14

Thoughts on the Apocryphon Theory

FUMIMASA-BUNGA FUKUI, a leading Japanese scholar on the Heart Sutra, wrote in his book that the Nattier theory “shook the Japanese academic world.”¹ He also remarked: “As the Prajna Heart Sutra is one of the most revered sutras in Japan, it would be a matter of grave concern if this were proved to be an apocyrphon produced in China.”²

Fukui outlines the Nattier theory, saying that her way of dividing the sutra into three sections is her own invention and does not accord with traditional scholarly divisions. He insists that it is a mistake to assume that the mantra was added to the main scriptural text.

According to Fukui, there has not been a single record or argument in Chinese history that suggests that the Heart Sutra is an apocryphal text. On the other hand, there are many documents acknowledging translation by Xuanzang, so Nattier would need to prove all of these documents to be apocrypha. He claims her paper is “driven by theory and not convincing.” Finally, he poses the question: Does Nattier’s hypothesis, based on a very small number of materials comparing the three texts — the Hridaya, the Sanskrit 25,000-line Prajna Paramita, and the Chinese Heart Sutra — prove that it is an apocryphal text?³

The U.S. translator and writer Red Pine also doubts Nattier’s theory:⁴

Despite the brilliance and depth of scholarship involved in Nattier’s presentation of this thesis, we are shown no proof that the Heart Sutra was originally composed or compiled in Chinese, that any part of the first half was extracted from the Large Sutra or any other Chinese text, or that the mantra was added later. Instead, we are asked to believe that this is what must have happened because certain lines in the two Chinese texts agree and those in the two corresponding extant Sanskrit texts don’t, and it should be the other way around, with the Sanskrit texts agreeing and the Chinese texts diverging in the usual course of translation.

My own solution to this apparent inconsistency is to assume that the lines in question in the Sanskrit texts of the Heart Sutra and the Large Sutra used by Kumarajiva and Xuanzang were identical. Thus, there was no need, nor any basis, for divergence in the Chinese. In fact, there is no evidence, only speculation, that the two Sanskrit texts used by Kumarajiva and Xuanzang differed at the time they made their renderings of this passage in these two sutras. The differences we see today in the two Sanskrit texts, I would suggest, were the result of subsequent corruption or simply reflect the existence of variant editions.

I myself believe that Nattier presents a revolutionary finding in the origination of the Heart Sutra by her careful and thorough argument. Because of her work, scholarly understanding about the history of the Heart Sutra will never be the same.

In summary, I agree with Nattier that the Heart Sutra was compiled in China. We can further argue that the so-called Kumarajiva Chinese translation is likely the original version of the sutra. Xuanzang seems to have played a key role in the formation of the Hridaya. It does not, however, mean that he literally back-translated the sutra from Chinese to Sanskrit, as Nattier indicates. A later record suggests that he received the Sanskrit version “from Avalokiteshvara.’’⁵ Thus it is possible to regard the emergence of the Hridaya as Xuanzang’s mystical experience during meditation. Below is the reasoning behind my thoughts on the Nattier theory.

For your reference, all the Hridaya and Heart Sutra texts, as well as earlier texts quoted by Nattier and myself, are presented in “Texts for Comparison” in appendix 1. (If the following arguments seem too cumbersome, you are welcome to skip to chapter 15, “Roles of Ancient Chinese Translators.”)

Comparing the two Sanskrit texts — the 25,000-line Prajna Paramita and the Hridaya — I endorse Nattier’s suggestion that the Hridaya did not come directly from the Sanskrit 25,000-line Prajna Paramita, or vice versa.

Her hypothesis is a new finding: that the core section of the so-called Kumarajiva translation of the Heart Sutra is a composition of extracts from Dazhidu Lun, a popular commentary containing parts of the Kumarajiva translation of the Maha Prajna Paramita Sutra. After examining the original Chinese texts included in the Taisho Canon, I verify Nattier’s theory in this regard. In other words, we can no longer call that version of the Heart Sutra Kumarajiva’s own translation, although its core section comes from the Maha Prajna Paramita Sutra translated by him through Dazhidu Lun.

