Law of palatals is not a triumph (trump?) card

"…how far a,e,o are preferable as original Indo-European vowels as
against a. I have discussed at length elsewhere (1975, New Lights on
the Indo-European Comparative Grammar, Varanasi, p.46-49) that Indo-
European e_/i_ show some effect on the preceding labiovelars in
Centum and velars and labiovelars in Satam languages. The details of
such changes in Indo-Iranian, Armenian, Slavonic, Hellenic and
Germanic have been shown with illustrations. But there are many
exceptions to palatalization which have also been shown there, e.g.,
Sanskrit hatah, Av. Jato < Indo-European g(w)hn-tos which has no
palatal vowel and the palatalization in such cases has been explained
as generalisation forms like Sanskrit hanti, Av. Janti. Similarly,
there are exceptions in other languages. With a reconsideration of
the whole situation it may be said that the exceptions which are also
too many might be pointing out the fact that this law was hastily
formed to satisfy the anti-Sanskrit inferiority feeling of some
scholars. The present position of Indo-European primary vowels a,e,o
needs revision with the discovery of Anatolian languages. Indo-
European a,e,o (accepting the present reconstruction) Sanskrit a, Av
a, OP a, Germanic a,e, Baltic a,e, Slavic o,e, Hittite a,e, other
Anatolian languages a (Lycian has e, which corresponds to Indo-
European a as well). Latin and Celtic show a confused state of things
which needed the reconstruction of a,e,o in Indo-European, so that
Lat a, e/I, o/u can be derived at least from three vowels a,e,o than
one vowel a. If we take the number of languages, most of the
languages (Sanskrit, Av, OP, Luwian, Palaic, Hieroglyphic Hittite)
show a. Many languages (the Germanic, the Baltic and Hittite) show
a,e, Slavonic o,e and a few (Hellenic, Italic) need the
reconstruction of a,e,o. One question has never been solved. Why
should a,e,o become a in Sanskrit? Since Sanskrit shows change of a
to e,o in some sandhi forms of ah. In Balto-Slavic there are so many
similarities. But Baltic shows a < Indo-European a,o and Slavic shows
o < Indo-European a,o. It is easier to explain if both would be
derived from Indo-European a only. There are many such problems which
need detailed discussion…Therefore, law of palatalization should in
no way be used as a triumph card to show that Sanskrit has lost its
importance. Sanskrit is so much important for Indo-European
comparative grammar that even now any new research has to take the
help of the Sanskrit evidence.

"…The evidence from the Gypsy language…clearly shows a marked
preference of e for Indo-Aryan original a in European Gypsy. Besides
sometimes a > o in Gypsy. Since there is no doubt now that Gypsy is
an Indo-Aryan language and that Gypsies had India as their original
home, it is now high time to reconsider the problem whether Indo-
European a > a,e,o in Greek and some other Indo-European languages
and remained a in Sanskrit. Therefore, from a linguistic point of
view Sanskrit has a very archaic stage…" (SS Misra, 1992, The Aryan
Problem, a linguistic approach, Munshiram, pp. 88-89).