Articles by alphabetic order
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
 Ā Ī Ñ Ś Ū Ö Ō
1 2 3 4 5 6 7 8 9 0


Tibetan Texts of Garbhāvakrāntisūtra: Differences and Borrowings

From Tibetan Buddhist Encyclopedia
Revision as of 06:25, 7 October 2020 by VTao (talk | contribs) (Created page with " Robert KRITZER Tibetan Texts of Garbhāvakrāntisūtra: Differences and Borrowings The International Research Institute for Advanced Buddhology Soka University...")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search



Robert KRITZER Tibetan Texts of Garbhāvakrāntisūtra: Differences and Borrowings

The International Research Institute for Advanced Buddhology Soka University Tokyo・2012・Hachioji JAPAN


Garbhāvakrāntisūtra (“Scripture on the Entry into the Womb”) is a Buddhist text that describes the rebirth process. It begins by explaining the conditions necessary for conception and why a woman might fail to conceive. It then describes the oedipal fantasies of the transmigrating being and its state of mind as it enters its mother’s womb.

The heart of the sūtra is a very detailed, week-by-week account of the 38 weeks of gestation, which is followed by descriptions of both miscarriage and successful childbirth. The remainder of the text is devoted to the suffering that befalls the newborn, including infestation by worms, attack by demons, and affliction by many illnesses.

The longer versions of the sūtra also include a variety of additional topics, most notably an exposition of the four different ways of entering the womb. The first three types of garbhāvakrānti are those of virtuous people, some of whom are aware while entering, dwelling in, and emerging from the womb, some while entering and dwelling, and some only while entering. The fourth is that of the unvirtuous person, who loses his memory and awareness at the moment of death in the previous lifetime and is unaware at all three stages.

Garbhāvakrāntisūtra is significant for a number of reasons. Its description of gestation is the most detailed in Indian medical and religious literature. Moreover, it appears to be the first account divided into thirty-eight weeks; the nine-month account in the medical literature and in non-Buddhist religious texts is the norm. The sūtra seems to have both influenced and been influenced by the Indian medical tradition, with which it perhaps shared common sources (see Kritzer “Life”). The text also illustrates certain Buddhist attitudes toward the body: its detailed and pejorative descriptions of the mother’s vagina are representative of the more misogynistic strains of Buddhist asceticism (see Kritzer “Childbirth”).

The sūtra is also important as a source for descriptions of rebirth in Buddhist texts, including: meditation texts such as the Yogācārabhūmi of Saṃgharakṣa (Hsiu-hsing tao-ti ching 修行道地經, translated by Dharmarakṣa, and Tao-ti ching 道地經, translated by An Shih-kao); abhidharma texts such as the Vibhāṣā (A-p’i -ta-mo ta p’i-p’o-sha lun 阿毘 達磨大毘婆沙論), Pañcavastukavibhāṣāśāstra (Wu-shih p’i-p’o-sha lun 五事毘婆沙論), and Abhidharmakośabhāṣya (Api-ta-mo chü-she lun 阿毘達磨倶舍論); Yogācārabhūmi (Yü-ch’ieh  shih-ti  lun 瑜伽師地論) (see Yamabe; Kritzer “Garbhāvakrāntau”).

Furthermore, together with Vāgbhaṭa’s Aṣṭāṅgahṛdaya, Garbhāvakrāntisūtra is one of the most important Indian texts for Tibetan embryology (Garrett 23-31). In particular, it is the main source for Tibetan theories of winds as a primary factor in fetal 131 PDF Version: ARIRIAB XV (2012), 131-145development (Garrett 146-148). Unfortunately, there are only a few Sanskrit fragments of the sūtra available in the form of quotations in abhidharma and Yogācāra texts, but there are six translations, three in Chinese and three in Tibetan. De Jong has pointed out that the various translations must be carefully compared in order to clarify differences among the versions of the text, and this paper is a preliminary attempt to do just that.

Table 1—Translations of Garbhāvakrāntisūtra Title Translator Date Pao-t’ai  ching 胞胎經 (T. 317) Dharmarakṣa (Chu fa-hu 竺法護)


Fo  wei  a-nan  shuo  ch’u-t’ai  hui 佛爲阿難説 處胎會 (Ratnakūṭasūtra [T. 310, no. 13]) Bodhiruci (P’u-t’i-liu-chih 菩提流志) 703-713

Tshe  daṅ  ldan  pa  dga’  bo  la  mṅal  du  ’jug pa bstan pa (Translation of Ch’u-t’ai  hui) Tohoku 58

Chos grub (Chinese name Fa-ch’eng 法成) ninth century

Fo shuo  ju-t’ai-tsang  hui 佛説入胎藏會 (Ratnakūṭasūtra [T. 310, no. 14])(also found in the Mūlasarvāstivādavinayakṣudrakavastu [T. 1451: 251a14-262a19])

I-ching 義淨 710 Dga’  bo  la  mṅal  na  gnas pa  bstan pa (Translation of Ju-t’ai-tsang  hui) Tohoku 57 Unknown (perhaps Chos grub) ninth century

Mṅal  du  ’jug pa  źes  bya  ba’i  chos kyi  rnam graṅs (found in the Tibetan Mūlasarvāstivādavinayakṣudrakavastu) Tohoku 6

Vidyākaraprabha, Dharmaśribhadra, and Dpal ’byor ninth century

Pao-t’ai ching is by far the earliest version of the sūtra that we have. It is very different from the other versions, and the Chinese is difficult to read and probably corrupt in places. Although it is very important for the study of the sūtra, I will not discuss it in this paper.

