A Study of the Four-Fascicle Lankavatara Ratna Sutram
In a Set of Four Texts:
A Sanskrit Restoration, English and Japanese Translations with Introduction,
and the Collated Gunabhadra Chinese Version with Japanese Reading
An English Translation (Not for Sale)

Published by Gishin Tokiwa (Professor emeritus, Hanazono University, Kyoto)
4-17 -1, Nishi-awaji, Higashi-yodogawa-ku, Osaka, Japan
Printed by the Meibunsha Printing Co. Ltd., Kyoto, Japan.
June 2003,

Lankavatara Sutram

A Jewel Scripture

of Mahayana Thought and Practice

Translated by Gishin TOKIWA


What I am presenting to the modern world is a mahayana Buddhist scripture that belongs to the early fifth century, the Lankavatara-ratna-sutram ("The Jewel Scripture [Named] Entering Lanka"), in my tentative English translation of the Langga-abadala-baojing, four fascicles, its earliest Chinese version translated in A.D. 443 in the dynasty of Liu Song by a Buddhist monk from India, named Gunabhadra. Speaking more correctly, this translation of mine was' made from a Sanskrit text restored by me from the Gunabhadra Chinese version through a thoroughgoing revision of the current Sanskrit text, Lankavatara Sutra, edited by Dr. Bunyiu Nanjio and published from The Otani University Press in 1923.

The scripture is considered to have been compiled in Sri Lanka, a land of Theravada Buddhism in the days when mahayana was


prevalent even under the Theravada reign. This situation seems to explain why it is full of critical thoughts. It is critical of the religious thoughts of both Buddhist traditionalists and non-Buddhists, especially the Samkhya philosophers. It seems to be from its criticism of the latter's thought that it unfolded its "tathagata-garbha" thought, a very important way of thinking that does not seem to have received a fair appreciation from among most modern Buddhist scholars. In its manner of treating the "tathagata-garbha" as the original mode of being of the "alaya-vijnana," our ordinary manner of being, the present scripture manifests a very critical attitude not only theoretical but practical as well. I understand the Lankavatara-ratna-sutram represents an authentic mahayana standpoint.

Unfortunately there have appeared few appropriate introductions of this scripture from its Gunabhadra version, and there are reasons for that. Let me explain why I have adopted the Gunabhadra Chinese version as the standard text.

As was already introduced by the late Dr. Daisetz Teitaro Suzuki seventy years ago,*1 the mahayana scripture Lankavatara sutram in printed form is available in one Sanskrit text, three Chinese versions, and two Tibetan versions.

The single Sanskrit version, available in printing, is an edition by Bunyiu


Nanjio, Kyoto 1923. By the way. No. 3 of the Buddist Sanskrit Texts, Saddharmalankavatdrasutram, edited by Dr. P.L. Vaidya, published by The Mithila Institute of Post-Graduate Studies and Research in Sanskrit Learning, Darbhanga 1963, is a handy book for students as it provides them with the Nanjio text, without its footnotes, and an index of the first pada (eight-syllable half-line) of all the verses in the ten chapters ("prathamam parisistam slokasuci") at the end of the book. At the beginning of the book it has rough contents of all the chapters and an introduction in English and Hindi by Mr. Srisitamsusekhara Vagaci (Bagchi, in English). Air. Bagchi praises Dr. Suzuki's introduction to the world of this scripture through the latter's English translation and Studies. Mr. Bagchi, then, poses several questions on the basic standpoint of the scripture. His questions, though posed according to his understanding of the text which had not gone through almost any text-critique, seem to have found not a few who shared them with him. I hope his questions have fully been responded to in my introduction of this scripture. Edition by Dr. Vaidya, as far as the text is concerned, seems to mean no more than his choice of more approproate wordings from among those shown in Dr. Nanjio's footnotes, omitting the latter from his text.

In connection with this, mention must be made about an attempt that has already been begun to check all the available manuscripts to have a more reliable text. It is: "A Revised Edition of the Lankavatara Sutra Ksanika-Parivarta" (Sixth Chapter), Tokyo 1981, by Dr. Jikido Takasaki, the then professor of Tokyo University. It was "A Report of the General Research C for the Years 1978~80" by a group represented by Prof. Takasaki (with 4 pages of preface and


74 pages of the text). In 1993 a faculty colleague member of mine at Hanazono University let me know about this revised chapter. Upon my request, though it was twelve years after publication. Prof. Takasaki promptly sent me a copy, so that I could make reference to it in my work. According to Prof. Takasaki, most of the manuscripts and microfilms made use of for this research belong to the time after the nineteenth century besides the three, which belong to the eighteenth century. In his preface Prof.. Takasaki writes (P. 4) that he has made more use of the Gunabhadra Chinese version than the two other Chinese versions in his research for the reason that, with its style strange as a Chinese translation but more faithful to the Sanskrit word order, and with the date earliest of the extant versions, one could expect of it to offer powerful resources for the attempt to seek an original form of the Lankavatara sutra. He writes (P. 2) that the present report, which covers pages 220~239 of the Nanjio edition, is just part of his research. I am ignorant of this work of his thereafter.

The three Chinese versions are those translated: (1) by Gunabhadra in Liu Song A.D. 443, four fascicles, Taisho Tripitaka vol. 16, no. 670; (2) by Bodhiruci in Wei A.D. 513, ten fascicles, Taisho Tripitaka no. 671; and (3) by Siksananda in Tang A.D. 700-704, seven fascicles, Taisho Tripitaka no. 672.

The two Tibetan versions are those translated: (1) from Sanskrit, now in the Tibetan Tripitaka Peking edition, vol. 29, no. 775; and (2) from Gunabhadra' Chinese version by Facheng of Dunhuang (Chos-grub in Tibetan), in the reign of King dPal-lha gTsan-po, possibly an early period of the ninth century, Tibetan Tripitaka Peking ed., no. 776.

According to Fazang (643-712), a well-known Huayan philosopher


in the early Tang dynasty in China,*2

Siksananda, in 698 when he finished translating the Avatamsaka (Huayan) sutram at the Foshouji Temple in the divine city Luoyang, was ordered by the Empress Zetian Wu to translate the Lankavatara sutram. Succeedingly receiving the royal order, he also made a translation of it. Before finishing it, Siksananda entered the capital Zhang'an by cart, with the order to stay near the palace, and settled down at the Qingchan Temple. When he finished a rough translation, but before he checked it, Siksananda left for home country with the royal permission. Then in 702 Mituoshan (Mitasana?), another Buddhist monk-scholar from Tukhara, who had stayed in India for twenty-five years and who, having studied the three pitakas, was well versed in the scripture Lankavatara sutram, was ordered, with the help of sutra-translator monks, Fuli, Fazang, and others, to check the translations to make a final version. .Fuli is to make a composition of the royal introduction to the scripture. I shall mention in praise as follows....

Fazang mentions the reason why this new translation was started under the royal order of Empress Wu:*3

The four-fascicle [Gunabhadra] version has wrong compositions which are observed endless and whose word-order is that of the western tongue, so that a supreme person of outstanding wisdom does not know how to understand it, while ignorant people and mediocre persons forcibly make wrong conjecture and understanding. Meanwhile, the ten-fascicle [Bodhiruci] version is known to be slightly furnished with literary quality, but it hardly manifests the noble meaning of the Buddha. Placing additional characters and mixing up constructions,


it obscured the meaning or caused errors. Its use of peculiar wordings finally resulted in preventing the undoubtedly evident principle from prevailing. Her Majesty the Queen, who lamented this hardness of understanding, ordered another attempt of translating this scripture. This time, provided with the detail of five Sanskrit manuscripts, we will check the two Chinese versions so that we can adopt what was good and correct what was wrong. Years of excellent job will exhaust the core of it, so that students will be happy being free from errors.

This certainly means that the seven-fascicle Siksananda-Mitasana Chinese translation was considered to be the finest version in China. Because of this understanding Dr. D.T. Suzuki adopted this version for his English translation. That being the case, how should we understand the reason that the Tibetan translation by Facheng, Peking edition no. 776, was made from the four-fascicle Gunabhadra Chinese version a century after this? My answer to this question is that the Gunabhadra version conveys the earliest, original Sanskrit text-form whereas the two other Chinese versions as well as the extant Sanskrit manuscripts that include the Nanjio-edition did not go through any kind of appropriate text-critique. Let me refer to an example for the complete lack of text-critique in these latter versions.

I invite the readers to have a look at the second division of the second fascicle of Gunabhadra's Chinese version in my translation:


"NII Section Fifteen: The Four Conditions That Make Great Practitioners.

The underlines there show how the other versions, i.e., one Sanskrit and two Chinese, which correspond to one another as far as this section is concerned, suffer corrections when they are corrected in accordance with the Gunabhadra version. The corrections include shifting the order of words or passages, supplementing the text with passages which are lacking, and correcting words. Correcting words in two of the three cases there results from the shifting of passages, which means a change in the context. Seen from the Gunabhadra version, the need for shifting of passages means how the other versions have missed right places for those passages. For details I ask readers to check them in my translation. Here I shall make a brief explanation.


(G) Gunabhadra.'s Chinese version, Taisho Tripitaka, no. 670, vol. 16, pp. 489b ~ 490a;

(T) Facheng's Tibetan rendering of the above, Tibetan Tripitaka Peking edition, no. 776, vol. 29, pp. 96, 235c8~97, 237c1;

(N) Nanjio's Sanskrit version, pp. 7913~ 827;

(B) Bodhiruci's Chinese version, Taisho Tripitaka, no. 671, pp. 529c ~5 30a;

(S) Siksananda's Chinese version, Taisho Tripitaka, no. 672, pp. 599c ~600a .

According to the Gunabhadra version, the Buddha told Mahamati, a representative of the audience, that Awakening great beings (my


rendering of bodhisattva-mahasattvas) will become great practitioners when they are equipped with four conditions:

(1) Clearly ascertaining that one's own mind is seen as something external,

(2) (N801-2) Observing that external beings are non-existent,

(3) (N7918~801) Desisting from the views of arising, staying, and breaking up, and

(4) Delighting in the noble wisdom personally attained.

