Sacred Texts  Buddhism  Index  Previous  Next 
Buy this Book at

Chinese Buddhism, by Joseph Edkins, [1893], at

p. 398



Changes in Chinese sounds since the time of the Buddhist transliteration of Indian words—Examples of Sanscrit words in old and new Chinese—The importance of translations made in A.D. 60 to A.D. 76 for reading the Four Books—The Hindoo translators did not speak pure Sanscrit—Sanscrit was the language of the books—No Pali books in China—The translators spoke Pracrit—The term po-li, "glass"—Use of Sanscrit words in magic—Dharani—Inscription in six languages at Kü-yung kwan.

THE Chinese characters have been written in the same form and with the same sort of pencils since the time of Wang Hi-chi, A.D. 350.

During these fifteen centuries, while the writing taught in all schools has been unchanged, the sounds attached to the characters have been in a state of slow and constant flux.

Thus, the translator Kumarajiva wrote his name with four Chinese characters then called Ku-ma-la-zhip. They are now Kieu- (or Chieu) mo-lo-shï.

All sonant initials, such as g, d, b, z, zh, j, have changed in the interval to surds, viz., k, t, p, s, sh, ch. In words pronounced with the tone called hia-ping, the aspirates k’, t’, p’, ch’ come in place of k, t, p, ch, which occur in words pronounced with the tones hia-ping, hia-ch‘ü, and hia-ju.

Final m has changed to final n. Finals k, t, p have been dropped; also the vowels have all changed their values, a to o, ya to e, u to yeu, &c.

The compilers of Kang-hi's dictionary have provided tables of the old sounds, with characters chosen to represent

p. 399

the pronunciation as it formerly was. They are to be read with the powers of the letters of the Sanscrit alphabet. Natives not knowing the Sanscrit letters cannot escape from the confusion in which they are involved by the difference between the old and new pronunciations. The foreign student will find that the principle here laid down is a key to unlock the difficulties of the subject.

The following examples will help to familiarise the learner with the method:—


Old Chinese.

New Chinese.




Amogha Vajra

A-mo-ga bad-ja-ra

O-mo-k‘ia po-che-lo











Pat-la-mit-ti 1












Ma-ha-pa-de-ma 2











Shakradeva Indra

Shak-ka-la-de-ba In-da-la

Shï-kia-lo-t‘i-p‘o Yin-t’o-lo




The admission of the principle that the Chinese pronunciation has changed, and that the recognised Mandarin orthography is nothing more than that of a modern dialect, will be found to throw a light much needed on the use of Sanscrit by the Chinese Buddhists.

It is also necessary to recognise the principle, that the Hindoo Buddhists in China were men who spoke the dialects of Central India, Northern India, &c.

p. 400

M. Stanislas Julien's Méthode pour Déchifrer et Transcrire les noms Sanscrits makes no allusion to these subjects. The consequence is that all his immense industry has failed to make a window that would have illuminated this dark room. Yet that does not hinder his work from being indispensable to the student on this subject. It gives the Sanscrit words. It gives the modern Chinese pronunciation. These two factors are tabulated alphabetically. The student can with this help proceed rapidly. But if he wish to understand why such and such Chinese characters were chosen, and not prosecute his researches mechanically, he must allow for the influence of dialects and the incessant change of language on both sides of the Himalayas.

It is necessary to take the finals of the Southern Chinese dialects, and the initials of the dialects spoken in Central China at the present time, as our sign-posts, pointing out to us what was the pronunciation of the T‘ang dynasty and of the previous age; and this must be done with the addition of aid from the Japanese and Corean transliteration of Chinese sounds, through the spread of Buddhism many centuries ago.

It was about three hundred and forty years after the death of Mencius, and five hundred and fifty after the death of Confucius, that the translations from Sanscrit were made. By learning the powers of the Chinese syllabary with the help of the transliterations then made, we can come quite near to the classical age of Chinese literature, and approximate to the actual pronunciation of the great Chinese sages. For the method and proofs, I may here refer the reader to my Introduction to the Study of the Chinese Characters.