Bodhiruci’s translation, Ch’u-t’ai hui, is said to be a translation of the same Sanskrit text that underlies Pao-t’ai ching.1 However, there are many differences between the two.

For example, while Pao-t’ai ching calls the interlocutor Nanda four times at the beginning of the text before changing to Ānanda, Ch’u-t’ai hui is consistent in using Ānanda throughout. Furthermore, Ch’u-t’ai hui includes a pair of verses in its account of Week 13 of gestation that are not found in Pao-t’ai ching. These and other differences

See, for example, K’ai-yüan  shih-chiao  lun 開元釋教錄 (T. 2154: 586c20-12).

PDF Version: ARIRIAB XV (2012), 131-145make it seem unlikely that the two are translations of the same text.2 In many respects Ch’u-t’ai hui is more similar to I-ching’s translation, Ju-t’ai ching.3 Chos grub or Fa-ch’eng, the famous Chinese translator based for much of his career in Tunhuang, translated Bodhiruci’s Chinese text into Tibetan (Tohoku 58).4 All the editions of this

translation contain the phrase mṅal du ’jug pa bstan pa (“the teaching about entering into the womb”) in their titles,5 but this phrase does not correspond to the Chinese title, which can be translated as “the chapter on dwelling in the womb.” Conversely, the title of the Tibetan translation (Tohoku 57) of I-ching’s Chinese translation (“the chapter on entering the womb”) generally

contains the element mṅal na gnas pa bstan pa (“the teaching on dwelling in the womb”).6 Thus, it seems likely that the Tibetan compilers of the Ratnakūṭa mistakenly switched the titles of the two texts. Because of this confusion of titles, I refer for the sake of clarity to the two Tibetan translations from Chinese by their Tohoku numbers: Tohoku 58 is the translation of Ch’u-t’ai hui, while Tohoku 57 is the translation of Ju-ta’i  ching.

I-ching’s translation of the sūtra was originally part of his translation of Mūlasarvāstivādavinaya. Bodhiruci also inserted it into his compilation of Ratnakūṭa, and these two versions are almost, but not completely, identical. A comparison of the Tibetan translation (Tohoku 57) with the two Chinese versions shows that Tohoku 57 was based on the version in Ratnakūṭa. According to

Ueyama Daishun, the Tibetan translation of Ju-t’ai ching also may well have been Chos grub’s work (“Dai Bankoku,” 178). It seems as though this is the version of the sūtra best known in Tibet. Finally, the Tibetan translation that is found in Vinayakṣudrakavastu (Tohoku 6) is the only one made directly from the Sanskrit. I

refer to this text, too, by its Tohoku number. Currently I am preparing a critical edition and annotated translation of this version, which is the longest extant version in either Tibetan or Chinese. The language of this translation is more difficult to

read than that of the other two, and it contains a considerable number of archaic words and words not found in any dictionaries. An examination of the extant texts suggests that there are two basic versions (or groups of versions) of the sūtra, a longer and a shorter one.

A detailed discussion of these differences will be included in the introduction to my forthcoming translation of the version of the sūtra in Tohoku 6.

Although Fo shuo ju-t’ai-tsang hui 佛説入胎藏會 is the title of the chapter of Ratnakūṭa that contains the sūtra, the sūtra proper is referred to in both versions of I-ching’s translation as Ju mu-t’ai ching 入母胎 (T. 1451: 253a21; T. 310 no. 14: 328a9) and Ju-t’ai ching 入胎經 (T. 1451: 260b29; T. 310 no. 14: 335b1). For convenience, I use the short title Ju-t’ai  ching in this article.

Tohoku 58 and Tohoku 57 are catalogued as separate texts in some Kanjurs but as chapters of Ratnakūṭa in others. For details, see bibliography.

The full titles are as follows: NFS dga’ bo mṅal du ’jug pa bstan pa źes bya ba theg pa chen po’i mdo; P ’phags pa tshe daṅ ldan pa / dga’ bo mṅal du ’jug pa bstan pa źes bya ba theg pa chen po’i mdo; D ’phags pa  tshe  daṅ  ldan  pa  dga’  bo  la  mṅal  du  ’jug pa  bstan pa  źes  bya ba  theg pa  chen  po’i  mdo.

