derivation from O.E. hearh, Old Northumbrian herg. He next contends that "names
like Alston, Ardington, Ardley, Ardwick,
Arley, Armsworth, Arrington, Arding, Ogden,
Orton, &c., have in many instances, like the
pronoun 'it,' and such common
as 'ostler,' 'arbour, 'ability,’ 'ermine,'
abundance,' &c., discarded an initial h."
The first statement is somewhat vague, but
there is nothing in these modern local names
to necessitate the assumption that they have
lost an initial aspirate. The reason why the
neuter pronoun has lost its h is that it is
the unstressed form that has survived, the
stressed form still preserving its aspirate in
some of the Northern dialects. The other
words cited by Mr. HarrisoN are of French
origin, and were introduced into the lan-
guage without an aspirate. He next gives
examples of the loss of h at the beginning of
the second member of a compound, but this
is altogether beside the question, the treat
ment of h in such a position being, as is well
known to philologists, different, even in early
times, from that of initial h. The physio-
logical nature of the English aspirate makes
it exceedingly difficult to pronounce it in an
unstressed position or at the beginning of a
second member of a compound
MR. HARRISON sophisticates my suggestion that Edgett may be derived from "edge" in the sense of "bluff," &c., by assuming that the word "seems to imply partial if not entire inaccessibility," and by then arguing that "there certainly does not appear to be much sense in a road over a cliff or precipice." I have said nothing about a road ; it is MR. HARRISON who has introduced it by an unscientific confusion of the meaning of O.E. geat with that of Old Norse gata (Gothic gatwo, German Gasse). A reference to the 'New English Dictionary have prevented this confusion. My words were "a gap or opening in high ground, a narrow pass.’ It is conceivable that a pass might be used as a road, just as the drifts are in South Africa, but a road going through such a gap would not be "a road over a or a precipice," as Mr. HARRIS0N assumes, cliff and it could not be called an "edgeway," for that, on the analogy of "ridgeway," would mean a road along an "edge," not one pro- ceeding across it by means of a pass or a gully
MR. HARRISON's not very recondite explanation of this name is that it is a corruption of "hedge-gate," which to me is a very unlikely compound either as a local or a personal name. Was not almost every gate a "hedge-gate"1 The difficulty about the loss of the h in America MR. HARRISON gets over by the airy generalization that the admixture of races in the United States renders them "exactly the country where a name is likely to go 'etymologically wrong.’
conflict with MR. HARRISON's conclusion, for
the American dialects are remarkable for their preservation of the initial aspirate, a characteristic that foreign immigrants are naturally unable to affect.
With regard to the name Eddiet in Domes- day, MR. HARRISON states that I have " been the unsuspecting victim of an erratic entry in Mr. W. G. Searle's 'Onomasticon Anglo- Saxonicum," and he adds that the context shows that the name is feminine. Both these assertions are wrong. I have been engaged for many years upon the compilation of an O.E. name-book and upon the study of the phonology of Domesday, and I am therefore hardly such a confiding novice as Mr. HAR- RISON assumes me to be. In the Survey in the vast majority of cases -iet represents the masc. name-stem -geat, but in one or two in- stances it stands for the fem. -gyo, which is normally represented by -id or -ida. Domes- day seldom indicates the sex of a name. In one instance only does Eddiet appear from the context to be fem.; it occurs thrice without any clue to its gender. Ed(d)ied
appears thrice as fem. and four times without any indication of sex. We cannot conclude that Eddiet is fem. in these cases in face of the fact that Leviet occurs both as masc.and as fem. (once only). The former represents the masc. Leof-geat, the latter the fem. Leof-gyo I therefore claim that Eddiet represents an O.E. Ead-geat. It is true that this name is not recorded in "any Anglo- Saxon docu- ment"; but this does not prove that it was not an O.E. name. There are scores of O.E. names recoverable from Domesday and later writings that are not recorded in the imper- fect O.E. remains. When MR. HARRISON says that the name strikes him as being "a rather improbable combination," I can only conclude that his acquaintance with the Germanic (and Indo-Germanic) name-system is imper- fect. There is nothing improbable in the combination, which is recorded beyond all cavil in its O.H.G. forms as Otkoz, Aotcaoz, both of which descend from a Germanic Audo-gauto-z, the prehistoric form of O.E Ead-geat.
The use of O.E. eca in the sense of " sword" is an instance of the figure synecdoche, and, indeed, the similar use of the Latin mucro for ensis is the first example of this figure given by Quintilian (lib. viii. c. 6, sec. 19).
imstances Here the facts of the case are in direct would