The numbers after the capital N in the brackets show those of pages and lines of the Nanjio-edited Sanskrit text. The order of the two conditions, (2) and (3), which had to be exchanged according to the Gunabhadra version, makes the whole passage quite consistent.

In the above (4) the two words "Delighting in" being underlined shows that the term "abhilaksanataya (by seeking for)" in the Sanskrit text (N802)(B529c28 %% S599c8 %%) had to be corrected to "abhiramanataya (by delighting in)," according to G (%%%%%%; T: "bde ba gya nom pa thob ste").

After this the Buddha's exposition of each of the four conditions continues. In the text the exposition of (2), which should come after (1) (N80s~12), was shifted from N8117~825, a part located more than a


page after (1) in the Sanskrit text, with the addition of a closing part ("he gets versed in the non-existence of external beings") by the translator. This is followed by (3), (4), and (5), but the final part of (3) (N8013~813) had to be supplemented with two lines from N8115~16. As for (4), it is further divided into three parts, a, b, and c. The beginning part of (4a) and the whole (4c), which are lacking in the text, had to be supplemented by the translator. The end of (4b) (N8115) had to be supplemented with a line from N824-5, with an accompanied correction from "abhilasate (seek for)" to "abhiramate (delight in)." As for the last correction, it is justified by the context itself. To summarize: The order of divisions of the Buddha's exposition in the Sanskrit text before corrections was (1), (3-), (-4a), (4b-3ending), (2), (4c), (5). (Underlines show lacunae.)

I wish my readers may imagine this section before all these corrections were made. It would be easy to see how hard it is to read through all the passages. But that is the real case with the extant Sanskrit, the Bodhiruci and the Siksananda Chinese versions (and the Tibetan translation from the Sanskrit). No wonder Dr. D.T. Suzuki, who made use of the Siksananda version for his English translation, commented on this scripture as full of inconsistencies. It is quite natural that Facheng ("Chos-grub" in Tibetan) had to


attempt a Tibetan translation from Gunabhadra's Chinese version a century after the Siksananda version came into being. Let me quote from Dr. Suzuki's Introduction to his Studies in the Lankavatara Sutra (p. 1718~p. 184):

As I noted elsewhere (Essays in Zen Buddhism, Vol. I, p. 75) the whole Lankavatara is just a collection of notes unsystematically strung together, and, frankly speaking, it is a useless task to attempt to divide them into sections, or chapters (parivarta), under some specific titles. Some commentators have tried to create a system in the Lankavatara by making each paragraph somewhat connected in meaning with the preceding as well as the succeeding one, but one can at once detect that there is something quite constrained or far-fetched about the attempt. If this, however, is to be done successfully, the whole arrangement as it stands of the paragraphs must be radically altered; and this redaction is possible only by picking up and gathering together cognate passages which are found promiscuously scattered throughout the text, when for the first time a kind of system would be brought into the text. As the present form stands, passages of various connotations are juxtaposed, and a heading indicating one of the ideas contained in them is given to the whole section, thus artificially separating it from the rest. Gunabhadra had done the wisest thing by simply designating the entire sutra as "The Gist of the Buddha's Teaching" (buddhapravacanahridayam).

I think Dr. Suzuki was almost right in this remark. What he missed is that he did not recognize the Gunabhadra version as the


very standard upon which all the work of rearranging the other versions could be carried out. But failure in this recognition was not limited to Dr. Suzuki alone. It was, honestly speaking, mine as well until I finished a tentative Japanese translation (from the first through the ninth of the whole ten chapters) of the Nanjio-ed. Sanskrit text in December 1994.*4 Immediately after this, when I was about to make an English translation of this scripture, I was still thinking of translating from the Sanskrit text with references to the Gunabhadra Chinese version, for I had realized the importance of the latter to some extent. In the next year when I was preparing an article on the Lankavatara sutra's concept "the body made of thought (manomayakaya)" and the Chan-founder's idea of "wall-contemplation (biguan)," I realized my Japanese translation of the very passages quoted above from the scripture lacked thorough text-critique.*5 Only then did I begin translating Gunabhadra's Chinese version, correcting the Sanskrit text so as to have the latter express the former. In the process I have come to know that the extant Sanskrit text is full of defects, examples of which were shown above. Some of the defects are due to confusion in the order of manuscript page numbers, and others derive from faulty copying. Copyists seem to have lost good manuscripts in a very early period after


compilation. This situation was not improved in the Bodhiruci and Siksananda versions and the Tibetan translation from Sanskrit.

Dr. Suzuki, who adopted the Siksananda version for his translation and study, refers to the work of the Japanese commentator Kokan Shiren (1278-1346), Butsugoshin-ron, completed in 1325. He mentions (Studies pp. 43~44):

We can thus almost say that there are as many subjects treated in the Lankavatara as it can be cut up into so many separate paragraphs, each paragraph consisting sometimes of a prose part and its corresponding verse, but sometimes in long or short prose part only, not accompanied by verse. The same subjects are sometimes repeated more or less fully. The Japanese commentator Kokwan Shiren, who is also the author of a history of Japanese Buddhism known as the Genko Shakusho in thirty fasciculi, divides the Gunabhadra version of four fasciculi into eighty-six sections including the last chapter on "Meat Eating." This is the most rational way of reading the sutra, as in each of his sections only one subject is treated.

Through understanding one subject clearly that is treated in each section Shiren must have believed that he could approach the message of the whole scripture; I agree to the method adopted by this patient, wise, and profound commentator in the Kamakura period; I have learned much from him.

Dr. Jikido Takasaki, the then professor of Tokyo University,


published a book entitled Ryogakyo, as one of the volumes in a series of Buddhist texts, in 1980. In this book he chose the Gunabhadra version from among the extant versions of the Lankavatara sutram, and, making use of the division of the scripture which Shiren had applied in the Butsugoshin-ron, wrote a series of lectures on this scripture. Dr. Takasaki agrees with Dr. Suzuki in appreciating Shiren's approach, since, according to Dr. Takasaki, this scripture is a collection of fragments of various mahayana teachings. He writes (in my translation):

The title of the second chapter of the Sanskrit text, "a collection of all the dharmas as many as thirty-six thousand," is what acknowledges itself to be just a patchwork of dharma-teachings; most of the chapters have one subject in it representing the whole chapter, like that of the third chapter, "Impermanence;" besides, there is no necessity for the chapter on Impermanence to come in sequence after the chapter on the "collection of all the dharmas."*

In the prefatory note to the same book Dr. Takasaki writes about his "unexpected discovery of the importance of the Song translation [Gunabhadra version, made in the Liu-Song dynasty in the south, A.D. 420-479] as a text." He writes (in my translation):

The Song translation is a Chinese text hard to read, indeed, but as it is nearest to the original text, it has a merit in offering conveniences for assuming the original text-form. Since there are occasions in which it offers materials for


correcting the extant Sanskrit text, I have come to think it might be possible to suppose a recension different from the extant Sanskrit text to be the original text of the Song translation. It is I that chose the Song version for the text, but, in that it has given me a motive to enter the study of the Lankavatara sutram to take charge of the present article in the series for lecturing on Chinese versions of the Tripitaka, I feel much obliged to the editorial staff of the publishing company. (Preface, pp. 2-3)

According to my understanding, the original text from which Gunabhadra made his four-fascicle Chinese version in China in A.D. 443 had been lost for some political reason in Lanka at an early stage after he left for China; instead, only imperfect manuscripts were preserved somewhere else among people who had little knowledge of the scripture. About one hundred years later, when Bodhiruci made his ten-fascicle Chinese version in Wei, the text he used is considered to have been little different from the extant Sanskrit text; the latter corresponds to the seven-fascicle Chinese version translated in Tang, with three additional chapters attached to the four-fascicle Chinese version .*7

According to "Gunabhadra Biography," the eighth of the ten biographies given in the earliest extant record of the translated Buddhist scriptures already issued, Chusanzang-jiji, compiled by


Sengyou in Liang, Gunabhadra, who left Lanka aboard a ship sailing eastward, landed Guangzhou, after extreme hardships, in the twelfth year of Yuanjia, in the reign of Emperor Wen of Liusong (A.D. 435). He gained the emperor's awe and respect, had two influential persons as disciples, and upon request from Buddhist monks made translations of many scriptures with more than seven hundred persons helping his work (Taisho Tripitaka 55, 105c). The translation, Langga-abatara-paojirig, according to Daoxuan's Datang-neidian-ji Fascicle 4, was made in the twentith year of Yuanjia (A.D. 443) (T 55,258c).

Quotations from and references to this sutra, like %%%%, zixinxian-liang, a rendering characteristic of the Gunabhadra version for "svacittadrsyamatram ('what is seen as something external is nothing but one's own mind')," are seen in a collection of teachings of the "Dharma-master," founder of Chan, directly recorded by one of his disciples, Tanlin, together with those introduced through teachings and words by other disciples of the same Dharma-master, including Huike, also recorded by Tanlin, in one book.*8 From this we know that this sutra in the Gunabhadra version has had deep influences on the formation of the Chan/Zen thought.

My own interest in this sutra has been aroused and strengthened


as I have come to think it certain that sources of the Chan/Zen thought are fully unfolded in it. Quotations by Chan masters have been made from the Gunabhadra version, so that for students of the Chan thought understanding of this scripture has been essential. But it has many peculiar wordings which have prevented people from having good understanding of it. I am convinced that it is worthwhile to introduce this scripture in a readable form to the modern world. In that sense Dr. Takasaki's Ryogakyo is a precious forerunner for those concerned.*9

One of the difficulties one encounters in reading the Gunabhadra version seems to derive from an unnatural way of translation. Here is an example:

(Taisho 16, 487b19) [%%. %%%%. %%%%%%%%%%%.] %%. %%%%%%%%%%%%%%%.

(N651') "[punar aparam mahamate agotram kim yaduta icchantikanam anicchantikata moksam /] tatra icchantikanam punar mahamate anicchantikata moksam (N66 ) kena pravartate."

(Tokiwa I. N Section 8-5) Mahamati, [by the non-approach (agotram) I mean "the nature of not desiring emancipation of those who desire it (icchantikanam anicchantikata moksam)."] As for the nature of not desiring emancipation of those who desire it, (N66) why does it take place (kena pravartate)?


