All the Germanic languages agree in using the dative inflection to express what comparative grammar has shown to be an IE ablative function of comparison. It is not my intention to treat the broader aspects of the question further than to emphasise the fact that this function is not one that belongs to the genuine IE dative; but it was taken over by the Germanic dative along with other ablative, instrumental, and locative functions when the case inflections that were originally associated with these functions in IE disappeared during the Proto-Germanic period. What we call the Germanic dative case is therefore in function an ablative, an instrumental, ora locative, as well as a dative; and from the very beginning we may see one of the causes for the early disuse of the construction here studied, namely, the ambiguity in the context beside these other dative functions.

In attempting to establish beyond a doubt the existence of the case of comparison as a comparatively frequent and idiomatic locution in Old English, I am fully aware of the not uncommon tendency of the special investigator to color the context of a given passage to conform to a favorable interpretation. This is often an unconscious habit and perhaps excusable in one who must by close examination at all times seek to discover the subtle relationships that bind the elements of thought together. I have found upon careful examination that I had fallen into such a danger many times where grammatically the case might stand in two or more relationships and where the general meaning was not especially clear. I have avoided such passages whenever I could bring semantics to the aid of syntax, but, on the other hand, I hope I have not failed to record any occurrence in which both syntax and semantics pointed to the comparative relationship.

This is part of the Offa passage, which may not be quite as old as the original Widsiđ (see Chambers 202), though certainly not a very late interpolation. No verb is expressed in this sentence, and the five words of line 40 permit of certain variations in syntax. From the predication that has preceded, however, something like ‘achieved’, ‘won’ or ‘had’ is understood, thus : ‘No man of the same age won greater valor in single combat than he.‘ I scan line 40a as type C. The word efeneald, whenever it is used as a adjective, may also govern the dative, but it does not govern the dative when used as a substantive, which function it undoubtedly has here. Only the genitive accompanies the substantive in OE : his efeneald, etc., and him would be superfluous if taken with efeneald.

Since the extant prose (with the exception of slavish glosses) is by its very nature closer to the vernacular than the poetry, we should probably not expect to encounter in it as large a proportion of those constructions associated with compact, synthetic speech as might be found in the poetry. Also, if there is a progression or a shifting with respect to any phenomenon in the language, we might reasonably expect to find it reflected in the prose rather than in the poetry. And so, the decline of our case-construction in OE speech is probably more accurately indicated by its use in prose writings.

A study of the case of comparison would not be complete without taking into account the remarks that scholars, beginning with Grimm, have made about a possible genitive or an instrumental of comparison. One cannot read a grammatical treatise touching upon comparison without encountering rather annoying hints about the possibility of a true genitive of comparison in Germanic.

The material in the foregoing chapters has been presented with a view to advancing our knowledge of comparison in Germanic and particularly in English. It therefore seems desirable to sum up briefly those points in grammar that have been the subject of demonstration, or that may be said to be better understood because of the light thrown upon them i1n the course of the present study. This investigation was begun solely through a kind of intellectual curiosity aroused in the course of a previous study, and so there is no thesis to defend or hypothesis to justify.