When they have reached the age of fifteen, they begin to study a commentary on the grammar (Sûtra), and spend five years on learning it. And here I-tsing gives the following advice to his countrymen, many of whom came to India to learn Sanskrit, but seem to have learned it very imperfectly. "If men of China," he writes, "go to India, wishing to study there, they should first of all learn these grammatical works, and then only other subjects; if not, they will merely waste their labor. These works should be learned by heart. But this is suited for men of high quality only.... They should study hard day and night, without letting a moment pass for idle repose. They should be like Confucius, through whose hard study the binding of his Yih-king was three times cut asunder, being worn away; and like Sui-shih, who used to read a book repeatedly one hundred times." Then follows a remark, more intelligible in Chinese than in English: "The hairs of a bull are counted by thousands, the horn of a unicorn is only one."

I-tsing then speaks of the high degree of perfection to which the memory of these students attained, both among Buddhists and heretics. "Such men," he says, "could commit to memory the contents of two volumes, learning them only once."

And then turning to the heretics, or what we should call the orthodox Brahmans, he says: "The Brâhmanas are regarded throughout the five divisions of India as the most respectable. They do not walk with the other three castes, and other mixed classes of people are still further dissociated from them. They revere their Scriptures, the four Vedas, containing about 100,000 verses.... The Vedas are handed down from mouth to mouth, not written on paper. There are in every generation some intelligent Brâhmans who can recite those 100,000 verses.... I myself saw such men."

Here then we have an eye-witness who, in the seventh century after Christ, visited India, learned Sanskrit, and spent about twenty years in different monasteries—a man who had no theories of his own about oral tradition, but who, on the contrary, as coming from China, was quite familiar with the idea of a written, nay, of a printed literature: and yet what does he say? "The Vedas are not written on paper, but handed down from mouth to mouth."

Now, I do not quite agree here with I-tsing. At all events, we must not conclude from what he says that there existed no Sanskrit MSS. at all at his time. We know they existed. We know that in the first century of our era Sanskrit MSS. were carried from India to China, and translated there. Most likely therefore there were MSS. of the Veda also in existence. But I-tsing, for all that, was right in supposing that these MSS. were not allowed to be used by students, and that they had always to learn the Veda by heart and from the mouth of a properly qualified teacher. The very fact that in the later law-books severe punishments are threatened against persons who copy the Veda or learn it from a MSS., shows that MSS. existed, and that their existence interfered seriously with the ancient privileges of the Brâhmans, as the only legitimate teachers of their sacred scriptures.

If now, after having heard this account of I-tsing, we go back for about another thousand years, we shall feel less skeptical in accepting the evidence which we find in the so-called Prâtisâkhyas, that is, collections of rules which, so far as we know at present, go back to the fifth century before our era, and which tell us almost exactly the same as what we can see in India at the present moment, namely that the education of children of the three twice-born castes, the Brâhmanas, Kshatriyas, and Vaisyas, consisted in their passing at least eight years in the house of a Guru, and learning by heart the ancient Vedic hymns.

The art of teaching had even at that early time been reduced to a perfect system, and at that time certainly there is not the slightest trace of anything, such as a book, or skin, or parchment, a sheet of paper, pen or ink, being known even by name to the people of India; while every expression connected with what we should call literature, points to a literature (we cannot help using that word) existing in memory only, and being handed down with the most scrupulous care by means of oral tradition.

I had to enter into these details because I know that, with our ideas of literature, it requires an effort to imagine the bare possibility of a large amount of poetry, and still more of prose, existing in any but a written form. And yet here too we only see what we see elsewhere, namely that man, before the great discoveries of civilization were made, was able by greater individual efforts to achieve what to us, accustomed to easier contrivances, seems almost impossible. So-called savages were able to chip flints, to get fire by rubbing sticks of wood, which baffles our handiest workmen. Are we to suppose that, if they wished to preserve some songs which, as they believed, had once secured them the favor of their gods, had brought rain from heaven, or led them on to victory, they would have found no means of doing so? We have only to read such accounts as, for instance, Mr. William Wyatt Gill has given us in his "Historical Sketches of Savage Life in Polynesia,"[274] to see how anxious even savages are to preserve the records of their ancient heroes, kings, and gods, particularly when the dignity or nobility of certain families depends on these songs, or when they contain what might be called the title-deeds to large estates. And that the Vedic Indians were not the only savages of antiquity who discovered the means of preserving a large literature by means of oral tradition, we may learn from Cæsar,[275] not a very credulous witness, who tells us that the "Druids were said to know a large number of verses by heart; that some of them spent twenty years in learning them, and that they considered it wrong to commit them to writing"—exactly the same story which we hear in India.

We must return once more to the question of dates. We have traced the existence of the Veda, as handed down by oral tradition, from our days to the days of I-tsing in the seventh century after Christ, and again to the period of the Prâtisâkhyas, in the fifth century before Christ.