This is the beginning part of explanation of the last of the five kinds of approach to truth. The other four are: approaches to truth by the faithful follower-disciples' vehicle (sravaka-yana), by the solitary practitioners' vehicle (pratyekabuddha-yana), by the tathagatas' vehicle, and by the unfixed ones. The fifth is, according to the Gunabhadra version, a special approach (prthag-abhisamaya-gotra), and according to the Sanskrit text, the non-approach (agotra). In the above I adopted the Sanskrit expression, for it seemed to match the context better.

Here we have various problems. Gunabhadra uses the same term "—%% (yichandi)" twice without rendering it into a meaningful Chinese term, first for a Sanskrit noun in the genitive [plural] case, made from the verb "icchanti ([they] wish)" (icchantikanam, "of those who desire"), and then for a Sanskrit abstract noun of the same term "icchantika" in the negative form (anicchantikata, "the nature of being one who does not desire"). The four characters "% %%%" ("emancipation from worldliness" for "moksam"), which serve as a common object of "desire" and "don't desire,"being placed after the two terms,"%%%%%%%," look like independent of them. Then, the final six characters will come to make one group, and the whole fourteen characters might be read as: "An icchantika is not an


icchantika. As for emancipation from worldliness, by whom will it be made to prevail?" The Sanskrit interrogative, "kena," which has the meaning of "by whom," would better be taken here to mean "how" or "why," in its another meaning. Only through such analysis based on the comparison do we have a pretty good way of reading this passage. But that is impossible for those who read the Chinese rendering alone. Here is a corrected form of the above Chinese passage:


Another difficulty we experience in reading Gunabhadra's version lies in the peculiar word-order he uses in rendering Sanskrit into Chinese. Very often he follows the Sanskrit word-order besides the Chinese. Here is one example:

(Tl6,p.495b27) %%. %%%%%%%%%.

(Nanjio 122" ) "etaya catuskotikaya mahamate rahitah sarvadharma ity ucyate."

(Tokiwa II Section 31-1) Mahamati, those in which this set of four alternative propositions are freed from are called all that have their own characterisitcs.

In the ordinary word-order of the above expression in Chinese, the character % comes before %%% (%%%%, li-cisiju); Gunabhadra,


however, follows the Sanskrit word-order in this regard ("rahitah," i.e., "freed," for % comes after "etaya catuskotikaya," meaning "from this set of four alternative propositions," for %%%: cisiju-li). Meanwhile, in the latter half, %%%%% (shi ming yiqiefa), he follows the Chinese word-order for "sarvadharma (for —%%) ity ucyate (for %%)." In the above rendering by me I agree to Gunabhadra's way of reading. Although the Sanskrit passage may possibly be read as: "All that have their own characteristics are said to be free from this set of four alternative propositions." But this is inappropriate as a remark by the Buddha, as it would prove to be a bystander's utterance. The corrected form of the above passage is:


Anyway, most of the difficulties we encounter in reading Gunabhadra's version can be overcome by carefully comparing it with Sanskrit expressions. Where we have no Sanskrit expressions, the Tibetan rendering by Facheng serves as a good help, though often the Tibetan translator misunderstood the Chinese expressions.*10

As I read through the Gunabhadra version in the manner I describe above (that is to say, comparing it carefully with the extant Sanskrit


text and the Tibetan translation by Facheng), the result of which I show in my tentative English translation, I feel that we should be free from modern man's arrogance to blame it as being inconsistent either in the manner of connecting sections and chapters or in that of expounding the contents of each section or chapter. Instead, I feel it most appropriate for us to carefully listen to the unfolding of a unique logic through the Gunabhadra version by the sutra-compilers. Blaming this earliest Chinese version as inconsistent without making any attempt to read it according to its original context means nothing but modern man's negligence and shame. That is why I have tried making poor efforts to restore the Sanskrit form of the Gunabhadra Chinese version throughout my translation work, by collating the extant Sanskrit text. I sincerely hope some specialist in Sanskrit may check my restored Sanskrit text. Also in my way of reading the Chinese version I am afraid I have left not a few parts incorrectly understood. Gratitude is mine when they be corrected.

As it was with my Studies Report for IRIZB (1994), throughout my present translation work I kept conferring with Dr. D.T. Suzuki's English translation (1932), Dr. Kosai Yasui's Japanese translation (1976), Dr. Akira Suganuma's Japanese translation (1977, 78, and


81), in which he kept referring to Jnanasribhadra's commentary extant in Tibetan, as well as Shiren's commentary in Chinese, Butsugoshinron (1325), as mentioned above. Besides, I often conferred with a Chinese commentary made in the eleventh year of Hongwu of Emperor Taizu (1398), Ming Dynasty, Langga-abatara-baojing, by two monks, Zonle and Ruji (Taisho 39, no. 1788). The two commentaries in Chinese are both very excellent. This time I did not take up Jnanasribhadra's Arya-lankavatdra-vrtti (Tibetan Tripitaka, Peking vol. 107, no. 5519), as I did not think it helpful for collating the extant Sanskrit text, though he seems to have offered his own view on the main themes dealt with in every chapter, before the tenth, Sagathakam, of the scripture.

After finishing this English translation and restoring the Sanskrit form of the Gunabhadra Chinese version, I have made my Japanese translation of the same text. I hope publication of these may help people have better understanding of the mahayana thought which has deeply inspired practitioners of Chan/Zen throughout its history.

Finally let me express my hearty gratitude to those who have encouraged me in finishing this translation work by giving some of their names: Late Mr. Soko Morinaga (Rinzai-zen master, former president of Hanazono University), Mr. Seizan Yanagida, professor


emeritus (Chinese Chan and Japanese Zen Buddhist history) of Kyoto University, and former director and life-member of the International Research Institute for Zen Buddhism, Hanazono University, (IRIZB); Mr. Urs App, former assistant professor (Chan Buddhism) at IRIZB; late Mr. Shun Murakami, research fellow (Chinese Buddhism) at the IRIZB; Mr. Noritoshi Aramaki, former professor (Indian and Chinese Buddhism) at the Institute for Research in Humanities, Kyoto University, now professor at Otani University, Kyoto, Japan; and Mr. Lambert Schmithausen, professor (Indian Buddhism) at Hamburg University, Germany.


Osaka, October 2002



I. Central Messages Conveyed by the Lankavatara-ratna-sutram

The Gunabhadra Chinese version has four fascicle-chapters, whereas the Nanjio-edited Sanskrit text in the part which covers the same contents has seven chapters, beginning with the second and ending with the eighth. (The former does not have the latter's first, ninth, and tenth chapters.) The Gunabhadra version's first and second fascicle-chapters correspond to the second chapter of the Sanskrit text, and the former's third and fourth fascicle-chapters to the latter's remaining six chapters, beginning with the third and ending with the eighth.

The Chinese version's first and second fascicle-chapters, which, thus, correspond to the Sanskrit second chapter, represent a conclusive whole, as the latter's second chapter's title, "A Collection of All Teachings Thirty-Six Thousand," indicates. The number 36,000 seems to stand for that of the Buddha's whole teachings for bodhisattvas. (Consider the Buddhist interpretation that each of the six perfections of bodhisattva practice is said to include all the six perfections. Then divide 36,000 by 108, another number which is


regarded as representing all the passions-emancipations and which is used by the Buddha as the number of questions to be asked and of answers to them. We have the eternal continuation of the number, three: 333.333333...) Taking this into consideration, the present English translator has corrected his early division of this part, thirty-five sections, into thirty-six, as is seen in the list of contents shown separately.

The rest of the scripture, the third and fourth fascicle-chapters, which include six Sanskrit chapters or twenty-nine sections according to my division (i.e., twenty sections of Chapter Three, five sections of Chapter Six, and four other chapters, each of which constitutes one section), center on the subject "the buddha" from the Buddha's standpoint. It is no wonder that sections of the latter half of this scripture (e.g., the "Five Grave Sins"; "Tathagatas Utter Not Even a Word"; "The Two Directive Principles of the Awakened Truth: The Effected End (siddhanta) and Communication (desana)"; "Meaning as the moon and Sounds and Letters as the moon-pointing finger-tip") have invited deep concern of Chan/Zen practitioners.

Throughout the sutra various topics are discussed between the questioner Mahamati and his respondent, the Buddha. Unlike questioners in other sutras, Mahamati would not just sit and listen


and overjoy with the other's sermon. He often equates the Buddha's views with those of non-Buddhists, and tends to blame the other for the lack of uniqueness. To this the Buddha tries to clarify his point of view, comparing it with other views. This kind of discussion helps readers better understand the Buddha's viewpoint. It sounds consistent enough all through, and often very convincing. Here for the sake of introducing the contents of the scripture, a brief sketch will be attempted of a few subjects picked up from sections (which are attached with head numbers common to the two translations, the restored Sanskrit text, and the Contents) throughout the Gunabhadra version.

A. "The One Hundred and Eight Terms":

3. GI, NII Section One (2): Mahamati Asks 108 Questions

4.     NII Section One (3): The Buddha Recounts Mahamati's Questions

5.     NII Section One (4): The 108 Terms Shown by the Buddha

After his expression of respect to the Buddha at the beginning of the scripture, a mahayana practitioner, an Awakening being (bodhisattva) named Mahamati ("Having-a-Great-Wisdom") asks questions on everything he thinks of as serious enough to ask, both worldly and supra-worldly. Hearing this, the Buddha encourages the other to ask more, and finally cites one hundred and eight terms


as his answers to those questions. They can be represented by one term: "The term 'dharma' not being the term 'dharma' (dharmapadam-adharmapadam)," as is supplemented by the present English translator as the one hundred and eighth term, which is lacking in the extant Gunabhadra version (Cf. NII Section One (4): The 108 Terms Shown by the Buddha).

Here the term "dharma" means "something that holds its own characteristic(s)," so that the above statement can be expressed as "the term 'something that holds its own characteristic(s)' not being the term 'something that holds its own characteristic(s)'." What matters here is that the whole statement constitutes one term. That this statement be applied to everything worldly and supra-worldly is what is meant by the number, one hundred and eight, of the terms. One can confirm this by citing each and every thing and being and matter without any feeling of redundancy; for the number one hundred and eight of the terms does not limit the citer within them but opens him to the truth of all.