For all scholars of the Heart Sutra who had believed this version to be Kumarajiva’s actual translation, this finding was a big blow. In addition, Nattier’s denial of the prior existence of the third-century Zhiqian translation pushes the formation of the Heart Sutra two to four centuries later.⁶ I myself was astounded by this new understanding.

We can now say that an unidentifiable person in China compiled the shorter text Heart Sutra. This compilation was an assemblage of three parts — an introductory part starting at “Avalokiteshvara,” a core section from the Dazhidu Lun that quotes Kumarajiva’s Maha Prajna Paramita Sutra, and the mantra.

Nattier points out that the so-called Xuanzang version of the Chinese Heart Sutra lacks two sections that are contained in both the Large Sutra and the so-called Kumarajiva version.⁷ This means that the so-called Kumarajiva version of the Chinese Heart Sutra is closer to the earlier Maha Prajna Paramita Sutra than the “so-called Xuanzang version.”⁸ Nattier’s observation leads me to believe that the so-called Kumarajiva version of the Heart Sutra preceded the Chinese Xuanzang version. I therefore propose to rename the so-called Kumarajiva version of the Chinese Heart Sutra the “α version” (alpha version).

In this way, the philological flow of the text can be presented this way:

Sanskrit 25,000-line Prajna Paramita → Kumarajiva’s Chinese Maha Prajna Paramita Sutra /Dazhidu Lun → α version

Which text, then, came next — the Hridaya or the Chinese version “attributed to” Xuanzang? In other words, which text — the α version or the Xuanzang version — became the basis for the Sanskrit version?

Let us compare these three texts: What is unique in the Sanskrit text and lacking in the Chinese versions is the dedication in the first line. What is lacking in the Sanskrit version and common in the Chinese texts is the title at the beginning and a phrase in the first line, “becoming free from suffering.” These discrepancies, however, do not provide clues to the chronological sequence of the texts.

What is lacking in the α version but existent in the Sanskrit version is the phrase “a great mantra.”⁹ The “Xuanzang” version reads “a great spirit mantra.” It seems there is a progression from the absence of the phrase to “a great mantra,” and then to “a great spirit mantra.”

What seems to be a deciding factor, however, in determining the order is the section right before “Form is no other than shunyata. Shunyata is no other than form.”¹⁰ This section is identical in Kumarajiva’s Maha Prajna Paramita Sutra, Dazhidu Lun, and the α version. It can be translated as follows: “Form is shunyata, therefore (it has) no mark of angst and destruction. Perception is shunyata, therefore there is no mark of perception. Feeling is shunyata, therefore there is no mark of knowing. Inclination is shunyata, therefore there is no mark of doing. Discernment is shunyata, therefore there is no mark of awakening. Therefore, Shariputra . . .”

In the corresponding section, the Sanskrit version says, “Form is shunyata. The very shunyata is form.” The so-called Xuanzang version has no part corresponding to either of these Chinese or Sanskrit passages. Do we not see a direction toward simplification from the earlier Chinese texts, then via the Sanskrit version, and finally up to the “Xuanzang” version? It would be odd if the α version were immediately followed by the “Xuanzang” version and then evolved to the Sanskrit version. It would, in effect, mean that these passages from the earlier Chinese versions were erased in the “Xuanzang” Chinese version and were brought back in the Sanskrit version. Thus I believe it is logical to assume that the α version became the basis for the Sanskrit version.