Particularly is the "Sutra of Forty-two Sections" worthy of attention, on account of its being the translation of Kashiapmadanga and his friend Chu-fa-lan. It is highly important for fixing the pronunciation of the Chinese, at the time when they taught Buddhism at Lo-yang, in the reign of Ming-ti, A.D. 58 to A.D. 76. From their use of

p. 401

characters it is clear, that at that time the modern Fo was But; O-lo-han was A-la-han; ch‘an, "contemplation," was dian or dan; Nie-p‘an, "Nirvâna," was Nit-ban or Nir-ban; Kia-she, the name of "Kashiapa" Buddha, was Ka-shap or Ka-shiap; P‘u-t‘i, the word Bodhi, "knowledge," was Bo-di. Sha-men, the "Shramana," was Sha-men, having about the same sound as now. Pi-k‘ieu or Pi-ch‘ieu, the "Bikshu," was Bi-k‘u. Ch‘iau-ch‘en-ju, for "Godinia," was Go-din-nia. O-na-han, a certain grade in discipleship, was A-na-gam, agreeing with the Sanscrit "Anagama." Pi-chï, for the Sanscrit "Pratyeka," was P‘ak-tie, the Pali being Patiekan. So it was probably not Pali that Kashiapmadanga spoke, though he was a native of Central India. Sü-t‘o-hwan, for "Srotapanna," another grade of discipleship, was So-da-ban or Su-da-wan. The last of these is the more likely, for the character is the same as that used in writing "Nirvâna." The Pali is Sotapan; so that the translator did not speak Pali.

The greatest initiator of change in the choice of characters was Hiuen-tsang, about A.D. 645. He altered the characters according to his opinion of what the selected symbols ought to be. His selection of characters is a gauge of the pronunciation of his time. His translations, however, have not become popular. The older usage of words has kept its place.

The language in which the Buddhist sacred writings were first compiled may have been Pali; but that from which they were translated into Chinese was Sanscrit. The Pali books were a separate set of originals. The Sanscrit originals alone are known to the Chinese. The manuscripts, the inscriptions, the charms cut on copper mirrors, the lucky sentences under eaves and over doors in monasteries, are in Sanscrit; and in polyglot books printed at Peking, Sanscrit is the language employed.

Koeppen, page 186, in saying that the Chinese also have a number of Pali texts, has been misled by Gutzlaff.

p. 402

[paragraph continues] This missionary had lived in Siam, where Pali is the sacred language, and was there accustomed to the idea that Pali was the original language of Buddhism. This view he brought with him to China, and when he saw Sanscrit inscriptions in the island of P‘u-to, he took them to be Pali. From him the opinion spread, but it is an error. The Buddhists of Birmah, Siam, and Ceylon have never spread their religion in China or Japan, or introduced their sacred books into those countries.

The Nepaul Buddhists preserve the sacred books in Sanscrit, and not in Pali. But Burnouf also found certain portions of the Nepaulese books written in Pracrit. The groundwork was Sanscrit. The language occasionally used was Pracrit. The language known by the Chinese as the Fan language was shown to be undoubtedly Sanscrit, by Julien's version of the works of Hiuen-tsang, the traveller who visited India, and who has described the Sanscrit language in his autobiography. It is the language of "Brahma" (Fan; old sound, Bam).

Brahmanical ideas form a strong element in Buddhism. Sanscrit words and Sanscrit writing are peculiarly sacred in the view of the Brahmans. This idea has been borrowed by the Buddhists. They preferred to use the words and writing which were most sacred. With this Shakyamuni would naturally have nothing to do. His instructions were oral. He was a great moral teacher and metaphysical logician. It was his disciples in the centuries that followed him that introduced Sanscrit writing, as the chief medium of recording his instructions. It is they that are responsible also for the charms, and for the faith in magic which stimulated their use.

So many Brahmans announced themselves believers in Shakyamuni's doctrines, that Sanscrit became at once a favourite medium for the embodiment of his teaching by writing, even though Shakyamuni himself spoke Pali or Pracrit, as he probably did.

In the same way it may be said that Pali was then so

p. 403

extensively spoken, that it was inevitable that it should, in the region watered by the Lower Ganges, become also a medium for the preservation of the sacred books.

