SHANSCRIT. 95 
place in the “ Puranas,’’ or Indian theogonies. According 
to the Puranica legends, Panini was the grandson of Devala, 
an inspired legislator; but whatever may be his history, to 
him the Sutras, or succinct aphorisms of grammar, are attri¬ 
buted by universal consent. 
His system is grounded on a profound investigation of the 
analogies in both the regular and the anomalous inflexions 
of the Shanscrit language. He has combined those analogies 
in a very artificial manner; and has thus compressed a most 
copious etymology into a very narrow compass. His pre¬ 
cepts are indeed numerous, but they have been framed with 
the utmost conciseness; and this great brevity is the result 
of very ingenious methods which have been contrived for this 
end, and for the purpose of assisting the student’s memory. In 
Panini’s system the mutual relation of all the parts marks 
that it must have been completed by its author; it certainly 
bears internal evidence of its having been accomplished by 
a single effort, and even the corrections, which are needed, 
cannot be interwoven with the text. It must not be hence 
inferred, that Panini was unaided by the labours of earlier 
grammarians; in many of his precepts he cites the authority 
of his predecessors, sometimes for a deviation from a general 
rule, often for a grammatical canon which has universal 
cogency. He has even employed some technical terms 
.without defining them, because, as his commentators remark, 
those terms were already introduced by earlier gramma¬ 
rians. None of the more ancient works, however, seem 
to be now extant; being superseded by his, they have 
probably been disused for ages, and are now perhaps totally 
lost. 
The inaccuracies of the Paniniya grammar were corrected 
by Catyayana, an inspired saint and lawgiver, whose history 
is involved in the impenetrable darkness of mythology. The 
amended rules of grammar have been formed into memorial 
verses by Bhartri-hari, whose metrical aphorisms, entitled 
“ Carica,” have almost equal authority with the precepts of 
Panini, and emendations of Catyayana. Bhartri-hari is said 
to have lived in the century preceding the Christian era. 
The text of Panini being concise and ambiguous, many com¬ 
mentaries were composed to elucidate it, of the chief of 
which Mr. Colebrooke has given an account. The best and 
most concise commentary now extant, is entitled the “ Casica 
vritti,” or commentary composed at Varanasi. Within a 
few centuries past, a grammar, well adapted for aiding the 
student in acquiring a critical knowledge of the Shanscrit 
tongue, has been compiled by Ramachandra, entitled 
“ Pracriyacaumudi.” 
When Shanscrit was the language of Indian courts, and 
was cultivated not only by persons who devoted themselves 
to religion and literature; but also by princes, lawyers, sol¬ 
diers', physicians and scribes; in short, by the first three 
tribes, and by many classes included in the fourth; an easy 
and popular grammar must have been needed by persons who 
could not waste the best years of their lives in the study of 
■ words. Such grammars must always have been in use; those, 
however, which are now studied are not, we believe, of very 
ancient date. The most esteemed is the “ Saraswata,” toge¬ 
ther with its commentary named “ Chandrica.” It seems to 
have been formed on one of the Caunmdis, by translating 
Panini’s rules into language that is intelligible, independently 
of the gloss, and without the necessity of adverting to a dif¬ 
ferent context. 
Another popular grammar, which is in high repute in 
Bengal, is entitled “ Mugd’habod’ha,” and is accompanied 
by a commentary. It is the work ofVopadeva, and pro¬ 
ceeds upon a plan grounded on that of the Caumudis; but 
the author has not been content to translate the rules of 
Panini, and to adopt his technical terms. He has, on the 
contrary, invented new terms, and contrived new abbre¬ 
viations. The same author likewise composed a metrical 
catalogue of verbs alphabetically arranged. It is named 
“ Cavicalpaaruma," and is intended as a substitute for the 
“ D’hatupata.” 