The former part of this double-natured term, "the term dharma," seems to find its detailed explanation in the so-called "sevenfold self-nature of being (bhava-svabhavah)":

Coming together (samadaya-svabhavah), beings (bhava-), characteristics


(laksana-), gross elements (mahabhuta-), causes (hetu-), co-operating causes (pratyaya-), and being brought about (nispatti-).

The latter part, "not being the term dharma," can be paraphrased as the "sevenfold ultimate way of being (paramarthah):

The ultimate way of being relating to mind (citta-gocarah), relating to knowledge (jnana-), relating to insight into voidness (prajna-), relating to the twofold view (drstidvaya-), relating to the twofold view surpassed (drstidvayatikranta-), relating to surpassing the bodhisattva-stages (sutabhumy-atikramana-), and relating to the tathagata's own attainment (tathagatasya sva-prapta-).

(Cf. NII Section Two (2): The Seven Characters Each of the Ordinary Being and the Ultimate Way of Being)

According to the Buddha, all this is,

"the core of the self-nature of being and the ultimate way of being, for all the past, future, and present tathagatas, the most worthy, the rightly awakened ones."

This twofold "core" is the very standpoint of all the buddhas for establishing truths both worldly and supra-worldly:

"Fully furnished with die core both of the self-nature of being and the ultimate way of being, tathagatas establish truths of the world, those that surpass the world, and those that superatively surpass the world, with their noble wisdom-eyes penetrating into the characteristics specific and general."


In this scripture the Buddha, standing on this viewpoint, goes on criticizing wrong views, and presents his own. The self-nature of being does not stand as it is, but is always open to criticism by the ultimate way of being; meanwhile the ultimate way of being does not stand outside of the self-nature of being, but constitutes its original mode of being, its true self. Roughly speaking, this is considered to be the structure of the twofold "core," and also it explains what is meant by the one hundred and eight terms, represented by "the term dharma not being the term dharma."

According to the Buddha, further, truths both worldly and supra-worldly are established by the buddhas in the way theirs "may not be equal to the non-Buddhists' wrong views." (Ibid.) In that case, the Buddha answers the question, "How would one's view be equal to the non-Buddhists' wrong views?" as follows:

It is because one does not realize that while one sees one's own mind one falsely discriminates it as something external. Since the discerning faculties don't realize what is seen to be something external as nothing but one's own mind, ignorant, common people come to embrace twofold views, since for them being does not have the [above-mentioned] self-nature of being and the ultimate way of being (bhava-a-bhava-svabhdva-paramdrtha-).

B. "What is Seen as Something External is Nothing But One's


Own Mind"; and

C. "Entering Lanka":

This further clarifies the relation between the two kinds of core. One should not see beings just as anything external; they are nothing but one's own mind, or more precisely oneself as mind, seen as something external (svacitta-drsya-matram). This does not seem to mean that one affirms one's own being while rejecting the independence of external beings, considering them as reflections of one's own mind. What is meant here seems to be that having penetrating insight into beings leads one to the understanding that everything is free from the concepts of externality and internality. Everything, in the sense that they are beyond such concepts, is nothing but me, that is not an ordinary, individual "me." Readers need to seriously think of the reason why we see the expression, "what is seen to be something external is nothing but one's own mind," being repeated throughout the present scripture. The title of the sutra, "Lankavatara (entering or attaining to Lanka)," seems to have something to do with this basic way of thinking and practice, though the Gunabhadra version apparently does not offer much help in this regard.

The title seems to have derived from the legendary story


transmitted among the Theravada practitioners of the island, which is recorded in the Dipavamsa ("Royal Lineage of the Island Lanka"),* n compiled among them. According to the first two chapters of the story, Gotama Buddha came to the island three times: (1) nine months after his attaining Awakening, (2) five years after, and (3) eight years after.

(1) When he saw all the world with his fivefold eyes, he saw the island, where yaksas ("something quick," spiritual apparitions) and raksasas (when yaksas get angry they are said to be flesh-eating goblins or raksasas, "anything to be guarded against") were abiding and afflicting people, groaning loudly and sucking human blood; Gotama was afraid some strange teachings might flourish in that situation to worry people further. Using supernatural power, from India he came, expelled the terrible yaksas and furious raksasas by having them shift their dwelling place to a lonely island named Giri far out in the ocean. Then he returned to Urvela in Magadha (Chatper I).

(2) After he left, in the island mountain snakes and marine snakes struggled for sovereignty over the island, both being nagas with supernatural power, violent and cruel, arrogant and drunk with power, though different in their size. The situation worsened to the extent diat, wherever they went, everything got contaminated and burned out. Gotama, far away in India, felt he could not leave things as they were. Again he came to Lanka, which he had emptied of yalsas. He put bodi parties of snakes under his control, brought them into reconciliation, and returned to the Jeta forest (II).


(3) Three years later, king of the Lanka snakes, Maniakkhika, invited Gotama together with his five hundred disciples to the island in return for the Buddha's contribution as peace-maker. The party came flying from the Jeta forest. Gotama came to the Mahamegha forest, and predicted that in a future time the very Bodhi tree beside which he had attained buddhahood would be planted at the site in Lanka where Bodhi trees had grown for previous Buddhas (II).

There is no doubt that such stories were made on the basis of other, more historical stories, also recorded in the Dipavamsa, that transmission of the Buddha's teaching to the island had begun in the reign of King Devampiyatissa (B.C. 241-207). In response to the gift of treasures from the Lanka king Tissa, King Asoka sent messengers from India with gift and a message that he had taken refuge in the Buddha, Dhamma, and Sangha. Asoka's son, Mahinda, as an elder monk (thera), came (XI), and King Tissa of the island had a temple, Mahavihara, built in the suburbs of the capital, Anuradhapura, as the centre for practice and study for monks under Mahinda's guide (XIII, XIV). Mahinda had a messenger sent to King Asoka, and had him bring back part of the bones of Gotama Buddha, and had a dome (stupa) erected for the relics (XV). Mahinda's sister, Then Sanghamitta, also came to Lanka. She brought a branch of the Bodhi tree, and had it planted in the wood of Mahamegha,


where the Mahavihara was located (XVI). Mahinda died in B.C. 199 (XVQ), and Sanghamitta, the next year (Mahavamsa XX).

The legendary stories of Gotama's three visits to the island, however, seem to derive from one of the famous epics of India, Ramayana.*u Rama, the hero, came to attack raksasas of the island, killed Ravana, their chief, and returned home with his beloved wife Sita, who had been taken away to the island. Gotama was a hero equivalent to Rama, an avatar of Visnu, but, unlike Rama, Gotama killed none; he expelled evil spirits that had been devastating Lanka. The part of peace-maker played by Gotama for two snake-groups also seems to have its forerunner in the Ramayana; where two groups of monkeys followed Rama and helped him in his attack against Lanka-demons, since he had worked as peace-maker for them when they had been in mutual conflicts in the continent. Thus we know that the Theravada document, the Dipavamsa, invented the idea, the Buddha's entering Lanka, on the basis of history and legends. But we need to consider what was meant by the mahayanists' use of the title, "Entering Lanka," for their scripture.

The Lankavatara sutram in the Gunabhadra version begins with the description of the spot where the Buddha, the bhiksu-samgha, and bodhisattvas met, and how one bodhisattva named Mahamati


from among other bodhisattvas of mahamati (see note below) stood up and asked the Buddha for teaching.

On one occasion die Buddha stayed for a while in the town of Lanka on the mountain top, on the coast of the southern sea ...

Now the Awakening great being, Mahamati, who was together with [other] Awakening beings of mahamati (i.e., of great wisdom), an attendant in every Buddhaland, through the Buddha's influence stood up from his seat, ...

As is known from the note attached to it, the word "mahamati," used here both as a common noun and a proper noun, reveals a close connection between the Lankavatara sutram and the Dipavamsa. In the latter, the word was used only as a common noun, to show a deep respect when excellent mendicants were referred to:

(1) "That mahamati Upali, after naming a learned man named Thera Dasaka as a responsible person for Vinaya, passed away." (V. 90)

(2) "A Greek mahamati named Thera Dhammarakkhita, by introducing the Aggikkhandhopamasutta ("Scripture on a Simile of Big Fire"), led natives of Aparanta to the Buddha's teaching." (VIII. 7)

(3) "In former days mendicants of mahamati had transmitted the Pali Tipitaka and their commentaries orally." (XX. 20)

According to the same Lanka record, in this land of Theravada practitioners, another Buddhist centre named Abhayagiri-vihara was built by King Vattagamini Abhaya (B.C. 29-17), at the site of a Jaina


temple, in the town of Anuradhapura (XIX). The Dipavamsa closes its record of the royal lineage of Lanka with the description of how King Mahasena (A.D. 334-361) died under the influence of "the shameless, evil monks" of this Abhayagiri-vihara, and had to receive punishment for his life-time evil conducts, and warns readers to avoid such evil people as beings like snakes (XXII).

