30S 
REPORT— 1847. 
speech in general. One of two characteristics is always required to eniWf 
us to pronounce an element apparently Celtic, which we detect in 
language, to be really so; either an internal one, which consists in the iv 
congruity or imperfect connexion of that element with the mass of ik 
language ; or an external one, consisting in the traceable history of its um- 
duction. 
W'ith reference to the first of these two characteristics, the most luiqne- 
tioDcd mark by which a great number of Celtic words in English and Gfraa 
betray their origin, Is their exhibiting, by the mode in which the sfrwjjdl 
quantity) and form of the one or several consonants (mutes) which tw; 
conlain are combined, the scale of articulation which belongs to the (dk 
and not to the Teutonic. My hearers will understand that I am aUudiHi: 
the interesting tact, discovered by James Grimm, that the Teutonic lo 
guagcft by a certain regular deviation from the ijhonrtico*elymological sjsw 
of the Sanscrilic languages—a deviation best known in Uiis counltTu# 
the appellation of Grimms /oro—have adopted a scale of arlicul^ ‘ 
their own, which in one portion of the Teutonic dialects, the High Gef» 
has undergone a second regular alteration. As we shall have still w 
on siwcral occasions to this fact, I think it necessary to correct prfTipti-; 
one great error, not loss iit matter than in name, by which the dmnousma* 
of the law' referring to it has hithciio been obscured*. Tins error fouw* 
in the undofined and confused signification given to the terras in'- 
metfiat aspiruta,^ which, instead of denoting, as they do, not only acconli^ 
to the nature of the matter, but abo to the definition given to them hj 
Greek and Latin grammarians, the three dift'erent degrees of 
{h\>v,tyt$) or (pianiUy of air with which every loiter may be uttered. anj«^ 
by the modern grammarimis, the first and second (/fTiww and 
noting the difiorence of the svrd mul rvcnl form of the consoiiaol, 
third {aspiratn) its alliance with a guttural sound, merely 
Greek and Latin, according to the scale of articulation adopted bj^ 
two languages, the tenuis, i. e. the feeble or short consonant, is, when » 
generally surd-, the uusUa, i. e. the consonant of middle strength 
of air, generally rural ; and the aspirata, i. e. the long or strong coW<' ^ 
generally atfectecl by a yvilnru-l articukiliov\ . But every disce«“»? 
will at once perceive that the English ih in dm, and German 
as well tenues, i. e. feeble, as the Erench tm tv, and in like manner, • 
the word deer, German tkier, Greek Oifp, the three mutes, d, th, B. 
apnrattr, i.strong, as well as in the word tm, french devx, Gernno • 
die three mutes, t, d, z, equally mediee. 
11, now, we apply Grimms law, thus corrected, to the Celtic langt^ 
find that they have prcservwl on the whole the scale of articulation 
mg to the classical languages, more particularly that of theSan^j^ 
which they correspond in attaching the long quantity, not as the 
Latin do, to the surd form of the mute, but to the vocal. I shall lU 
tins fact, with reference to the English and German also, by a table of 
panittve examples. 
V 1>aper, as far as he knows, was the first to discover 
+ •rl!*’ 9®^«brte Anzeigen, 18-11, No. 238, p. 877. . 
teiuies”* ** Sanserif, where surd mutes allied with « » 
