________________
DECEMBER, 1932]
NASALIZATION AND DENASALIZATION
(12) +ĭ, ǎ+, become i and i, respectively. Thus :Ap. Pr.*plihaï O.H. pthai. dùúnàü
pásyati, he will drink
dvigunakaḥ, two-fold
IAV. dúna.
1 For further particulars, see Grierson and Barnett, Lalla-vákydni (London, 1920), 137 ff.
Nasalization and Denasalization.
184. The nasalization of long vowels, by means of anunāsika, is a striking feature in all IAVs. It varies with the personal equation of the speaker or according to locality. The nasal tone is strongly heard in Audh, where Eastern Hindi is spoken, and in Northern Lahndā. In the latter, every final long vowel may optionally be nasalized, as in bahu or bahu, much; nā or na, of. Similarly (NP.Gr. v), in Northern Panjabi there is a tendency to nasalize the vowel of a syllable containing n or n. Thus, saņē or sānē, jāņā, jānd, or jana. Again, in Sindhi, nasalization is frequent, and imparts a distinct tone to the language. Long vowels, and especially final long vowels, are most subject to this, as in mih", rain (Skr. meghaḥ); pri, a friend (Skr. priyah). In the latter, there has been loss of a final syllable, with the old penultimate vowel lengthened in compensation. So, in Gujarati, final long vowels are often nasalized, as in mha, a face (mukha-), or duniya, the world (Ar. dunya) (GLL. 332).
[ §§ 184-185
Just as we have the above instances of spontaneous nasalization, so we sometimes come across sporadic cases in which an original nasal has been lost. This, again, is often a mere matter of personal equation, but in some languages it has crept into literature. Thus :
rinkhaï, ringai
samvi
rinkhati or ringali, he crawls sampayati, ho harasses
H. V rig, M. Vrag-, crawl; but G. Vrikh, roam.
0. yan; but M. Vasei, H. y satá-, G. Vsatáv-; P. V salau-, S. yadi.
paña, fifty
pannasam
M. pannas, A.Bg.O. pañcás, P.L. pañjáh, S. pañjáh; but H.G. pacás. In the last two instances, the denasalization has been induced by the stress-accent in the following syllable, but no such explanation is available for the first."
185. We have seen, under the head of lengthening of vowels (§ 168), how, when a conjunct consonant of which the first member is a class-nasal, the class-nasal is liable to be dropped, while the preceding vowel is lengthened and nasalized in compensation. But this compensatory nasalization is not confined to words which, in the literary Sanskrit or Prakrit, have conjuncts involving a class-nasal. In the IAVs. we sometimes come across cases in which there is nasalization of a vowel which has been lengthened in compensation, although there was no class-nasal in the literary Prakrit gonjunct which has been simplified. In the vulgar speech, not found in dictionaries, this is everywhere very common, but this unexpected nasalization also finds its way into literary words. E. g. :
áksikam, an eye
akkhiam
99
H.G. akh, B. akh', Bg.O. khi, EPh. CPh. #kha, but A. dkhi, P.L. akkh,
S. ákh, etc.
J. Bloch (FLM. § 70) and Turner (JRAS., 1921, 344) look upon this nasalization as due to the length of the vowel, which, according to Bloch spontaneously develops a nasal resonance, but I am unable to accept this explanation (see Grierson, JRAS., 1922, 381 ff.). Such spontaneous nasalization could occur only if it was introduced in the present stage of the development of the IAVs. in which the vowel had become long. But this is not the case, for the nasal was introduced not later than the Prakrit stage, and has nothing to do with the length of the vowel,