On the Letter ' r.' 
125 
been placed by, I belieTe, every gi'ammarian who has ever wiitten, 
among the consonants. The vocal organs are rather active than 
passive during the utterance of the r ; and it is always used either 
before or after a vowel, never as a word by itself. 
The non-vibrant r, however, like two other letters which we are 
accustomed to regard as consonants, the w and y, though it possesses 
most of the characteristics of a consonant, does not, as prolonged, 
result from any very distinct action of the vocal organs. In sound- 
ing the y, there is an approximation of the back part or body of the 
tongue to the back part of the palate, but scarcely more than in 
sounding the vowel e. In sounding the non-vibrant r, there is a 
precisely similai* approximation of the tip of the tongue to the centre 
of the palate. In sounding the w, there is an approximation of the 
lips to each other, but hardly more than in sounding the vowel oo. 
Yet in all these cases at the moment of transition to a following 
vowel, there is a more decided pressure, and a distinct action, of the 
organs. We can easily recognize such action even where the y is fol- 
lowed by its cognate vowel, as in year (compare ear), or the w, as in 
wooes (compare ooze) ; and precisely similar action takes place in ray 
and rohe, provided the r is not rolled. Now suppose these three 
sounds respectively to follow theii' vowels instead of preceding them. 
Then we find them pronounced so feebly that one can hardly affirm 
there is any distinct action of the vocal organs. At present we make 
no distinction in the pronunciation of the diphthong in boy and noise, 
or between the ow of how and the ou of house ; yet it is possible that 
when the fashion was first established of writing a final y or w in 
these diphthongs rather than i or u, there may have been a quasi- con- 
sonantal force in the closing sounds, which is now lost. The r, 
however, though feebly sounded, yet maintains its existence even in 
this position ; for Dr. Guest's statement, unless he refers to the 
vicious pronunciation of uneducated Londoners, is certainly not ac- 
curate, when he says, " In his ordinary conversation the southern 
Englishman never pronounces a final r, unless it is followed by a 
vowel," while he acknowledges that even a Londoner would do so 
" if his attention were alive upon the matter."- 
There are however so many points of similarity between the non- 
vihrant r and the w and y, and they all three so closely approach the 
nature of vowels, resembling these in formation, while they resemble 
consonants in their use, that they seem to have a better claim than 
any other letters to the name of semivowels, by which I shall speak 
of them henceforth. Their proper order, as arranged on the princi- 
ple of beginning from the throat and advancing along the palate to 
the lips, will be r, w. 
But r is commonly spoken of as a liquid, a name which it, in 
* Phil. Soc. Proc, Vol. Ill, p. 7. 
