66

With respect to those considerations unconnected with astronomy, from which the Reviewer says he could distinctly prove, "that the high antiquity "attributed to the Hindu records is founded on evi"dence of a nature almost conclusive," we wish he had stated those weighty considerations, or told us where we might find them; for the astronomers and others now engaged in investigating the antiquities, arts, and sciences of India, are unwilling to take his ipse dixit for it; particularly as he had but the moment before totally destroyed the credibility of those very records he would wish to support, by saying, that no work of any great antiquity can " exist in a country where the art of printing is un"known, free from interpolation." How is it possible then, that they are to be considered as ancient records, when every line of them may be interpolated? who can pretend to judge of those parts which are genuine, and those which are not? for certainly it is not necessary that a part that is interpolated should have any date or mark annexed to it, by which it might be known; therefore the authenticity of works so interpolated, must be as fully to all intents and purposes destroyed, as if the whole were an actual forgery.

The Reviewer should only judge for himself,for that evidence which he may think is of a nature almost conclusive, may be no evidence at all to others. And I am afraid, that unless his gymnosophists find a better advocate in their cause, their pretensions to superior antiquity, to arts, and to sciences, must soon fall to the ground. Lastly, the Reviewer says,

" By exhibiting the mean result only, we have "given Mr. BENTLEY's argument an advantage to "which it is not entitled; the individual results "from each of the ten data vary from 300 to 1100

t

the knowledge of the key to their interpretation, the student finds them ambiguous. In the application of them when understood, he discovers many seeming contradictions; and, with every exertion of practised memory, he must experience the utmost difficulty in combining rules dispersed in apparent confusion through different portions of PAN'INI'S eight lectures. A commentary was therefore indispensably requisite. Many were composed by ancient grammarians to elucidate the text of PANINI. A most copious one on the emendations of his rules was compiled in very ancient times by an uncertain author. This voluminous work, known by the title of Mahábháshya, or the great commentary, is ascribed to PATANJALI, a fabulous personage, to whom mythology has assigned the shape of a serpent. In this commentary every rule is examined at great length. All possible interpretations are proposed: and the true sense and import of the rule are deduced through a tedious train of argument, in which all foreseen objections are considered and refuted; and the wrong interpretations of the text, with all the arguments which can be invented to support them, are obviated or exploded.

Its defi

VOLUMINOUS as it is, the Mahábháshya has not exhausted the subject on which it treats. ciencies have been supplied by the annotations of modern grammarians. The most celebrated among these scholiasts of the Bhashya is CAIYATA, a learned Cashmirian. His annotations are almost equally copious with the commentary itself. Yet they too are loaded by numerous glosses; among which the old and new Vivaranás are most esteemed.

THE difficulty of combining the dispersed rules of grammar, to inflect any one verb or noun through all its variations, renders further aid necessary. This seems to have been anciently afforded in vocabularies, one of which exhibited the verbs classed in

the

acknowledged it. For, as POPE very justly says, respecting the moral qualities of a good Critic: "Tis not enough WIT, ART, and LEARNING join; In all you speak, let TRUTH and CANDOUR shine. It is much to be lamented, that the very reverse of this is but too often the case, and that men suffer their judgment to be biassed by their prejudices.

By exhibiting the mean result of ten different operations *, viz. 731 years for the age of the Súrya Siddhanta, the Reviewer conceived he did me more justice than I was entitled to; and therefore, to counteract it, as he thought, instead of giving the whole of the different results, from which his readers would be enabled to form a just opinion, he makes choice of the two extreme results, as differing most from the mean, and concludes from thence, that either the heavenly bodies were so inaccurately observed by the author as to furnish no basis for calculation, or that the observations were made at a period prodigiously anterior to that given by me.

Now, it must be immediately apparent to any man of common sense, that by taking the two extreme results only, no other inference could, consistently with truth, be drawn from thence, but that the work must have been written at some period between these extremes; the mean of which 1105+340-722 years.

In computations, depending on a number of observations, it is well known that astronomers reject such as are found to differ most from the mean results; for in all cases some of the data, from their

* These were the results which the Reviewer ought to have given his readers.

MOON's apogee, gave 605 years. || JUPITER,

......

875 years.

nature, will be more erroneous, and less to be depended on than others. Had the Edinburgh Reviewer, therefore, adopted this plan, and rejected the extremes, 1105 and 340, as too incorrect, no fault whatever could be found with him for so doing; for the remaining eight results would still have been more than sufficient to answer the purpose required.

But his views, as may be easily seen, were to endeavour, if possible, to discredit any investigation that should in the smallest degree tend to open the eyes of the public with respect to the true antiquity of Hindu books; and therefore he asserts, that the heavenly bodies must have been so inaccurately observed by the author, as to furnish no basis for calculation, or that the observations were made at a period prodigiously anterior to that assigned by me. Why did he not point out what these errors were, that his readers might judge of the truth or falsehood of his assertions?

But in order to shew the fallacy of the Reviewer's argument, let us endeavour, if possible, to ascertain the quantity of the errors from the years only, on which the Reviewer grounds his notions.

The years are obtained by dividing the error in the position of the planet, at a certain instant, by the error in the mean annual motion, which, by its gradual accumulation, is supposed to have caused the error in position. Therefore, suppose we denote the error in position by æ, and that in the mean annual motion by y, and that =1105; it is required from thence, to determine the quantities r and y, which the Edinburgh Reviewer would wish to make his readers believe, must be so extraordinarily great as to leave no basis for calculation : I say it is absolutely impossible, nor does the nature of the case admit of such an unjust inference. For

1

any two quantities whatever, whether large or small, that are in the proportion of 1: 1105, will give the same quotient. Thus, suppose 2=1105 minutes, and y=1 minute, then, 105-1105. Again, suppose r=1105 seconds, and y=1 second, then, 5=1105, as before. Or, suppose r=221", and y=0, 2", then,=1205, as before. Hence it evidently follows, that as 1105 may be deduced from any two quantities, however small, that are in the proportion of 1: 1105, so may 340 from any other two quantities whatever, small or large, that are in the proportion of 1: 340. It is, therefore, the heighth of absurdity to pretend to draw any conclusion relative to the supposed quantity of error from the years exhibited; and if we wish to shew the errors, it must be done by a direct computation, and not by ideal notions or sophistry.

The Reviewer perhaps conceived that all the results should come out exactly the same; if so, it is more than he had a right to expect from the most correct European tables extant. If we examine the second edition of LA LANDE's tables, we shall find that one of the data will give us 318 years for the age of it, and another 243 years: but would this be a sufficient ground to assert, that either the heavenly bodies were so inaccurately observed by the author as to furnish no basis for calculation, or that the observations were made at a period prodigiously anterior to that assigned to LA LANDE's second edition? The error from which the 243 years arise, only amount to about one minute and half, which may shew the Reviewer, that he is not to assume the quantity of the error from the number of years. There are, perhaps, no astronomical tables in existence, that do not contain errors, but these errors are always less at or near the time the work is written than at any distant period whatever. Therefore, to put this matter out of dispute, I shall exhibit, in the