*-ōzô

From a Proto-Germanic innovation built on the model of the inherited comparative suffix *-izô, with the corresponding superlative *-ōstaz formed secondarily after *-ōzô in the same way that *-istaz was formed from *-izô. The inherited model ultimately continues Proto-Indo-European *-yos- ~ *-is-, the primary comparative/elative suffix. In PGmc, the zero-grade *-is- was generalized in the comparative; since the pre-Verner accent fell on the root, the suffixal *s became *z by Verner’s Law, yielding *-izô. (Fulk 2018: 218–20; Meier-Brügger 2003: 219–22; Ringe 2017: 316–17).

The long vowel of *-ōzô is specifically Germanic. Fulk (2018: 218) follows the traditional explanation that *-ōz- originated when comparative *-iz was added to adverbs in *-ō, after which the resulting adverbial comparative was extended to adjectives. Ringe (2017: 317) is more cautious, saying that the exact preform is not recoverable, but the suffix complex appears to have arisen by contraction with a preceding vowel and then spread by morphological remodeling. As a result, *-ōzô has no exact non-Germanic cognate (Fulk 2018: 218–19; Meier-Brügger 2003: 220–22).

In PGmc, *-izô was used with i-, u-, and ja-stem adjectives and also with some a-stems, whereas *-ōzô was restricted to a/ō-stems. The distribution among a/ō-stem adjectives was partly lexical and analogical rather than fully predictable. In the Northwest Germanic branches, *-ōzô/*-ōstaz became the normal productive pattern, while *-izô/*-istaz often survived only in relic forms (Fulk 2018: 218–20; Ringe 2017: 317).