The best and most esteemed vocabulary of the Shanscrit 
is the “ Amc-ra cosha,” which, like most other Shanscrit 
dictionaries, is arranged in verse to aid the memory. Numer- 
< ous commentaries have been written on this vocabulary ; 
the chief object of which is to explain the derivations of the 
nouns, and to supply the principal deficiences of the text. 
Shanscrit etymologists scarcely acknowledge a single primi¬ 
tive amongst the nouns. When unable to trace an etymo¬ 
logy which may be consistent with the acceptation of the 
word, they are content to derive it according to grammatical 
rules from some root to which the word has no affinity in 
sense. At other times they adopt fanciful etymologies from 
Puranas or from Tantras. But in general the derivations are 
accurate and instructive. 
Amera’s dictionary does not contain more than ten thou¬ 
sand different words. Yet the Shanscrit language is very 
copious. The insertion of derivatives, that do not at all 
deviate from their regular and obvious import, has been very 
properly deemed superfluous. Compound epithets, and other 
compound terms, in which the Shanscrit language is pecu¬ 
liarly rich, are likewise omitted; excepting such as are 
especially appropriated, by a limited acceptation, either as 
titles of deities, or. as names of plants, animals, &c. In fact, 
compound terms are formed at pleasure, according to the 
rules of grammar; and must generally be interpreted in strict 
conformity with those rules. Technical terms too are mostly 
excluded from general dictionaries, and consigned to sepa¬ 
rate nomenclatures. The “ Ameracosh’’ then is less defective 
than might be inferred from the small number of words ex¬ 
plained in it. Still, however, it needs a supplement. The 
remaining deficiences of the Ameracosh are supplied by con¬ 
sulting other dictionaries and vocabularies, which are very 
numerous. 
The Shanscrit language is very copious and nervous; but 
the style of the best authors wonderfully concise. It far ex¬ 
ceeds the Greek and Arabic in the variety of its etymology, 
and, like them, has a prodigious number of derivatives 
from each primary root. The grammatical rules are also 
numerous and difficult, though there are not many anomalies. 
“ The Shanscrit language,” says Sir William Jones, (Asiat. 
Res. vol. i. p. 422.) " whatever be its antiquity, is of a 
wonderful structure, more perfect than the Greek, more 
copious than the Latin, and more exquisitely refined than 
either; yet bearing to both of them a stronger affinity, both 
in the roots of verbs, and in the forms of grammar, than 
could possibly have been produced by accident; so strong, 
indeed, that no philologer could examine them all three 
without believing them to have sprung from some common 
source, which, perhaps, no longer exists. There is a similar 
reason, though not quite so forcible, for supposing that both 
the Gothic and the Celtic, though blended with a very dif¬ 
ferent idiom, had the same origin with the Shanscrit, and the 
old Persian might be added to the same family.” 
The fundamental part of the Shanscrit language is^divided 
into three classes; viz., dhaat, or roots of verbs, slinid , or 
original nouns, and ev?/a, or particles. The latter are always 
indeclinable, as in other nations; but the words compre¬ 
hended in the two former classes must be prepared by certain 
additions and inflexions to fit them for a place in composi¬ 
tion. Here the art of the grammarian interposes, as not a 
syllable, nor a letter, can be added or altered but by regi¬ 
men, nor the most trifling variation of the sense in the 
minutest subdivision of declension or conjugation can be 
effected without the application of several rules; and all the 
different forms for every change of gender, number, case, 
person, tense, mood or degree, are methodically arranged for 
the assistance of the memory; resembling, though on an 
infinitely more extensive scale, the compilations of propria 
qua; maribus and as in praisenti. 
In the Shanscrit language, the three distinctions of genders, 
viz., masculine, feminine, and neuter, are preserved in their 
common number and order. A Shanscrit noun, in its first 
formation from the general root, exists equally independent 
of case as of gender. It is neither nominative, nor genitive, 
nor accusative, nor is impressed with any of those modifica¬ 
tions, which mark the relation and’’connexion between the 
several members of a sentence. In this'state it is called an 
imperfect: 