The Dipavamsa is considered to have been compiled during the period between A.D. 361, the year when King Mahasena died, and 429, the year when Buddhaghosa, who had come from India to abide near the Mahavihara in Anuradhapura, began writing Samanta-pasadika, a commentary on the Vinaya, basing his description of history for its preface on the Dipavaima.*14 The other history-book on Lanka, the Mahavamsa, is said to have been compiled around the middle or end of the fifth century with the purpose of refining and supplementing the Dipavamsa's expression up to XXXVTI. 50. According to this newer record, at the Abhayagiri-vihara mahayana studies and practice were conducted in a critical manner against the Theravada way of thinking represented by the practitioners at the Mahavihara, and the latter hated the former so much that they tried removing the mahayanists by means of the political power. In the Mahavamsa mahayana was called "Vetulya (=vaipulya)-vada":


(1) King Voharikatissa (A.D. 269-291): "Suppressing the Vetulya-doctrine, and keeping heretics in check by his minister Kapila, he made die true doctrine to shine forth in glory. "(XXXVI. 41)

(2) King Gothabhaya (=Meghavannabhaya, A.D. 309-322): "He seized bhikkhus dwelling in die Abhayagiri (vinara), sixty in number, who had turned to the Vetulya-doctrine and were like a thorn in the doctrine of the Buddha, and when he had excommunicated them, he banished them to the further coast."(XXXVI. 111,112)

In A.D. 410-411, Faxian, a Chinese monk, who had left China in 399 to seek for Vinaya texts, and who had made long travels by land, stayed in Lanka for the two years. Then he returned by sea with Buddhist texts on board a ship, and back home wrote down a detailed record of the travels for himself (Taisho Tripitaka vol. 51, no. 2085). According to this record, in Anuradhapura five thousand monks were abiding in the Abhayagiri-vihara, and for ninety days annually the Buddha's teeth were carried from the Buddhadanta- vihara to the Abhayagiri-vihara to receive people's offerings. In the Mahavihara three thousand monks were abiding. The present English translator surmises that the Lankavatara sutram was compiled at the Abhayagiri-vihara some time between A.D. 411, when Faxian left Lanka for home, and 435, when Gunabhadra reached China, possibly bringing its Sanskrit text from Lanka.


In the Lankavatara sutram in the Gunabhadra version there is another place where the name Lanka is cited in connection with the Buddha's teaching (NII Section Three: The Seven Discerning Faculties and the Subtle Root Discerning-Faculty). Mahamati asks the Buddha as follows:

For those who abide in the land of Lanka on the Malaya mountains in the sea, headed by Awakening beings, (N44) please declare what has been celebrated in song by tathagatas, the original way of being of the root discerning-faculty compared to the ocean of ocean waves (udadhi-taramga-alayavijnana-gocaram), which is the Awakened self itself (dharmakayam).

Here the ocean-waves are compared to the discerning faculties, while the ocean is to the root discerning-faculty, and their original mode of being is to the Awakened self. A particular geographical place Lanka is referred to, so that its ocean-waves and ocean may be compared to discerning faculties and the root discerning-faculty, and their original mode of being to the Awakened self. Here Lanka is nothing but the here-and-now of the islanders, and it is when they realize the truth of their here-and-now that the Buddha enters Lanka to teach the Awakened truth. What they see as Lanka is nothing but their own mind appearing as such. That seems to be the reason the compilers of this mahayana sutra named it as "Entering Lanka." The geographical name can be replaced by any other name


according to where one abides in, for entering it means "entering and really penetrating one's here-and-now." Possibly that is why the Gunabhadra version makes no other explanation about the title except for the ocean-waves and so on.

However, not long after the compilation of the original text, because of the strong inclination towards diffusion of discrimination among those who inherited it, people came to feel like having more explanations about the title. They invented a new story as an introductory chapter, though it is not certain whether the compilers were still abiding in Lanka or not, when one takes into consideration the terrible condition of the text-preservation seen in the Bodhiruci Chinese version (tr. A.D. 513) and the extant Sanskrit text which shares almost the same condition with the Bodhiruci version.

In the introductory chapter the yaksa-king of the island, Ravana ("One Who Cries Loud") came out from nowhere to the shore to receive the Buddha to the mountain-top town Lanka, and, helped by Mahamati, asked about "dharma" and "adharma." Certainly the new-compilers did not like the way yaksa-raksasas of Lanka were treated both in the Indian epic and the Theravada records. In Chapter I of the extant Sanskrit text Ravana proclaims that they have had buddhas teach them, and that their sons and daughters


need to be taught in the same way. Indubitably this idea could never have come from those living in the tradition of the Ramayana or of the Theravada; it must have come from the mahayanists of the Abhayagiri-vihara, who had to criticize those Theravadins who rejected the mahayana thought and practice.

D. "The Tathagatagarbha-Alayavijnana":

This subject is treated in two places:

38. Gil, NII Section Fourteen: the Tathagatagarbha Thought;

112. GIV, NVI: Section One:

The Womb for Tathagatas (tathagatagarbha)

Free From Transmigration

Transmigrates As the Root Discerning-Faculty (alayavijnana).

In the first place of reference, the term "tathagatagarbha" is presented in two ways: (a) Mahamati asks if it does not mean the same as the nonBuddhists' "atman" theory; In this case the term is understood to mean "the tathagata in the womb (tathagato garbha- avasthitah)." (b) The Buddha does not accept this view; instead, he offers his understanding of it as "the selfless womb for tathagatas (tathagata-nairatmya-garbhah)." In both understandings the womb (garbha) stands for humanity. While in the former humanity conceals transcendence (tathagata), in the latter, humanity in its selflessness


or void, that is, "depth humanity" (a term used in his personal talk by Dr. Shin'ichi Hisamatsu, 1889-1980) works as the source of transcendence (tathagata-activities).

In the second place of reference the above (b) understanding is made clearer. Here the two concepts "tathagatagarbha" and "alayavijnana" constituting one term, instead of being joined and identified, is seen in the expression "the womb for tathagatas—the root discerning-faculty (tathagatagarbha-alayavijnanah)." This reminds one of the typical expression of the one hundred and eight terms: "the term dharma not being the term dharma (dharmapadam-adharmapadam)." The present term may be considered to represent "the term alayavijnana not being the term alayavijnana." In a more modern way of expression the two terms will be related as: "the alayavijnana as the unawakened tathagatagarbha" and "the tathagatagarbha as the awakened alayavijnana."

What is meant in all this is that the present scripture shows its criticism of the Samkhya view on emancipation, in which separation of the Purusa from the Prakrti through the latter's transformation is asserted, whereas the Lankavatara sutram advocates realization by the alayavijnana, as it were, of how it, as it is, ultimately fails, and having its turning over, and returning to its original mode of being,


i.e., the tathagatagarbha. This relation of the two concepts finds its practical expression in a Chan/Zen maxim, "Directly pointing to the mind, Having it see its original nature and attain buddhahood."

E. "The Five Grave Sins":

86. GUI, NIII Section Two: Five Grave Sins: (1) The Internal Ones

87.         NIII Section Two: (2) The External Five Grave Sins

The title "the five grave sins" is taken up in the context where the essential nature of the buddha is discussed, and the apparently strange interpretation, the internal five grave sins, serves as an explanation of the nature of the Awakened ones. Their so-called internal interpretation is not peculiar to this scripture; it is shared by other mahayana texts as well:

Manjusri-parivarta-aparaparyaya Saptaiatika prajnaparamita sutram (Buddhist Sanskrit Texts no. 17, p. 3499-15);

Avaivartacakra sutram (Taisho 9, no. 266, Dharrnaraksa tr., pp. 214c~215a; Tib. Trip. Peking vol. 36, no. 906 p. 118, 290ab);

Supratisthitamati-devaputra-prasna-sutram (Taisho 11, no. 310, p. 589a; Tib. Trip., Peking vol. 24, no. 36, p. 142, 347b~348a).

Buddhasamgiti sutram (Taisho 17, no. 810, 768c; Tib. T. Derge vol. 13, no. 228, 445; Peking vol. 35, no. 895, 234a).

In the last scripture accomplishing the five grave sins is said to be the reason why a practitioner leaves household to acquire


monkhood. The buddha, in whose presence this was uttered by a woman practitioner answering Manjusri's question, is called Jug-sred-kyi-rgyal-po" (Avatararuciraja?), and %%% ("Buddha as King of All Deities") in Chinese. These names of the buddha as well as the structure of the whole scripture suggest that he seems to be the buddha "Devaraja," a future buddha Devadatta was predicted to be by Sakyamuni in the Saddharma-pundarika sutram Chapter 11.

Among traditional buddhists Devadatta was an abominable person who had committed the latter three of the five grave sins, causing death of a woman arhat, splitting the buddha Sakyamuni's sangha, and causing blood to be shed from the buddha's body (See the chapter "Sanghabhedavastu" of the Mulasarvastivadavinayavastu II, Buddhist Skt. Texts no. 16, p. 188). The main reason Devadatta was hated was that he had started his own sangha independent of Sakyamuni's. Even among scholars of mahayana Buddhism there have been some not willing to accept the passages as authentic of the Saddhamia-pundarika sutram Chapter 11 that mention Sakyamuni's prediction of how Devadatta will attain buddhahood in the future. But we need to know that the so-called internal interpretation of the five grave sins was prevalent among mahayana Buddhist scriptures, including the Saddharma-pundarika sutram. The


present translator is of the opinion that the five grave sins in their internal interpretation represent the very mahayana view, instead of being anything accidental. It is well known that the first two of the five grave sins, killing mother and father, have already been given the so-called internal interpretation in the ancient text, Dhammapada (nos. 294 and 295).

F. "Meat-Eating Be Stopped":

120. GIV, NVIII: "Meat-Eating (mamsa-bhaksanam)"

The chapter "Sanghabhedavastu" of the Mulasarvastivadavinayavastu II mentions that Devadatta prohibited mendicants from eating meat so as not to kill animals, while criticizing Gautama for eating meat (Ibid., p. 19020~21). This means that the Lankavatara sutram shares the same view on meat-eating with Devadatta's sangha. Meanwhile, Sakyamuni's prediction in the Saddharma-pundarika sutram that Devadatta will become a buddha in the next life ought to be considered consistently in connection with the typically mahayana concept of internal five grave sins. This leads us to think that Devadatta's sangha had had a typically mahayana character as far as the latter's monastic discipline was concerned.

G. "The Non-Approach to the Truth (agotram)":

26. GI, NII Section Eight (5): The Non-Approach


The Buddha in the present scripture cites five ways of approach to the true mode of being (abhisamaya-gotrani): approach through the follower-disciples' vehicle (sravaka-yana); through the solitary attainers' vehicle (pratyekabuddha-yana); through the tathagatas' vehicle (tathagata-yana); through the unfixed, either-one way of approach (aniyata-ekatara-gotram); and through the fifth, non-approach (agotram). Among these, the tathagata's vehicle represents the mahayana or bodhisattva-vehicle, as far as the monastic discipline is concerned. The prohibition of meat-eating will be applied to this approach. The mahayana vehicle has another aspect, freedom from monasticism as well as from secularism. That may be represented by the fifth approach, the non-approach (a "special" one, according to the Gunabhadra version, but "the non-approach" is adopted for the present translation, as already mentioned above).