The full titles are as follows: P ’phags pa dga’ bo mṅal na gnas pa bstan pa źes bya ba theg pa chen po’i mdo; D ’phags pa dga’ bo la mṅal na gnas pa bstan pa źes bya ba theg pa chen po’i mdo; N bcuṅ mo’u dga’ bo źes bya ba theg pa chen po’i mdo; S gcuṅ mo’u dga’ bo źes bya ba theg pa chen po’i mdo; F tshe ldan pa dga’  bo  mṅal  na  gnas pa  bstan pa  źes  bya ba  /  theg pa  chen po’i  mdo.

PDF Version: ARIRIAB XV (2012), 131-145Table 2—Versions of the sūtra

Short version: 1. Pao-t’ai  ching (about 4.6 Taishō pages) 2. Ch’u-t’ai  hui (about 4.3 Taishō pages) 2a. Mṅal  du  ’jug pa  bstan pa (Tohoku 58: about 11 folios [22 sides] in Derge edition) Long version: 1. Tohoku 6 (about 28.6 folios) 2. Ju-t’ai  ching (about 7.3 Taishō pages) 2a. Mṅal  na  gnas pa  bstan pa (Tohoku 57: about 21.3 folios)

As usual, we do not know the dates of the Sanskrit texts on which these translations are based. Pao-t’ai ching and Ch’u-t’ai hui are said to be based on the same Sanskrit text and seem to represent an earlier version or versions. Ju-t’ai ching and Tohoku 6 are definitely translations of somewhat different Sanskrit texts. We do not know which Sanskrit text was earlier, but on the principle of “longer equals later,” we might suppose that Ju-t’ai  ching was based on an earlier text than Tohoku 6.

To give some idea of the differences among the versions, I provide the text of the explanations of the second week of gestation in the three Tibetan translations. The different lengths of the passage are typical of the different translations. The wordiness and repetitiveness of Tohoku 6 is evident at a glance: it uses more than twice as many near-synonyms as the other two translations to describe the “cooking” of the embryo in the womb.

Table 3—Account of Gestation, Week 2

Tohoku 6 dga’  bo  źag  bdun  pa  gñis  pa  la  ni  /  mṅal  gyi  zug  rṅu  mi  gtsaṅ  ba  rul  pa rab  tu  ’bar  ba’i  ’dam  gyi  naṅ  na  ’dug  pa  lus  kyi  dbaṅ  po  thams  cad  rab tu tsha ba  /  śin  tu  mi  bzad  pa  daṅ  /  dog  pa  daṅ  /  ñam  ṅa  bar  gyur  pa sdug  bsṅal  bar  ro gcig  pa’i  rnam  par  śes  pa  can  de  la  ma’i  ltor  naṅ  gi rluṅ  sṅon  gyi  las  kyi  rnam  par  smin  pa  las  skyes  pa  kun  sdud  ces bya  ba dag  ldaṅ  ste  /  de’i  reg  pas  mṅal  gyi  zug  rṅu  de  mer  mer  por  ’gyur  ro  // dga’  bo  de  la  mer  mer  po’i  raṅ  bźin  ji lta  bu  źe  na  /  dga’  bo  dper  na  /  źo ’am  mar  mkhraṅ  po  lta bu  ste  /  mer mer po  ni  de  lta  bu’i  raṅ  bźin  yin  no //  de  der  źag  bdun  du  gnas  śiṅ  /  de  źag  bdun  po  der  gduṅ  ba  daṅ  /  kun  tu gduṅ  ba  daṅ  /  tsha ba  daṅ  /  kun  tu  tsha ba  daṅ  /  yoṅs  su  tsha  bar  ’gyur te  /  de  la  sa’i  khams  ni  sra ba  ñid  du  /  chu’i  khams  ni  gśer  ba  ñid  du  / me’i  khams  ni  dro  ba  ñid  du  /  rluṅ  gi  khams  ni  yaṅ  źiṅ  g.yo  ba  ñid  du  ñe bar  gnas  so  //

PDF Version: ARIRIAB XV (2012), 131-145Tohoku 57 dga’  bo  źag  bdun  phrag  gñis  pa  la  ma’i  mṅal  na  mi  gtsaṅ  ba’i  gnas  su  ’di lta  ste  /  dper  na  zaṅs  kyi  naṅ  du  btsos  śiṅ  brṅos  pa  bźin  du  lus  kyi  dbaṅ  po  daṅ  rnam  par  śes  pa  lhan  cig  tu  sdug  bsṅal  mi  bzad  pa  myoṅ  ba  na ma’i  mṅal  nas  kun  tu  reg pa  źes  bya  ba’i  rluṅ  sṅon  kyi  las  kyi  dbaṅ  gis raṅ  byuṅ  nas  mṅal  de  la  reg pa  gaṅ  yin  pa  de  ni  nur nur po  źes  bya’o  // de’i  dbyis  ni  źo  ska  ba’am  /  mar  ’khyags  pa  ’dra    ste  /  źag  bdun  gyi  bar du  btsos  rab  tu  btsos  pas  na  khams  bźi  po  dag  mṅon  du  snaṅ  bar  ’gyur ro  //

