoken of later.
Gower did not use ȝ for h or gh. Such forms as miȝte, riȝt, uȝte, wrouȝt, are practically unknown in the best MSS. (F has nouȝt once.) On the other hand initially in such words as ȝe, ȝer, ȝive (forȝive), ȝong, &c., ȝ is regularly used. Only late and inferior MSS. have y. In regard to this letter Gower’s usage is exactly the reverse of that which we find in the Ayenbite of Inwyt. We have þ for th regularly except in the case of a capital letter being required, as at the beginning of a line, or in connexion with some foreign words and names as thalemans, thevangile, rethorique, Athemas, Anthenor, Thebith. Cases of th for þ in ordinary English words are very rare in F (but i. 2890, v. 2319, vii. 4203).