This last approach is explained to mean "the nature of not desiring emancipation of those who desire it (icchantikanam anicchantikata moksam)." And practitioners of this strange nature are of two kinds: One of them is those traditional buddhists who are criticized to "abandon all the roots of virtue" by blaming the mahayana sutras and vinayas or monastic rules for not leading to emancipation; they are criticized for the reason that they seek for their own salvation


and have no concern with others' emancipation. The other kind is those mahayanists who realize their vows since the beginningless time for sentient beings. They are bodhisattvas who desire emancipation of all beings but who do not desire it for themselves; they are not to attain perfect nirvana to the end, since they realize that all that have their own characteristics have originally been in perfect nirvana. They are called bodhisattva-icchantikas. A Japanese Zen monk in the Edo period, Hakuin Ekaku (1685-1768) used a Japanese rendering of this term, [Is-]Sendai, for his pen-name.*15 Since the term in its second of the two senses given above is peculiar to the present scripture, Hakuin must have meant the bodhisattva-icchantika without doubt. In those feudal days Hakuin continued to criticize those in power who neglected ordinary people's sufferings. While remaining a strict monastic practitioner, he was deeply concerned with emancipation of people both worldly and unworldly. When the non-approach of the bodhisattva-icchantikas is considered in connection with the internal five grave sins, most probably we are having the mahayana way of being that goes beyond the monastic character of the tathagata vehicle referred to above. The Vimalakirti-nirdesa sutram, among mahayana scriptures, seems to depict this mode of being. Students of truth of modern times


also may find something deeply democratic to learn from this mode of being that is free from both worldliness and unwordliness.

H. "An Appellation of the Tathagata (tathagatasya adhivacanam)":

107. GIV, NTH Section Nineteen (1):

Not-Arising Not-Perishing

as Another Name for the Tathagata

This subject is also remarkably characteristic of the present scripture, and is expressive of the basic character of Buddhism itself. The Buddha of this scripture asserts that all that have their own characteristics are free from perishing and arising, and that this very freedom from perishing and arising, the non-arising non-perishing of all, is the tathagata's appellation, or another name (parydya- vacarmm). He also asserts that the tathagata reaches people's hearing range through innumerable synonymous names without their awareness, like the moon reflected in water, which is neither in the water nor out of it. As synonyms of the tathagata he cites some such names as: tathagata, buddha, rsi, visnu, Isvara, pradhana, kapila, soma, bhaskara, varuna, sunyata, bhutata, satyata, dharmasvabhava, and so on. He explains that the manyness of the names does not indicate so many beings or the non-existence of the tathagata. All these are the tathagata's other names insofar as through them they


realize the non-arising non-perishing nature of all that have their characteristics. But people, fallen into the two extremes of alternatives, cling to letters and sounds, don't well discriminate the tathagata's name (abhinna-samjnah). Not well versed in their own directive principles (svanayam), people consider the non-arising non-perishing of all to be mere nonbeing (abhavam). They are not confident enough to realize the culminating point to which their own directive principles return (na svanaya-pratyavasthana-nisthan adhimoksanti), for they pursue teachings only through letters and sounds. They believe that there is no meaning apart from letters and sounds, and that what matters is sounds alone. Thus they don't penetrate the original nature of sounds. They don't realize that sounds arise and perish while meaning is free from arising and perishing.

The Buddha says one should be confident in meaning instead of letters since the Awakened truth is free from letters. One should not cling to sounds and letters. Meaning is compared to the moon in the sky while sounds and letters to the finger-tip which points to the moon. The ignorant take the non-arising non-perishing of all without cooking, that is to say, literally, thus stopping with the finger-tip and suffering from ill-digestion, for sounds and letters represent discriminative thought, clinging to which results in one's


suffering from birth-death transmigration. Meaning as truth (tattvdrthah) is gained in the presence of the deeply erudite (bahusrutanam sakaiallabhyate), for deep erudition means being versed in meaning (artha-kauialyam) instead of in sounds and letters (na ruta-kauialyam). The Buddha of this scripture urges us to penetrate into the meaning, as the Awakened truth, of the term: "The Non-arising Non-perishing nature of all," instead of stopping with its sounds and letters.

II. The Tathagatagarbha Thought Expressed by Other Texts

      in India and China (1) The Tathagatagarbha sutram,

       the Ratnagotravibhaga, and the Foxing-lung

In my footnote to Mahamai's question on the meaning of the tathagatagarbha thought in the following translation (Gil, NII Section 14, N77) I referred to the room left for a non-Buddhistic interpretation in the manner in which the Tathagatagarbha sutram presented nine illustrations and which was further extended by the compilers of a treatise named Ratnagotravibhaga Mahayanottaratantra-iastram ("A Treatise as A Section on the Lineage of Treasure, the Ultimate Mahayana Doctrine;" hereafter, the Ratnagotravibhaga)*^ What I mean by the room for a non-Buddhistic interpretation is as follows:

The nine illustrations of the Tathagatagarbha sutram are: (1) a


withered lotus calyx or padma-garbha which hides a shining buddha image in it, (2) the beehive which stores honey and which is guarded by bees, (3) husks which cover and protect grains in them, (4) dirt and mud which protect a mass of gold smeared by them, (5) the ground which hides a treasure below a poor man's shed without the latter's knowledge, (6) fruits of trees which contain seeds, (7) stinking, dirty clothes in which a jewel-made buddha-image is wrapped, (8) a poor woman pregnant, without her knowledge, of a royal inheritor, and (9) the mould of clay which holds a golden buddha-image inside. Of these illustrations the containers or covers which keep human attention away from the contents stand for the human beings in despair, unawakened to their true way of being represented by the precious contents. The purpose of the exposition by the use of these illustrations is considered to lie in having people penetrate their desparate way of being to cease holding to it as anything final, and in having them realize the ultimate Voidness that is the source of true activities. What matters here is the true darkness of the covering, instead of the deceptive brightness of the contents. In other words, our original Awakening takes the form of Unawakening, instead of anything brilliant. It is only through the realization of Unawakening that true Awakening presents itself. That seems to be


what is meant by the teaching of the Womb for tathagatas.

Meanwhile, the teachings of the two texts, the Tathagatagarbha sutram and the Ratnagotravibhaga, apparently place emphasis on the brilliancy of the contents at the sacrifice of the outer darkness: a buddha in a withered lotus flower, honey where bees are, grains in husks, gold in excrement, a treasure underground, seeds etc. in tree-fruits, a buddha image wrapped in dirty clothes, a royal prince in a poor woman's womb, and a golden image in a clay-mould. In the case of the Ratnagotravibhaga it is not very clear whether the Buddha who says he observes the presence of a buddha inside the darkness of humanity insists he would liberate the buddha by cutting through hindrances for the sake of humans or would have the humans under his guidance do the liberation. The latter is evidently the case, however, with the Tathagatagarbha sutram, where practitioners are repeatedly encouraged to follow the Buddha's advice to proceed in practice. On the contrary, the Ratnagotravibhaga gives an impression as if it kept on depreciating the darkness of coverings or the womb (garbha), by focusing on the brightness of the contents or the buddha nature (tathagata-dhatu. Cf. I., verse 122). When it speaks of the presence of the buddha nature which is compared to a royal successor as the future protector in a helpless woman's own being which


stands for human defilements (Ibid.), it sounds very weak because it does not seem to pay attention to the positive role of the womb in defilement.

The Foxing-lung, a "Treatise on the Buddha-Nature," was introduced as one of his unique "translations"to the people of China by Paramartha (A.D. 499~569), an excellent Buddhist thinker-translator from India, who came to China in 546. The well-known Dasheng-Qixing-lun, a "Treatise on the Mahayana Awakening of Faith," is another such "translation." By applying the term "translation" to such works he seems to have attempted to introduce authentic Buddhist views on subjects concerned, as he observed that inauthentic views on Buddhism were much in vogue. As for the Foxing-lung, it seems to have been composed for the purpose of introducing the subjects of the Ratnagotravibhaga, which had been introduced to China through its Chinese version by Ratnamati (tr. around A.D. 511 ~515), in a critical way in the context of the mahayana thought as he understood it ought to be. It consists of accurate introduction of important themes of the treatise in a systematic and supplementary way with the intention of clarifying their meaning originally intended. Although he attributed its authorship to Vasubandhu, when compared with the extant text of the


Ratnagotravibhaga, it is known to be his own compilation, serving as a critical commentary on the latter. He must have had some reason for the authorship-attribution.

In this commentary-like treatise Foxing-lung, Paramartha, concerning the eighth illustration on a poor woman pregnant of a royal prince, introduces the Ratnagotravibhaga s explanation that it was meant for manifesting the nature of the self-afflicting passions (klesa) at the first through the seventh bodhisattva stage (I v. 141ab). Then he adds, saying that just as the misery and helplessness of the woman would never defile the future world-king in her womb, the self-afflicting passions for the first through the seventh bodhisattva stage practitioners rather have three virtues. They are: (1) Being free from defilement since they are nurtured by wisdom and compassion; (2) Being free from faults since they never harm self or others; and (3) Being full of immeasurable merit since they bring to maturity both the buddha-dharma and sentient beings. Indeed, he says, if the self-afflicting passions should grow further passions, they would effect ordinary, ignorant beings, far form maturing the buddha-dharma. Meanwhile, if the self-afflicting passions should cut themselves off, that would mean effecting sravakas and pratyekabuddhas, far from maturing sentient beings [for mahayana]


(Taisho 31. no. 1610, p. 808a). There is no explicit explanation of such a role of suffering in the illustration of a helpeless, pregnant woman in the Ratnagotravibhaga.*17

The Foxing-lung cites three meanings of the term "tathagatagarbha" (Taisho 31, 795c~796a). They are: (1) All sentient beings being taken in by the tathagata without exception (tathagata-grhltah sarva-sattvah. Cf. I below v. 147); (2) The tathagata being hidden from all sentient beings (tathagato gudhah sarva-sattvanam. Cf. I below v. 148); and (3) The tathagata-nature being what is original to [all sentient beings and] tathagatas (tathagata-dhatur garbhah [sarva-sattvanam ca] tathagatanam [ca]. Cf. I below v. 152). These are how Paramartha explained the term according to the corresponding expressions in the Ratnagotravibhaga.*18 Certainly they show the meaning of the term "tathagatagarbha," respectively in their own way. Nevertheless, it is not clear how they are mutually related to constitute a concrete, religious meaning. They sound too abstract as they are to respond to our urgent religious quest, though we know that Chan and Zen Buddhists have taken them up in the religious context of their own to explain their religious thoughts.