Tohoku 58 źag  bdun  phrag  gñis  pa  la  ma’i  mṅal  du  gnas  pa’i  tshe  las  kyi  rluṅ  mṅon par  grub pa  gaṅ  yin  pa  de  ni  kun  sdud  ces bya  ste  rluṅ  śin  tu  phra  mo des  ma’i  rtsib  logs  g.yon  pa  daṅ  /  g.yas  pa  bus  te  mer  mer  po’i  lus  kyi mtshan ma  rim  gyis  snaṅ  bar  byed  pas  de’i  dbyibs  źo  ska  ba’am  /  mar ’khyags  pa  ’dra  ste  /  mṅal  gyi  naṅ  du  btsos  /  rab  tu  btsos  pas  nur  nur po’i  lus  su  gyur  nas  ’di  ltar  ’byuṅ  ba  chen po  bźi  po  dag  mthar  gyis mṅon  par  grub  par  byed  do  //

The remainder of this article is essentially a comparison of several passages in the two Tibetan translations from the Chinese and the two Chinese texts that are the bases of the translations. Scholarly interest in Tibetan translations from Chinese has been focused for the most part on the fact that the Tibetan used in the translations is often quite different from the Tibetan (broadly

speaking, the vocabulary prescribed by Mahāvyutpatti) used in translations from Sanskrit. Stein analyzes in detail some of the translations from the Chinese, and he identifies examples of what he calls “Chinese vocabulary” in contrast to the “Indian vocabulary” found in texts translated after Khri lde sroṅ btsan’s edict of 814, in which the standardization of the translation vocabulary was decreed (152). For example, the standard Tibetan translation of saṃsāra is ’khor ba, while in a number of texts

translated from Chinese, the word skye śi, a literal translation of the Chinese sheng-ssu 生死 (“birth and death”) appears (Stein 174-175). In the case of the Garbhāvakrāntisūtra, neither of the translations from Chinese seems to contain what Stein refers to as the “Chinese” vocabulary.7 However, Tohoku 57 and Tohoku 58 are of interest for what they may tell us about Tibetan translation practices in Tunhuang in the early ninth century.

In many cases, I believe, differences between Tohoku 57 and 58 and the Chinese translations on which they are based reflect the translator’s reliance on another text, not the one he is supposedly translating. There are many possibilities, including: the other Chinese text (e.g., the translator of Ju-t’ai ching may have been looking at Ch’u-t’ai hui and translated what he found there rather than in Ju-t’ai ching); one of the other Tibetan translations (e.g., the translator of Ju-t’ai ching may have preferred a reading from Tohoku 58 or from Tohoku 6); a Sanskrit manuscript, of the text from which the Chinese translation had been made, the text from which Tohoku 6 had been translated, or yet

I have not read Tohoku 57 and Tohoku 58 in their entirety, but the portions that I have read seem to conform to Stein’s observation that Chos-grub generally follows the dictates of Mahāvyutpatti (149).

PDF Version: ARIRIAB XV (2012), 131-145 another Sanskrit version of the text, unknown to us. Of course, there are also differences that are mere mistakes or that simply reflect the translator’s idiosyncratic understanding of the Chinese.

However, I believe that in most cases when a difference between a Chinese translation and its corresponding Tibetan is attributable to reliance on a different text, the Tibetan translator has deliberately accepted either the wording or the general meaning of the corresponding passage in Tohoku 6. In the following, I discuss a number of passages in Tohoku 58 and Tohoku 57 that appear to rely on another text.

Tohoku

Differences between Tohoku 58 and Ch’u-t’ai hui can be more or less convincingly explained as reflecting influence from Tohoku 6. When the reading in Tohoku 58 disagrees with Ch’u-t’ai hui but agrees with several other translations in addition to Tohoku 6, the influence of Tohoku 6 is less definite. When Tohoku 58 agrees only with Tohoku 6, the influence is fairly clear.

In the first group of passages, Tohoku 58 differs from Ch’u-t’ai hui but agrees with all other versions. Table 4—Passages in Tohoku 58 that disagree with Ch’u-t’ai  hui (1) Differences probably explainable by the reliance of Tohoku 58 on Tohoku

Tohoku 58 Ch’u-ta’i  hui
(處胎會)
Ju-t’ai  ching
(入胎經)

Tohoku 57 Tohoku 6
interlocutor Nanda
(dga’  bo)
Ānanda
(阿難)

Nanda
(難陀)
Nanda
(dga’  bo)

Nanda (dga’ bo)
fault of the womb
oppressed by phlegm

(bad  kan...gis ñe  bar  non  par
gyur—237b1 [all references are to the Derge edition unless otherwise noted])
(missing) having phlegm

(或有黄病痰
癊—T.
1451:

253b16; T. 310 no. 14: 328b4)
affected by phlegm
(bad kan  gyis yoṅs  su  zin  par
gyur—211b7)

oppressed by phlegm
(bad kan gyis  bgags
pa—126a1)

