tongues; but a large number of them have come to us since 1500. Before
that date the chief languages from which it was possible for us to
borrow words were British or Gaelic, Irish, Latin, Greek (invariably
through the medium of Latin), Hebrew (in a small degree, through
the medium of Latin), Arabic (very slightly, and indirectly),
Scandinavian, and French. A few words as to most of these are
sufficient.

It is not long since a great parade was made of our borrowings from
"Celtic"; it was very easy to give a wild guess that an obscure word
was "Celtic"; and the hardihood of the guesser was often made to take
the place of evidence. The fact is that there is no such language as
"Celtic"; it is the name of a group of languages, including "British"
or Welsh, Cornish, Breton, Manx, Gaelic, and Irish; and it is now
incumbent on the etymologist to cite the exact forms in one or more of
these on which he relies, so as to adduce some semblance of proof. The
result has been an extraordinary shrinkage in the number of alleged
Celtic words. The number, in fact, is extremely small, except in
special cases. Thus we may expect to find a few Welsh words in the
dialects of Cheshire, Shropshire, or Herefordshire, on the Welsh
border; and a certain proportion of Gaelic words in Lowland Scotch;
though we have no reliable lists of these, and it is remarkable that
such words have usually been borrowed at no very early date, and
sometimes quite recently. The legacy of words bequeathed to us by the
ancient Britons is surprisingly small; indeed, it is very difficult
to point to many clear cases. The question is considered in my
_Principles of English Etymology, Series I_, pp. 443-452, to which I
may refer the reader; and a list of words of (probably) Celtic origin
is given in my larger _Etymological Dictionary_, ed. 1910, p. 765. It
is also explained, in my _Primer of English Etymology_ that, in the