Þa béodes hé beodéþ therínne. Pater Noster, 23.
Annd áȝȝ afftérr þe Góddspell stánnt.Orm. 33.
All þúss iss þátt hallghé goddspéll.ib. 73.
In most cases dissonant rhythmical accentuations of this sort are caused by the rhyme, especially in Middle English poetry, e.g.:
Sównynge alwáy th’ encrés of his wynnýnge.
He wólde the sée were képt for ény thínge.
Chaucer, Prol. 275.
Cf. also: thing: writýng ib. 325–6; bremstóon: non ib. 629–30; ale-stáke: cake ib. 667–8; goddésse: gesse Chaucer, Knightes Tale, 243–4; herde: answérde ib. 265–6; assemblýnge : thynge Barclay, Ship of Fools, p. 20; similar examples are even to be met with in early Modern English poetry, e.g.: nothíng: bring Sur. 15; bemoaníng: king Wyatt, 206; welfáre: snare ib. 92; goodnéss: accéss ib. 209; manére: chere Surrey, 124, &c.
Sometimes it may be doubtful how a line should be scanned. In some cases of this kind the usage of the poet will decide the question; we know, for instance, that Orm never allows the omission of the first unaccented syllable. Where decisive evidence of this kind is wanting, the verse must be scanned in such a manner as to cause the least rhythmical difficulty. If a compound, or a word containing a syllable with secondary accent, does not fit in with the rhythmical accent, it is to be read, as a rule, with level stress when it occurs in the middle of a line (and, of course, always when it is the rhyme-word). On the other hand, if according to the rhythmical scheme of the line an unaccented syllable would be the bearer of the rhythmical stress, we must in most cases assume suppression of the anacrusis.