PDF Version: ARIRIAB XV (2012), 131-145fault of the womb
filled with medicine
(sman  gyis  gaṅ  

ba—237b1-2)
has salty disease (有醎8病—T.
310 no. 13: 322b1)

taking medicine (服藥—T.
1451:

253b17; T. 310 no. 14: 328b5)
taking medicine
(sman ’thungs—
212a1)

altered by medicine
(sman  gyis bsgyur
ba—126a1)

Week 15 the 20,000 channels in the front of the body are called
sa ga or sa  ka (lus  kyi  mdun na  yod pa  ñi khri  ni  sa ga
[FPS ka] źes bya’o— 241a5)
{sa ga and sa

ka probably are transliterations of Sanskrit
sakha

(Companion)}
the 20,000 channels in the front of the body are called Conch
(shang  ch’ia)
(身前二萬名
曰商佉—T.

310 no. 13: 323c12)
{shang  ch’ia
probably is a transliteration of Sanskrit
śaṅkha}
(the five types of channels in the front of the body) are called Companion (或名伴 T.
1451:

254c28; T. 310 no. 14: 329c9)
(the five types of channels in the front of the body) are called sa ga or
sa  ka (miṅ  ni  sa ga
[PS ka] źes
kyaṅ  bya—
216a2)

{sa ga and sa
ka probably are transliterations of Sanskrit
sakha

(Companion)}
there are channels called Companion
(grogs  źes bya ba yod
132a3)

In these cases, it is clear that Tohoku 58 does not correspond with the Chinese text that it is supposed to be translating. It is possible that Chos grub relied on either Ju-t’ai ching or its Tibetan translation. My working hypothesis is that Chos grub

translated Tohoku 57 as well, in which case, even if he had translated Tohoku 58 first, he would most likely have already been familiar with the readings of Ju-t’ai ching, and if he had translated Tohoku 57 first, he would have had his own translation to refer to. So it is not impossible that Ju-t’ai  ching or Tohoku 57 is the source of his translation in Tohoku 58.


However, as we can see below, there are other cases in which the reading in Tohoku 58 can only be explained with reference to Tohoku 6, which strongly suggests that it, too, was available to Chos grub. It is possible that the greater length of Tohoku 6, and the fact that it was already translated into Tibetan, prompted Chos grub to give it more authority than the Chinese texts he was

translating. Therefore in these passages, Tohoku 6 is probably the ultimate source of the reading in Tohoku 58. Thus, in the third example, my supposition is that Chos grub must have seen “salty disease” but been aware that, according to Tohoku 6, the condition had something

The Koryō Tripiṭaka reads 鹹 for 醎 (K 22 [13]: 455a20.) I am indebted to Karashima Seishi for this information.

PDF Version: ARIRIAB XV (2012), 131-145to do with medicine. As we can see, the translation “filled with medicine” does not correspond precisely with “altered by medicine” in Tohoku 6. This could mean that the source is a translation or manuscript unknown to us. However, the “fault of the womb” mentioned immediately before this one is described as “the space is filled with flesh” (mtshams śas gaṅ ba) in both Tohoku 58 and Tohoku 6, and the word “filled with” (gaṅ ba) may have been repeated in Tohoku 58 under the influence of the previous item.

The next passage is somewhat more certainly influenced by Tohoku 6. Table 5—Passages in Tohoku 58 that disagree with Ch’u-t’ai  hui (2) Difference very  probably explainable by the reliance of Tohoku 58 on Tohoku 6

Tohoku 58 Ch’u-t’ai  hui
(處胎會)
[Ju-t’ai  ching is
identical]
Tohoku 57 Tohoku 6
fault of the womb
center is like barley
(dbus  nas  ’dra ba)
has barley belly disease
(麥腹病)
center is like barley
(dbus  nas  ’dra ba)
center is like barley
(dbus  nas  ’dra ba)

In this case, the two Chinese translations agree. It appears as though both Tohoku 57 and Tohoku 58 have adopted the reading from Tohoku 6 or perhaps from a Sanskrit manuscript on which Tohoku 6 was based. Thus, the translator or translators of both Tohoku 58 and Tohoku 57 rejected “barley belly” disease in favor of a condition in which the center of the womb is like barley. The only extant full text of the sūtra with such a reading is Tohoku 6, although an abbreviated quotation from the sūtra in Yogācārabhūmi also suggests that center of the womb is being compared to a seed or grain.9

saced yonir vātopastabdhā bhavati pitto pastabdhā vā tilamadhyā vā... (Yogācārabhūmi 22.2-3; for a discussion of this passage, the Chinese and Tibetan translations of which mention barley as well as sesame, see Kritzer “Garbhāvakrāntau”).

PDF Version: ARIRIAB XV (2012), 131-145Finally, the following passages definitely seem to have been influenced by Tohoku 6.