At this point I will diverge from Nattier’s argument that Xuanzang did not translate the Heart Sutra; I do not think she has presented sufficient evidence to discredit Xuanzang as translator of the Heart Sutra. Although Xuanzang’s biography does not mention his translation of the sutra, his successor Kuiji’s introduction to the Sino-Sanskrit Heart Sutra says, “This Sanskrit text of the Prajna Paramita Sutra has been transliterated (i.e., into Chinese) by the Tripitaka Master of Great Tang.”¹¹

It is true that Xuanzang’s most comprehensive Maha Prajna Paramita Sutra does not include his translation of the Heart Sutra. For centuries, this monumental text was considered to be of Indian origin. If it is indeed an Indian text, is it not natural that the Heart Sutra, which is of Chinese origin, is not found in it?

As Fukui points out in his review of Nattier’s paper, Xuanzang’s Heart Sutra was carved on the monument erected by Emperor Gao in 672 at the Gaofu Monastery, Chang’an¹² (fig. 8). Those who endorsed Xuanzang’s translation of the Heart Sutra — Emperor Gao, along with his officials whose names were carved on the monument — were all contemporaries of Xuanzang.

It is theoretically possible that the translation of the Heart Sutra “came to be attributed” to Xuanzang later, as Nattier suggests.¹³ Were this indeed the case, it must have been done by one or more of his students during the eight years between Xuanzang’s death in 664 and the dedication of the monument. I must say that this is highly unlikely.

I do believe that Xuanzang did not make a completely new translation of the Heart Sutra. However, in discussing whether he “translated” the sutra or not, we should not use the contemporary Western concept of “translation,” which calls for unique interpretations and innovative renderings throughout the text. We need to understand what “translation” meant for olden-time Chinese Buddhists, including Xuanzang himself. As I demonstrate in the chapter, “Roles of Ancient Chinese Translators,” those “translators” who followed in later times — Prajna and Liyan, Dharmachandra, and Zhihuilun — did not make more than “minor editorial changes,” in Nat tier’s terminology.

From this perspective, it seems impossible to deny Xuanzang’s role as the translator of the Heart Sutra. As a matter of simple logic, acknowledging Xuanzang’s translation presupposes the existence of a Sanskrit original. (Without an original text, no “translation” would have been possible.)

What, then, was the original Sanskrit text? Was the Hridaya not the original for Xuanzang’s Chinese Heart Sutra? As I suggested earlier, if we compare these two texts, it is clear that the former preceded the latter; it is therefore reasonable to assume that the Hridaya was not translated from Xuanzang’s Chinese version. In summary, I suggest that the logical chronological sequence of the texts would read this way:

Kumarajiva’s Maha Prajna Paramita Sutra/Dazhidu Lun → α version → Hridaya → Chinese Xuanzang version

Assuming that we can accept this sequence of texts, where did the Sanskrit Hridaya come from? Nattier says, “Xuanzang may have quietly re-translated the text back into Sanskrit.” By saying so, and with her use elsewhere in her article of the word “forgery,” is she hinting that Xuanzang might have deliberately deceived or at least misled people of his time and later generations by composing and presenting the Hridaya as an authentic scripture? Wondering about this point led me to ask if any reputable translator would “quietly re-translate” a text into its source language and translate it back to his or her language. If I reflect on myself, although I sit on the lowest (and perhaps the newest) seat among translators, is there any chance that I would “create” or “forge” an “original” text and translate it back to another language? Never.

What may lead us to a conjecture on the initial emergence of the Hridaya is a passage in the Sino-Sanskrit edition of the sutra by Amoghavajra (see Appendix 1): “Avalokiteshvara Bodhisattva had personally taught and bestowed this Sanskrit text to Tripitaka Dharma Master Xuanzang.” Although this manuscript was copied after Amoghavajra’s death — over one hundred years after Xuanzang’s death — it may give us a clue to the “transmission” of the Hridaya to Xuanzang.¹⁴ How?

Consider this: What do meditators do when they need to know something of deepest importance? Do they not meditate on it, or pray to a guardian deity for it? When Xuanzang wanted to have the original Hridaya and could not get it, would he not have meditated and prayed to Avalokiteshvara, his guardian deity, to bestow the sutra upon him? (We know Avalokiteshvara was his guardian from Huili’s account that Xuanzang invoked this deity when he was in danger in the Gobi Desert.) Did he not then in truth “receive” it from the bodhisattva? Is this not the implication of the passage in the Amoghavajra manuscript stating that the bodhisattva personally gave the sutra to him?