By the way, none of the three explanations suggests me the strange interpretation, "all sentient beings being the tathagata-


embryos," which is very popular among modern Buddhist scholars, and which has largely contributed to the prevalent criticism against the tathagatagarbha thought as a non-Buddhistic, atman-doctrine.

(2) The Srimaladevi-simhanada-sutram, the Ratnagotravibhaga,

       and the Dasheng-Qixing-lun

For the purpose of clarifying the meaning of the term "tathagatagarbha," the Ratnagotravibhaga quotes many passages from the Srimaladevi-simhanada-sutram (hereafter, Srimala-sutram or just Srimala). Those Srimala passages sound more religious than those from the Tathagatagarbha sutram. Those quotations apparently cover almost all the important expressions given in the Srimala-sutram concerning the tathagatagarbha thought. Since the Srimala-sutram is extant only in Chinese and Tibetan versions, the Sanskrit quotations in the Ratnagotravibhaga are really helpful for the understanding of the text. Now we miss one statement, which seems to be decisive for the understanding of the tathagatagarbha thought, in those Sanskrit quotations. Let me explain it. What follows is my English rendering of part of Lady Srimala's words addressed to the Buddha, in Gunabhadra's Chinese version of the Srimala-sutram in the Taisho Tripitaka vol. 12, no. 353,11. 5-11, p. 252b:

The Most Revered One, die Womb for tathagatas is what the life-death


circulation (samsara) rests on. By virtue of the Womb for tathagatas, as the most Revered one has expounded and presented it to us, die life-death circulation has no starting point (purvakoti). The Most Revered One, the utterance that as the Womb for tathagatas exists, the life-death circulation exists is appropriate. The Most Revered One, by die life-death circulation I mean that as soon as the faculties that had been received perish, one would cling to the faculties not yet received as one's own. The Most Revered One, the names of the two concepts, dying and being born, are synonyms of the Womb for tathagatas.

(Tibetan: Tsukinowa 144, 9-146, 2; Peking 281a, 1 -3. The underlined part shows the Sanskrit quotation in the Ratnagotravibhaga: Z. Nakamura 143, 5-6)

Dying and getting born—diese are worldly uses of expression. Dying means faculties getting suppressed. Getting born means new faculties arising. In the Womb for tadiagatas is no getting born, no dying, no perishing, and no arising. The Womb for tadiagatas surpasses the sphere of the composite characteristics. The Womb for tathagatas is permanent, whole, and constant. (Tib.: Tsukinowa 146, 2-12; Peking 281a, 3 -6. Sanskrit quotation: Z. Nakamura 89, 11-17)

As is seen in my translation, when it quoted the latter part, the Ratnagotravibhaga did not include the final statement in the former part which preceded it:

The Most Revered One, the names of the two concepts, dying and being born, are synonyms of die Womb for tadiagatas.


But what was missed in quotation was essential, for, together with what follows, it conveys the meaning of the term "tathagatagarbha." The text mentions: What we call our birth and death, or our birth-death being, is nothing but what is called the tathagatagarbha, and this tathagata-garbha has no birth and death in itself; it goes beyond that. The whole statement well characterizes the tathagata-garbha. It can be summarized in one sentence: We, grasping ourselves as birth-death existences, do not realize that we are free from birth and death. This is no mere negative expression; it implies that we, birth-death existences as we are, are free from birth and death. This shows the direct connection between our Original Awakening and Awakening Attained, and clarifies the ground for the so-called "immediate Attaining of Awakening." But in actualities they are separated by Unawakening, for our Original Awakening necessarily takes the form of Unawakening. Only through the realization of Unawakening does our Awakening is Attained. And the meaning of sentient beings being the tathagata-garbha, "the Womb for tathagatas," should first be located in this direct relationship between Original Awakening and Unawakening. Also the kind of Unawakening which keeps to itself is no real Unawakening. Real Unawakening should penetrate itself to Awakening Attained,


"the Awakened self of tathagatas" (tathagata-dharmakaya).

It was to introduce this point, whose heartfelt appreciation constitutes the real meaning of mahayana faith, to people of China that Paramartha wrote the Dasheng-Qixing-lun, as his "translation," critically quoting from various sources, including the Srimala-sutram, the Ratnagotravibhaga, and the Lankavatara-sutram. He attributed its authorship to Asvaghosa possibly because the latter described Siddhartha as a criticizer of the Samkhya teacher Arada Kalama in the Buddhacarita ("The Acts of the Awakened One," Chapter 12, vv. 23, 64, 71, and 73. In Chapter I, v. 11, the prince is said to have been born from the womb of Lady Maya as if descending from the sky.). We know that in the Lankavatara sutram the tathagata-garbha thought was taken up in connection with its criticism of the Samkhya dualism with prakrti and purusa. The reason Paramartha did not attribute the authorship of the Dasheng-Qixing-lun to Vasubandhu, as he did for the Foxing-lung, is, it seems, that he did not ultimately acknowledge Vasubandhu's as well as the latter's elder brother Asanga's, view. We notice that point in his Chinese translation (A.D. 563) of Vasubandhu's commentary on Asanga's vijnana-vada treatise Mahayana-samgraha, especially concerning the way the alayavijnana is explained to "abide together (sahasthana-; sahacarin)"


with a transcendent element, "the influence of sacred knowledge derived from the sphere of the Awakened truth (suvisuddha-dharmadhatu-nisyanda-smta-vasana)"(Taisho 31, no. 1595, III-l, 5: p. 173bc, 175a; Nagao Japanese translation I, pp. 45, 46, 146, 149). The two heterogeneous elements are said to cohabit like water and milk, while one is of the perishing nature and the other nonperishing. The situation is compared to that in which a miracle-working goose drinks up the milk which is mixed with water.

In the Dasheng-Qixing-lun Paramartha, quoting from the Srima/a-sutram, states,

Because of resting on the Womb for tathagatas there is the mind that is of the arising-perishing nature. One may call this what is nonarising-nonperishing abiding together with what is arising-perishing, neither as one nor as different. And that is what is called the alayavijnana. (Taisho 32, no. 1666, p. 576b)

In this case the "abiding together" does not mean a mixture of two heterogeneous elements which will result in the exclusion of one with the preservation of the other. Nor does it mean a mere identification of the two. It means one and the same thing has its one mode of being hide itself while the other mode of being appear. The nonarising-nonperishing mode of being hides itself while the


arising-perishing mode of being appears. That is made possible, Paramartha seems to mean, by the latter resting on the former. Resting-on means ignorance, for not resting means Awakening.

This being the "co-habiting" character of the alayavijnana, the latter also is said to be made up of two meanings which are neither one nor different, and one of which depends on the other. Its two meanings are: Original Awakening and Unawakening. Paramartha says:

Because of resting on Original Awakening there is Unawakening; because of Unawakening there is Awakening Attained. (Ibid., 576b)

No sentient beings are Awakened. From the beginning they have had discrimination continue and have never been free from it. I characterize this as beginningless ignorance. If they get free from discrimination, immediately they will know arising, abiding, changing, and perishing as the characteristics of the mind. Because of the equality of no-discrimination, actually there is no distinction of Awakening Attained. The four characteristics [arising, abiding, changing, and perishing] being simultaneous, none of them has separate beings; originally Awakening is equal and one. (Ibid., 576bc)

Breaking of the co-habiting character of the alayavijnana has


"Awakening Attained" or "the Awakened self (dharmakaya) of tathagatas" fulfil itself. That is another way of exposition by Paramartha of the Srimala's on the attainment of the Noble Truth, "Extinction of Suffering" or "nirvana," as the working source of tathagata-activities. In Paramartha we see a penetrating expounder of the tathagata-garbha thought.

III. The Historical Significance of the Lankavatara Sutram

The first two fascicles of the Gunabhadra Chinese version, which divide the whole thirty-six sections of the Sanskrit second chapter into two parts — thirteen and twenty-three sections in order, roughly speaking, deal with the bodhisattva path. The other two fascicles, excluding the final eighth chapter in the Sanskrit text on prohibiting meat-eating, deal with the buddhas' path. In other words:

Gunabhadra's third fascicle includes seventeen sections of the Sanskrit third chapter; while Gunabhadra's fourth fascicle includes remaining three sections of the Sanskrit third chapter, the Sanskrit fourth chapter (one section), the Sanskrit fifth chapter (one section), the Sanskrit sixth chapter (five sections), and the Sanskrit seventh chapter (one section), besides the eighth.

The Lankavatara sutram, when it takes up the bodhisattva path, does so in a very critical way, and offers its critical, bodhisattva


interpretation of the traditional sravaka path. Then, in its consideration of the buddhas' path, the scripture shows a very thoroughgoing way to criticize both traditional and non-Buddhist paths, to suggest what is truly mahayana. Later, Chinese Chan Buddhists seem to have learned much from this manner of its presenting the buddhas' path. Its criticism of the Samkhya thought is outstanding, as is seen in connection with its "tathagatagarbha-alayavijnana" concept. This critical attitude must have been remarkable even among Buddhists contemporary to the Lankavatara-compilers. But this does not seem to have attracted the attention of many present-day Buddhist scholars, who tend to rebuke the scripture for its mixture of the two opposite natures, "tathagatagarbha" and "alayavijnana." As far as my limited knowledge is concerned, no investigators have ever noticed the supremely Buddhistic significance of the Lankavatara-criticism of the Samkhya thought. It is in this regard that I make much of the way Paramartha chose Asvaghosa as the author of his writing, the Dasheng-Qixing-lun, because, so I assume as mentioned above, in the hymn-story of Sakyamuni's early life, Buddhacarita, Asvaghosa described how Siddhartha before attaining Buddhahood had criticized Samkhya thought.*19

Against the general disregard towards this scripture among modern


Buddhist scholars, in India there is no reason that we must believe it was neglected; to the contrary, it seems to have been paid due regard by specialists, including Paramatha (sixth century, as we saw above) and Candrakirti (seventh century, who quoted two verses: 48th and 51st, G,NIII, from the Lankavatara-sutram in his Prasannapada).