Table 6—Passages in Tohoku 58 that disagree with Ch’u-t’ai  hui (3) Differences almost  certainly explainable by the reliance of Tohoku 58 on Tohoku 6
Tohoku 58
Ch’u-t’ai  hui
(處胎會) Tohoku 6
name of wind, Week 2
All-uniting
(kun  sdud)
All-completing
(遍滿)

All-uniting
(kun  sdud)
 Week 36 since the fetus gives rise to a thought of emerging (from the womb), it becomes unhappy
(de  ’byuṅ  ba’i  bsam pa  skye  bas  mṅon  par mi  dga’  bar  ’gyur
ro—244a5)

the fetus gives rise to a thought of weariness and is unhappy
(生厭離心不以爲
樂—T. 310 no. 13: 324c19)
wishing to emerge, the fetus is unhappy
(de  ’byuṅ  ’dod  ciṅ  mṅon  par  mi
dga’o—136b2)

Week 37 the last of five unmistaken notions is the notion of emerging due to being weary
(yoṅs  su  skyo  nas phyir  ’byuṅ  ba’i  ’du śes  skye ba  gaṅ  yin  pa
ste—244a6)

the last of five unmistaken notions is the notion of loathing (厭惡想—T. 310 no. 13: 324c22)
the last of three unmistaken notions is the notion of emerging
(’byuṅ  ba’i  ’du
śes—136b3)

explanation of why all activities are suffering
“In short, as for this body consisting of the five upādāna
skandhas...” (mdor  na  ñe  bar  len pa’i  phuṅ  po  lṅa’i  lus
’di  ni—247a5)

“This body consisting of the five skandhas...” (此五陰身—T. 310
no. 13: 326a2)

“In short, the five
upādāna skandhas...” (mdor  na  ñe  bar len  pa’i  phuṅ  po
lṅa—141a7)
Here, Ju-t’ai ching and Tohoku 57 do not correspond closely to the other three translations, so the most obvious source for Tohoku 58 is Tohoku 6. For example, in the second passage, Ch’u-t’ai hui mentions the fetus’s thought of weariness, but Chos grub follows Tohoku 6 in saying that it is a thought of emerging. Ju-t’ai ching and Tohoku 57

PDF Version: ARIRIAB XV (2012), 131-145mention neither weariness nor emerging but simply say that the fetus is unhappy.
On the other hand, there are some differences that can only be explained with reference to either Ju-t’ai  ching or Tohoku 57.
Table 7—Passages in Tohoku 58 that disagree with Ch’u-t’ai  hui (4) Differences explainable by the reliance of Tohoku 58 on Tohoku 57 or Ju-t’ai  ching
Tohoku 58 Ch’u-t’ai
hui

(處胎會)
Ju-t’ai  ching
(入胎經)
Tohoku 57
name of wind, Week 4 Inner Explanation
(naṅ  rab  tu  byed pa)
Gather and Receive
(攝取)
Inner Explanation
(内開)

Inner Explanation
(naṅ  rab  tu  byed pa)
name of worm that lives in the anus
Small Bundle
(po  ta  ra  ka)
{Sanskrit poṭalaka}
Breast Wrinkles
(臆皺)
Small Bundle
(小束)
Small Bundle
(po  ta  ra  ka)
{Sanskrit
poṭalaka}

In these cases, Tohoku 6 is quite different and cannot be the source for the differences between Tohoku 58 and Ch’u-t’ai hui. For example, in Tohoku 6 the name of the wind in Week 4 is “Drawing Inward,” and the name of the anal worm is, aptly, “Possessing the Rotten.” These are clearly not reflected in any of the other translations.
Therefore, rejecting the reading in Ch’u-t’ai hui, Chos grub must have based his translation on either Tohoku 57 or Ju-t’ai  ching.
Tohoku 57

Unsurprisingly, since the texts are longer, there are more differences between Tohoku 57 and Ju-t’ai ching than there are between Tohoku 58 and Ch’u-t’ai hui. Again, I present several groups of examples.
Table 8—Passages in Tohoku 57 that disagree with Ju-t’ai  ching (1) Differences probably explainable by the reliance of Tohoku 57 on Tohoku 6

Tohoku 57 Ju-t’ai  ching
(入胎經)
Ch’u-t’ai  hui
(處胎會)
 
Tohoku 58 Tohoku 6
name of wind, Week 3
Treasury Opening
(mdzod  kha)
Sheath Opening
(刀鞘口)
Treasury Opening
(藏口)
Treasury Opening
(mdzod  ka)

Treasury Opening
(mdzod  sgo)
140
PDF Version: ARIRIAB XV (2012), 131-145name of wind, Week 17
Yak Face
(’bri  gdoṅ)
Hair Wipe Mouth
(毛拂口)
Yak Face

(髦牛面)
Yak Face
(’bri  gdoṅ)
Yak Face
(’bri’i  gdoṅ)
Week 21 (simile)
like a plasterer, or a plasterer’s pupil, who prepares some mud with which to plaster a wall
(dper  na  źal źal  mkhan nam  /  źal  źal mkhan  gyi slob  mas ’jim  pa  legs par  sbyar nas  rtsig  ṅos la  źal  źal byed pa
217a7-b1)

like a plasterer, who prepares some mud with which to plaster a wall
(譬如泥師先
好調泥泥於
牆壁—T.
310 no. 14: 330a19-12; T. 1451:
255b11-12)
like a plasterer, or a plasterer’s pupil, who prepares some mud with which to plaster a wall
(譬如泥師及
其弟子。能