This double form of the sacred books had much to do with the separation that sprang up between the Northern and Southern schools of Buddhism. The peculiarities of the Chinese transcription deserve to be considered.

The Pracrit of the early Chinese translators was, for example, nearer to Sanscrit than to Pali in the sound of prajna, "wisdom." The characters adopted are directed to be pronounced pat-nia. The Pali is paññya.

There was also in the Pracrit of the early Chinese translators a very clear pronunciation of b for the Sanscrit and Pali v. This is shown by the constant selection of Chinese characters sounded with b or p, according to the old pronunciation. For example, the city "Vaishali," near the modern Patna, is spelt Bai-sha-li. The Pali sound is Vesali. Dr. E. J. Eitel, in his Hand-book of Chinese Buddhism, page 27, has said, that "Chinese texts consider Pali as the ancient and Sanskrit as the modern form even as regards the system of sounds." If he will direct his attention to these facts, he will perhaps admit that not the Pali, but a certain Pracrit form or forms of the Hindoo language, prevalent at the time in Central and Northern India, was or were at the basis of the Chinese old texts, The Hindoo translators in China would have Sanscrit texts chiefly before them, and Pracrit texts occasionally. Their pronunciation was not pure Sanscrit, but was modified by Pracrit peculiarities.

In the flourishing period of Buddhism, in the region watered by the Ganges, at the time of the Greek invasion, and afterwards, the art of writing lately introduced was put to extensive use in the Buddhist monasteries. Those institutions fostered education, which was then very much in Buddhist hands. While the people spoke Pali and Pracrit, Sanscrit was the language of education, and hence the fondness shown for it by the Buddhists.

p. 404

Burnouf held that there was a double text, a Pracrit text for the laity and a Sanscrit for the literati.

King Kanishka in Cashmere called a council, the fourth; and in the writings edited by this assembly, Sanscrit was the language.

The language of Magadha in the time of the emperor Ashôka was a Pracrit. This was probably much used by the Hindoo Buddhists who came to China.

An argument against Pali is to be found in the careful selection of Chinese words commencing with sh, to represent the Sanscrit sh; but sh is not a letter known to the Pali, just as it was wanting in the ancient Greek and Latin. The original text of the early Chinese translations before the days of Hiuen-tsang must have had sh fully developed. It probably dropped ra in sharira, "a relic," and sti in Shravasti, name of the capital of an ancient kingdom called Kosala, and lying near Kapilavastu.

I place here some remarks on po-li, "glass," a favourite word. In Buddhist glossaries, the Chinese po-li is derived from the Sanscrit word spatika, "crystal." Many of the Hindoo Buddhists who came to China—perhaps all—spoke dialects of Sanscrit, but not the Sanscrit itself. The s was dropped, and the final ka. The t in ti became l, as in the Turkish belur.

The rock crystal of China comes from Turkestan, and would bring its own name with it from that country. Buddhist makers of glossaries would prefer to derive the word from Sanscrit, as the mother of all knowledge. They have passed over without remark the possibility that the Chinese word may come from the Turkish.

The word po-li for "glass," formerly pronounced pa-li, has been in common use in China since the T‘ang dynasty. It came in with Buddhism and the international trade with Turkestan.

I believe that the initial s in spatika might be an accretion and not original, just as most probably smelt is later than melt, and sneeze than nose, and stannum than tin.

p. 405

Curiously we find in the Mongol vocabulary bolor, "crystal," "glass;" bolor daboso, "rock salt;" bolo ch‘ilagon, "a polishing stone," a rolling stone used in smoothing the clods of a ploughed field. Compare Turkish bileghi, "whetstone." Let it be noticed that glass-dust is used by polishers and grinders.

Whether the bôli or bali is of Turanian origin and has originated the Sanscrit spatika, it would be interesting to know. Ballur is Arabic for "crystal;" spashta is Sanscrit for "clear;" berrak is Turkish for "clear," "limpid." Probably here is the root; but who shall decide?