The Sanskrit eighth chapter, the final part of Gunabhadra's Fourth Fascicle, seems to have had a very important meaning for those mahayanists who had met with rebuke from traditional Buddhists against their mahayana way of thinking and practice. That, possibly, is the reason why the original form had to suffer so many rewritings as we see when we compare the Gunabhadra version to the Sanskrit text. We know that Santideva (seventh to eighth century) quoted nineteen of the whole twenty-four verses with a short passage from the Sanskrit eighth chapter, as we see in the extant text, in his Siksasamuccaya, in the course of his discussion on how to preserve practitioners' health. It must have been those practitioners who followed the practical instruction of the scripture faithfully that had inherited the Lankavatara-teaching with utmost honesty to themselves.

At the beginning of the fifth century in Lanka the Theravada school had Buddhaghosa, who had come from India to stay with


them, as a great spokesman of their tradition; his continued work for the Pali Tripitaka must have urged the then mahayanists to produce their own words to clarify what the mahayana meant. Or it may be quite the opposite: Because of the powerful contents of the Lankavatara-sutram, the Theravadins were afraid of the mahayana influence to become too strong, and invited Buddhaghosa to proclaim the authenticity of the traditional position.

Anyway, I believe that the influence of the mahayana movement in Lanka was very great, and that the mahayanists of the Abhayagiri-vihara must have had good communications with mahayana thinker-practitioners of the Indian continent. Considering such situations as well as the scriptural contents, I feel I can assert that the Lankavatara-sutram represents the most critical mahayana Buddhist thought ever attempted. I sincerely hope this impression of mine will be shared by other people through my poor efforts of presenting this translation to the contemporary world.


Notes to Preface and Introduction: --

1 Studies in the Lankavatara Sutra, Routledge & Kegan Paul Ltd., London 1930, and The Lankavatara Sutra, a Mahayana Text, R. & K., 1932.

2 %% [%%%%%%%], Taisho vol. 39, no. 1790, p. 430b.

3 Ibid.

4 The Lankavatara Mahayanasutram Rendered into Modern Japanese with Studies, by Tokiwa, Gishin, Studies of THE INTERNATIONAL RESEARCH INSTITUTE FOR ZEN BUDDHISM, vol. 2, published by the IRIZB, Hanazono University, Kyoto, Japan 1994.

This second volume of the Institute Studies Report consists of two books: (1) Japanese translation (277 pp.) and thirteen lectures on the scripture, rendered into Japanese from English, given at the University of Leiden, the Netherlands, Feb. through May 1993 (99 pp.); (2) Notes to the translation, in Japanese (131 pp.) and Sanskrit text expressions marked with numbers in the translation for reference sake (80 pp.).

5 Cf. Tokiwa, Gishin: "%%%%%%," Bulletin of the Faculty of Letters, Hanazono University, vol. 28, 1996, 18 pp.

6 Page 19, Takasaki, Jikido: Ryogakyo, Butten-koza, vol. 17, Daizo-shuppan, Tokyo 1980.

The author's lectures, based on his extensive knowledge and insight, covered almost one third of the whole text.

7 Cf. Tokiwa, Gishin: "The Historical Significance of the Opening Chapter Ravanadhyesana of the Lankavatara sutra, Journal of Indian and Buddhist Studies, vol. 40, no. 1, December 1991.

8 Compiled and published by Professor Emeritus Yanagida, Seizan, of Kyoto University and former director of the International Research Institute for Zen Buddhism, Hanazono University, from four manuscripts, two of which had been discovered from among Dunhuang manuscripts by Dr. D.T. Suzuki and published together with another manuscript that had been preserved in Korea. Prof. Yanagida gave it the title, Daruma no Goroku ("Dharma's Words Recorded") with a subtitle


Ninyushigyo-ron ("On Two Entries and Four Practices": Zen no Goroku series vol. 1, Chikuma-shobo, Tokyo. 1969). The term %%%% is seen once in Section 13 (p. 80), and two times in Section 19 (p. 103) of this Yanagida-edition. No other translators than Gunabhadra used it for rendering the Sanskrit term "svacittadrsyamatram," which is specific to the Lankavatara sutram.

The "Dharma-master" (as is called by Tanlin in the record) for Tanlin, Huike, and others, was later given the name "Bodhidharma" by Daoxuan, compiler of the "Continued Biography of Eminent Buddhist Priests" in the seventh century. Daoxuan records how Bodhidharma advised Huike to do practice with the Gunabhadra version of the Lankavatara sutra as a good help.

["Bodhi"] the "Dharma-master"[, if I rename him after Daoxuan's naming,] from South India must have been well acquainted with the scripture before he came to China, and also with the two Chinese versions, Gunabhadra's and Bodhiruci's, after he came to China. That is why, it seems, he recommended the Gunabhadra version to his Chinese disciples for practice. As Dr. Suzuki writes in his Studies, and Dr. Takasaki in his Ryogakyo, the Gunabhadra version was well studied, and sometimes commented upon, by Chan/Zen monks. The Sixth Patriarch Huineng is also recorded to have quoted from the sutra in the Platform sutra, against the legend propagated by Shenhui, one of his disciples, that the master ignored it.

9 Besides this work, Dr. Takasaki had another precious work on this scripture: A Revised Edition of the Lankavatara-sutra, Ksanika-Parivarta, ed. by Jikido Takasaki, Tokyo 1981, as a research report General Studies (C), supported by the Government subsidy for aiding scientific researches for 1978~1980. For this project he collected seventeen manuscripts: seven kept in the Tokyo University Library, two in the University Library, Cambridge, one in possession of the Royal Asiatic Society, London, and seven photographed under the Nepal-German Manuscript Preservation Project, Kathmandu. This goes in the direction of restoring the extant Sanskrit text.

10 By the way this Tibetan version has a lacuna of about twenty lines of the Taisho Tripitaka Chinese text for the section on the "pancanantaryani (%%%%, five grave sins)." Taisho Tripitaka vol. 16, p. 498a ~b; Tibetan Tripitaka, Peking edition, vol. 29, no. 776, p. 107,263b; Kokan Shiren's division in his Butsugoshin-ron, no. 52; Gunabhadra III; Nanjio Sanskrit text III. Section 2.


The situation is the same with another Tibetan manuscript sNar-thang version. The lacuna is in vol. 51, no. 96, folio 382b6. Although the spellings of this version are more accurate than those of the Peking edition, the order of folios (298a~456b) in the microfilm is in complete confusion; it took me much time to arrange them in order according to the Peking edition. By the way, the Derge edition of this sutra shows confusion at the beginning, and then is superseded by another text, possibly the translation of the Sanskrit text. This seems to mean that the Tibetan translation of the Gunabhadra version has never been used except for the Peking edition.

11 Hermann Oldenberg: The Dipavamsa, Asian Educational Services, New Delhi, Reprint 1982. In the former half of the book the Pali text, and in the latter half its English translation, are given.

12 The Ramayana of Valmiki, tr. Makhan Lai Sen, Munshiram Manoharlal, New Delhi 1978.

13 The word "mahamati" was originally a common noun, used as a call of respect to monks and nuns and lay people according to the Lanka histories: Dipavamsa and Mahavamsa. This is also seen in the Gandavyilha. Perhaps in the present scripture, too, it is used in the sense: "a person of mahamati."

14 Cf. W. Geiger, tr.. The Mahavamsa, Pali Text Society, London 1912, reprinted 1980, Introduction pp. 1-5.

15 Hakuin used this name at the age of fifty-seven (A.D. 1741) when he wrote comments on the Hanshan Poety, and called the commentary Kanzan-shi Sendai-kimon, three fascicles. He also called his residence in the Shoin-ji temple "Kokurin-Sendaikutsu." Cf. Hakuin-osho Zenshu vol. 4, Tokyo 1934.

16 Edited and published in 1950 by E.H. Johnston & T. Chowdhury. The present translator uses the Johnston-Chowdhury text romanized and contrasted with Ratnamati's Chinese version by Professor Zuiryu Nakamura, Tokyo (1961), second printing 1971. In 1967 Prof. Z. Nakamura published a Tibetan version collated by him, contrasted with his Japanese translation of it as Zowa-taiyaku Kukyo-ichijo-Hosho-ron-Kenkyu, Tokyo.

The Johnston-Chowdhury Sanskrit text has an English translation by Professor Takasaki, Jikido, with detailed annotations: JIKIDO TAKASAKI: A STUDY ON


THE RATNAGOTRAVIBHAGA (UTTARATANTRA) Being a Treatise on the Tathagatagarbha Theory of Mahayana Buddhism, Roma, Instituto Italiano Per II Medio Ed Estremo Oriente 1966, Serie Orientale Roma XXXHI, pp. 439.

17 The verse 141, Chapter I, states:

"The impurities that belong to the seven stages are likened to those of the womb;

Wisdom free from discrimination is like the growing embryo freed from the womb."

18 For the third definition by Paramartha of the tathagata-garbha, the Ratnagotravibhagaspeaks about tathagata-dhatu being "original to all sentient beings (garbhah sarvasattvanam)." Meanwhile, Paramartha defines it as being "original to tathagatas (garbhas tathagatanam)." And this is exactly what the Lankavatara sutram gives as the explanation of the term "tathagata-garbha" ("garbhas tathagatanam"). Cf. GIV, N6, v. 1.

Of the correspondence of the three definitions of the term "tathagata-garbha" between those of the Ratnagotravibhaga and of Paramartha, Prof. Jikido Takasaki takes notes in his English translation of the Ratnagotravibhaga, pp. 286, 287, and 290.

19 See Canto XTI, especially verses 23, 64, 71, and 73 in The Buddhacarita: Or, Acts of the Buddha, tr. E.H. Johnston. Motilal Banardass, reprint: Delhi 1972.