Meditation is, at times, described as a nondual experience, and some meditators say that giver and receiver become one in meditation. We could then say that the distinction between Avalokiteshvara and Xuanzang himself disappeared in his state of absorption, and thus Xuanzang received the Hridaya from the bodhisattva.

“Receiving” sutras from the Buddha or a bodhisattva, as Xuanzang’s followers — including Amoghavajra — claimed to do, is traditionally how the origins of Mahayana sutras were accounted for. From a contemporary academic perspective, the process of a Chinese monk conceiving a sutra during meditation may well be defined as “creation,” “composition,” “production,” “fabrication,” “back-translation,” “forgery,” or writing “an apocryphal text,” as Nattier suggests. However, one can equally well view the emergence of the Sanskrit version, according to Amoghavajra’s edition, as a result of the monk Xuanzang’s mystical experience.

These two views, scholarly and legendary, may not be mutually exclusive. Even if Xuanzang felt that he had been given the Sanskrit version by Avalokiteshvara, its contents were filtered through his knowledge of the earlier Chinese version. Of course, there would also have been an editing process, outside of nondual meditation.

The reader may get the impression that I have been making the case that Xuanzang did back-translate the Heart Sutra. However, I would not put it that way. In this particular case, the word back-translation can have the following implications:

1.  The normal flow of Buddhist texts at that time in China was from Sanskrit to Chinese. If the Chinese text comes first and then the Sanskrit version emerges from it, some scholars may say this is a back-translation.

2.  Xuanzang seemed to have had full knowledge of the α version, and at some point he had in his hand a Sanskrit version that corresponded to the α version. No one else seemed to have been responsible for the emergence of this new version. Again, some scholars may say Xuanzang must have back-translated the α version.

3.  If Xuanzang had had the intention and plan to translate the Chinese text into Sanskrit, I would say that he back-translated the Heart Sutra into Sanskrit. However, we cannot prove this. It is unlikely for an established and respected translator like Xuanzang to intentionally “create” an “original” text. For some scholars, the point may be too subtle, but in this case I personally would not use the term “back-translation.”

Suppose Xuanzang did feel that he had received the Hridaya from the bodhisattva, and then wrote it down (composed it), what was the probable date of his “transmission”? The reader may remember that according to a later account he completed the translation of the Hridaya into Chinese three days before the death of Emperor Tai. It might have been a moment when Xuanzang would have done anything to get the Sanskrit sutra for recitation, in order to acquire a higher magical power for healing the emperor than its earlier Chinese version carried. In this case, the sutra was conceived in China, so it deserves to be called an apocryphal text in the traditional Chinese sense. As Nattier reminds us, for a Buddhist scripture to be regarded as authentic in China, it had to be of Indian origin.¹⁵

The possibility that Xuanzang “received” the Hridaya in India cannot be entirely ruled out. Kuiji reported to the emperor that at Nalanda Xuanzang met again the sick monk who had given the sutra to him, whereupon the monk revealed that he was in fact Avalokiteshvara and blessed him for his safe journey home.¹⁶ Thus, Xuanzang may have had contact with Avalokiteshvara in India. He may even have consulted with Indian monks on vocabulary and grammar. This could explain why, in spite of Nattier’s observation that some lines in the Hridaya are not idiomatic, this text stood nearly uncorrected when it was expanded into the longer Hridaya.

Now we run into a peculiar dilemma: If Xuanzang had “received” the text in India, it would have to be seen as a scripture of Indian origin! Therefore, technically speaking, by the traditional Chinese standard that regards all sutras created in India as authentic, Xuanzang’s Sanskrit version could be accepted as an authentic scripture.