124 
On the Letter ' r.' 
Mr. Garnet is speaking, when he says that in Northumberland and 
some parts of Germany, the sound meant for r has no lingual vibra- 
tion at all, but becomes a deep guttural, neither very easy to describe 
nor to imitate, but almost exactly corresponding to the Arabic 
The following observations however concern almost exclusively the 
two sounds which we have in English, and which in this paper will 
be distinguished as the vibrant r and the non-vibrant r. 
In considering what place r ought to occupy in a well-classified al- 
phabet we must not lose sight of this distinction between the vibrant 
and the non-vibrant r. It is doubtless of the latter that one of our 
most advanced English scholars has said that he knows not how to 
distinguish it as other than a vowel. Let us briefly examine this 
point. 
We all know there are certain sounds which grammarians recognize 
as vowels, others which are called consonants, though definitions of 
these terms are rarely met with. According to the distinction im- 
plied in the words themselves, a vowel can be sounded by itself, 
while a consonant can only be " sounded with " a vowel. But s, sk, 
ih surd, th sonant, r, &c., can be perfectly sounded without any vowel; 
and if on these definitions r becomes a vowel, so must all these others. 
Or shall we, with Dr. Latham, define a vowel as a sound in the 
formation of which ''the passage of the breath is uninterrupted " ? 
Again the non-vibrant r becomes a vowel ; but so does Z, in the for- 
mation of which the passage of the breath is not interrupted, though 
it is restricted ; and m, in which the breath passes uninterruptedly 
through the nose, though restricted to that exit. Nor can we escape 
this difficulty by varying the above definition and substituting "un- 
restricted " for " uninterrupted," as no sounds whatever pass from 
the human mouth without some restriction or limit : even the most 
open vowels are denied egress through the nasal cavity, which, as I 
have shown elsewhere,! is open only to m, n, ng. 
The definitions which I venture to propose, are the following : — 
A vowel is a simple sound produced by the passage of the breath 
through the mouth, its character depending on the position of the 
tongue and lips, which organs remain passive during the utterance of 
the vowel ; and such sound, if prolonged, is capable of independent 
use as a word. A consonant is a simple sound emitted from the 
mouth alone or the mouth and nose, and resulting from some pressure 
or other action of the vocal organs ; such sound always preceding or 
following a vowel, and being incapable of being used as a word by 
itself. If these definitions are accepted, it will follow that r, whether 
vibrant or non-vibrant, is not a vowel, but must remain where it has 
* Phil. Soc. Proc, Vol. Ill, p. 111. 
+ See Transactions of the Philological Society for 1856, p. 21. 
