The cognates of the Gothic masculine i-stems

Since a large majority of the Indo-European i-stems consisted of action nouns (Brugmann 1906: 167-70), it thus appeared that at the time of the loss of the second element of long diphthongs the animate diphthongali-stems were perhaps more numerous than the corresponding zerograde forms or that the diphthongal abstracts had indeed remained productive until the occurrence of the phonological change. Since the evidence of the ti-forms supported the view that all of the inherited feminine i-stems were diphthongal in origin, the Gothic evidence examined thus far therefore indicates that at the time of the loss of the second element of long diphthongs the diphthongal /-stems were considerably more numerous than those of zero grade or that the diphthongal i-stem abstracts had indeed been productive on an appreciable scale in Proto-Indo-European. Since the o-stems were also associated with verbal abstracts (Brugmann 1906: 631) and newly derived o-stems of that and every other sort were being produced in large numbers at that time as part of the expansion which made them the most important Indo-European nominal type, it is plausible that the use of both of the two alternative methods available for the derivation of action nouns would sometimes have resulted in pairs of abstracts derived from the same verb. According to the original proposals concerning the history of the diphthongal t-stems prior to the loss of the second element of long diphthongs, the animate e(i)- and o(i)-stems were productive in the derivation of verbal abstracts on an appreciable scale in Proto-Indo-European until their expansion was halted by the phonological change.