Fortunatov's Law and tolkAppiyar's rules

[Vidhyanath Rao]

Sudalaimuthu Palaniappan <Palaniappa at AOL.COM> wrote:

> To me it looks as if Fortunatov's law, if we take into account Levitt's
> finding regarding the splitting of *l, is almost a restatement of Tamil
> morphophonemic rules beginning with tolkAppiyam.

I would disagree with this.

First of all, change of l occurs in other contexts as well:
kal vs ka_tka (_t = _r, the alveolar two-tap of modern school Tamil),
nuul vs nuu_tpaa, nuu_tpu (spinning) etc. This can be seen in Tamil
borrowings from Sanskrit: kalpanaa => ka_tpa_nai, alpa => a_tpam etc.

Secondly, in Tamil, l+t => _t_t, but Fortunatov's law says that
lt => .t, without compensatory lengthening of the previous vowel
(palta => pa.ta etc). It is this aspect that I find most mysterious
about this law. Most other changes seems to preserve heavyness of vowels.

Another problem is the occurance of `galda' in RV. This led Burrow to
put the lt => .t change at RV times.

---

BTW, what is the current opinion about Fortunatov's law? Gonda, in
``Old Indian'', puts quotation marks around the word law and otherwise
discounts it. Mayrhoffer, in KEWA, did not use it. Has Burrow changed
people's minds?