善調泥泥諸
牆壁—T.
310 no. 13: 324a22-23)
like a plasterer, or a plasterer’s pupil, who prepares some mud with which to plaster a wall
(dper  na  źal źal  mkhan nam  /  źal  źal mkhan  gyi slob  mas ’jim  pa  legs par  sbyaṅs nas  rtsig  pa rnams  la  źal źal  byed pa
242b1-2)

like a plasterer, or a plasterer’s pupil, who prepares some mud with which to plaster a wall
(dga’  bo  dper na  /  źal  źal mkhan  nam  / źal  źal mkhan  gyi slob  ma mkhas pa ’jim  pa  legs par  byas  pas rtsig  pa’i  gźi la  źal  źal byed  do—
134a2)

These, again, are cases in which the Tibetan translation differs from the Chinese text on which it is supposedly based and agrees with the other Chinese and the two other Tibetan translations. In the first example, we can imagine the underlying Sanskrit, something like kośamukha. Here, I-ching understands kośa in its meaning of a sheath for a sword. But the translator of Tohoku 57, probably based on Tohoku 6, prefers a translation of the other sense of kośa, a storehouse or treasury.
In the following examples, the probability of reliance on Tohoku 6 is even higher.

PDF Version: ARIRIAB XV (2012), 131-145Table 9—Passages in Tohoku 57 that disagree with Ju-t’ai  ching (2) Differences almost  
certainly explainable by the reliance of Tohoku 57 on Tohoku 6
Tohoku 57 Ju-t’ai  ching
(入胎經)
Tohoku 6

Week 4 (shape of the fetus)
like a whetstone or millstone
(DP: de’i  dbyibs  ni  bdar ram  mchi  gu  ‘dra  ste
214b6-7 [P 258a4]) [or like a piece of dirt or a millstone (NFS:

de’i  dbyibs  ni  boṅ  pa ’am  /  mchig  gu  ‘dra
ste—N 441b1; F 383a4-5; S 401a4)]
Like a shoe last or warming stone
(状如鞋楥。或如温
石—T. 310 no. 14: 329b3; T. 1451: 254b16-17)
like a whetstone or millstone
(gdar  ram  mchi  gu  lta bu  ste—130a6-7)

Week 27 (results of bad karma)
even if the being, when reborn, has the three karmas of body, speech, and mind, when he tells people, they will not believe or remember what he says

(lus  daṅ  /  ṅag  daṅ  /  yid kyi  las  ji  sñed  btsal  ba dag  gźan  la  smras  na yaṅ  gźan  gyis  yid  mi ches  śiṅ  yid  la  mi  ‘dzin pa  dag  mṅon  ’grub  par ’gyur  ro—218a5)
even if the being, when reborn, has the three karmas, when he tells people, they will not believe or remember what he says
(所有三業向人説時。
他不信受不將在意—T.

310 no. 14: 330b13-14; T. 1451: 255c8-9)
whatever little bit he accomplishes in the world through body, speech, and mind, none of it will be delightful or acceptable in the world
(des  ’jig rten  la  lus  daṅ ṅag  daṅ  /  yid  kyis  gaṅ  cuṅ  zad  bsgrubs  pa  de thams  cad  kyaṅ  /  ’jig rten  la  bzuṅ  bar  ’os  pa daṅ  /  mṅon  par  dga’ bar  mi ’gyur  ro—
135a1)

ten stages of life: stage 1
weak, he lies on his back
(daṅ  po  ni  byis  pa’i  gnas skabs  ste  ñam  chuṅ  źiṅ  gan  rkyal  du  ñal  bar
byed  do—222b5-6)
he lies in swaddling clothes
(初謂嬰兒位臥於襁
褓—T. 310 no. 14: 332a9; T. 1451: 257b7)
weak, he lies on his back
(gnas  skabs  daṅ  po  la ni  byis pa  dman  pa  gan rkyal  du  ñal  bar  ‘gyur
ro—140a1-2)

the desirability and difficulty of being reborn as a human
a “good world” refers to birth as a human
(bde  ’gro’i  ’jig rten  gaṅ  źe  na  /  ’di  lta  ste  /  de  ni mi’i  ’gro  ba  gaṅ  yin
pa’o—227a7)
a “good world” refers to birth as a human or god
(云何世間善趣。謂是

人天—T. 310 no. 14: 333b10-12; T. 1451: 258c8-10)
a “good world” of the gods is being born as a human
(dga’  bo  lha  rnams  kyi bde  ’gro’i  ’jig rten  gaṅ  źe  na  /  mir  skye ba  yin
te—147b2)

PDF Version: ARIRIAB XV (2012), 131-145These passages are clear evidence of the agreement of Tohoku 57 with Tohoku 6 rather than Ju-t’ai ching. In these cases, Tohoku 58 and Ch’u-t’ai hui do not correspond to the other translations. Taking the third passage as an example, we can see that Tohoku 57 could not possibly be translating what is in Ju-t’ai ching, while it very closely matches Tohoku 6.

Finally, in some cases the reading in Tohoku 57 can best be explained by reliance on Tohoku 58.
Table 10—Passages in Tohoku 57 that disagree with Ju-t’ai  ching (3) Differences probably explainable by the reliance of Tohoku 57 on Tohoku 58
Tohoku 57 Ju-t’ai ching
(入胎經)
Ch’u-ta’i hui
(處胎會)

Tohoku 58 Tohoku 6
fault of the womb
the womb is triangular like the joint of a carriage
(śiṅ  rta’i tshigs  bźin sogs  kar ’dug  pa—
212a1)
the womb is like a carriage shaft
(或如車轅—
T. 310 no. 14: 328b7; T. 1451:
253b19)

the womb is like the bent wood of a carriage shaft
(或如車轅曲
木—T. 310
no. 13: 322b2)
the womb is triangular like the joint of a carriage
(śiṅ  rta’i tshigs  pa bźin  sogs kar  ’dug
pa—237b2)

the opening of the womb is like a carriage
(sgo  śiṅ  rta  lta
bu—126a2)
Week 27 (results of bad karma)
the body of the fetus will be ugly and resemble a
preta (lus  mi  sdug ciṅ  yi  dags daṅ  ’dra  bas—
218a4-5)

the shape of the fetus will resemble a
preta
(形如餓鬼—
T. 310 no. 14: 330b12-13; T. 1451:
255c7-8)
the body of the fetus will be ugly and resemble a
preta
(其身醜陋猶
如餓鬼—T.

310 no. 13: 324b16)
the body of the fetus will be ugly and resemble a
preta (lus  mi  sdug  ciṅ yi  dags  daṅ  ’dra
ba—243a4)
the fetus will resemble
pretas (yi  dags rnams  daṅ  ’dra  bar  ’gyur
gyi—134b7)
In the first case, there is no other translation that agrees with Tohoku 57 and Tohoku 58. In the second case, Ch’u-t’ai hui agrees with Tohoku 57 and Tohoku 58 against Ju-t’ai ching and Tohoku 6. Perhaps this suggests that Tohoku 58 was translated before Tohoku

PDF Version: ARIRIAB XV (2012), 131-145Conclusion
The textual history of Garbhāvakrāntisūtra is very complicated, and the six extant texts are probably only pieces of the whole puzzle. Therefore, any conclusions about the relative age of the different Tibetan translations must remain tentative and subject to revision if new information becomes available.

One thing is definite: the Tibetan translator or translators of Ch’u-t’ai hui and Ju-t’ai ching must have used other texts of the sūtra, either other translations or Sanskrit manuscripts. My own speculation is that Tohoku 6 preceded both Tohoku 58 and Tohoku 57. However, it is difficult to establish the chronological order of Tohoku 58 and Tohoku 57. It seems possible, especially if both translations were indeed the work of Chos grub, that there was two-way contamination, either while the texts were being translated or during the period in which they were being revised and edited. Since the texts were found side-by-side in the Chinese Ratnakūṭa, it is very likely that the translator or translators were familiar with the contents of both texts before they finished translating either one.


References


Chinese texts Ch’u-tai  hui=Fo  wei  a-nan  shuo  ch’u-tai  hui  佛爲阿難説處胎會 (=  Garbhāvakrāntisūtra), translated by
Bodhiruci (P’u-t’i-liu-chih 菩提流志), T. 310 no. 13.
Ju-t’ai  ching=Fo-shuo  ju-t’ai  tsang  hui 佛説入胎藏會 (=  Garbhāvakrāntisūtra), translated by I-ching 義淨,
T. 310 no. 14.
Ju-t’ai  ching= Mūlasarvāstivādavinayakṣudrakavastu (Ken-pen  shuo  i-ch’ieh-yu  pu  p’i-nai-yeh  tsa-shih 根本
説一切有部毘奈耶雜事), translated by I-ching, T. 1451.
Pao-t’ai  ching 胞胎經 (=  Garbhāvakrāntisūtra), translated by Dharmarakṣa (Chu Fa-hu 竺法護), T. 317.
Tibetan texts Tohoku 6= ’Dul ba phran tshegs kyi gźi (=Vinayakṣudrakavastu). Derge ’dul ba tha 124b6-153a1 (sūtra
proper).

Tohoku 57= Dga’ bo la mṅal na gnas pa bstan pa (= Garbhāvakrāntisūtra). Derge dkon brtsegs ga
205b1-236b7 (sūtra proper: 210b5-232a7);. Peking 760 (14) dkon brtsegs wi 248a3-282b4 (sūtra proper: 253b4-277a7); Narthang 45 dkon brtsegs ga 426b6-477a6 (sūtra proper: 434b7-469a2); