The following passages from authors later than Aristotle have been cited by grammarians as illustrating the a. u. c. construction; they have not been used in the preparation of this paper and are given here without comment:

In Language 14.303, Professor Urban T. Holmes, Jr., noting that the Dean of Lismore represents Gaelic spirant g by z, queries whether this might represent the pronunciation of the Dean's day or dialect.

No convincing or satisfactory etymology has as yet been suggested for the English word big. Skeat properly compares it with the Norwegian bugge ‘strong man’ and says that it is probably a borrowing from Scandinavian. He further hazards a connection with NE bow, Goth, biugan, from PG *beugan, and there lets the matter rest. Webster's New International Dictionary connects it with bug and Puck, originally ‘hobgoblin, spectre’, and suggests a further relation with Lat. boccae ‘puffed cheeks’. But phonetic difficulties, to say the least, are at once apparent. The word bug is undoubtedly borrowed from the grotesque fancies of Celtic folklore; cf. Welsh bwg ‘hobgoblin’: Lith. búgti, baugùs ‘terrify(ing)’, Skt. bhujati ‘turn aside’, Gk. φ∈ὐγω, Lat. fugio ‘flee’, all from IE *bheug-. If big is from the Norse it must have a Germanic inheritance and could not be related to the IE root *bheug-at all. Puck, < ME pouke < OE pūca : ON pūki, is equally bad, if not worse. Nor is the semantic connection all that could be desired. Middle English (big has no Anglo-Saxon antecedent) big, bigge, bygly mean ‘large, rich, mightily’, and the Scandinavian bugge means, as we have seen above, ‘a strong or important man’. It is a far cry from ‘imp, hobgoblin’ to the conception of largeness, might, and wealth. Der Tag of proletarian ideologies had not yet arrived.

A noun and participle (or adjective) when neither has modifiers may be combined with the article in five different ways: apn, anap,nap; a η ρ, ρ a η. In the first three, the position of the participle is called attributive, in the last two, predicative. The commonest of the attributive positions is a ρ η, as in of the predicative, a n p, In one, the participle stands between article and noun, in the other, the noun stands between article and participle. Because these two orders are so common, one is always startled to find a noun standing between the article and a participle which, because of the context, can only be attributive, or, on the other hand, to find a predicative participle (or adjective) between article and noun. Now both of these phenomena are to be found in the best prose authors. Both are subject, however, to certain laws of order which, as I hope to show, are corollaries of the regular laws of attribution and predication.

My purpose in speaking of the attributive participle as ‘identifying’ has been to avoid the term ‘defining’, which applies to most but not all of the examples. There are some attributive participles that perform a different function, which I have called ‘characterizing’. In the standard grammars, the two types are not, as a rule, distinguished.1

By the time the Homeric poems were written the genitive absolute had already become productive; so any account of its origin must be based wholly on conjecture. It is reasonable to assume, however, that it did not spring spontaneously into being but grew up gradually from one of the earlier uses of the genitive case, and that it followed those laws of linguistic change that have been observed in constructions whose growth can be traced historically.