In Buddhist magic there has been extensive use of the Sanscrit characters. The doctrine of magic has been developed by the Buddhists very systematically, and to an almost unexampled extent. It arose from the same tendency in the Hindoo mind, which produced those vast fictions in the description of the universe, and in the narrative of the past, that distinguish the native literature of that country. The love of the wonderful led the Hindoo authors to forsake, at the same time, the fair bounds of history and the sober reality of nature. Here it is easy to perceive a similarity to the Arabians. There is, in their fictions, the same fondness for splendid scenes and striking supernatural effects. This would be poetry were it not very much overdone. The same circumstances of gaudy magnificence are again and again repeated, and the reader is wearied with the unending recital of marvellous events, invented after one model, and whose one object is to excite an undistinguishing admiration of the power displayed.

By magic is here meant the supernatural power attributed to the Buddhas and Bodhisattwas, or claimed by the ordinary priests, and which is exercised by charms, mystic formulæ, incantations, finger-postures, and such-like means.

It is not the power of God acting through nature that is here intended, but the power of the priest, through his charms, virtues, and superior knowledge. The magical result is effected by the never-erring retributive fate which

p. 406

is the cause of everything that occurs, and which is responsive in the most complete manner to Buddhist wisdom and goodness.

The use of the Devanagari writing for the purposes of magic is an instructive instance of the power of superstition to delude the human mind. The words used by the magician for the most part have no meaning. They are senseless clatter. The sounds are Sanscrit, but the words usually not so. These absurd compositions of unmeaning sounds are of various lengths. They occur frequently in the books of the Great Development. They are engraved on stone monuments on the way side, on imperial roads, and at places of resort for Buddhist pilgrims. They also form a chief part of the liturgies in use in the monasteries and at funerals.

Om-mani-padme-hum is one of the most common. Padme is "lotus;" mani is a "precious stone;" om is a sacred "Hindoo symbol." It is written in Sanscrit characters under the eaves of all the lama temples in Peking. In these temples it meets the eye everywhere.

The Thibetan character is based on the Sanscrit. It is also found cut on monuments, both for charms and for intelligible inscriptions. It is the chief language for liturgical use among the Thibetans and Mongol lamas in Peking, except in two instances. In the Mahakala miau, the Mongol sacred books are read. In a temple, Fa-hai sï, near the hunting park, the Manchu is read. The Chinese lamas in Peking read Thibetan prayers, while the Chinese priests of the old Chinese Buddhism read, of course, in Chinese.

In all these forms, the syllables of the charms are the same. They are written in Sanscrit, or in the other languages mentioned.

At the pass called Kü-yung kwan, near Peking, there is a stone monument containing a charm in six languages, viz., Sanscrit, Chinese, Thibetan, Nü-chih, Ouighour, and Mongol. It was cut in the time of the Mongol emperors.

p. 407

[paragraph continues] It contains the same charm written with the characters employed for all these languages. It was intended as a protection to the emperor in going to and coming from the summer palace, at that time beyond Tu-shï k‘eu, and also to all travellers on this much frequented road between China and Mongolia.

There are also some monuments inscribed with Sanscrit charms in Peking at the present time, which date from about seven hundred years ago. They are stone octagonal pillars. One is at the monastery called Hwa-yen sï, near the park of the Altar of Heaven and the city gate known as Kiang-cha men. These octagonal pillars are called shï-chwang, and they are placed in the courts of temples. There is one kept on the premises of the London Missionary Society in Peking.

Sanscrit inscriptions are supposed, like pagodas and monasteries, to have a lucky effect on the neighbourhood where they are found, and on those who erect them by their benefactions and goodwill.

A muttered charm is called "dharani," or, in Chinese, cheu.


399:1 S. Julien is wrong in making the first of these four characters end in n. It is pat in old Chinese; but pat was often par. See p. 201 of my Introduction to the Study of the Chinese Characters. Thus the famous word karma, "cause," "fate," was transliterated kat-ma, the t being heard as r.

399:2 The character de should be transliterated dek. That the k was then lost is shown by its use in this case. The loss of k final was beginning.

Next: Chapter XXVI. Books and Papers That May be Consulted for the Study of Chinese Buddhism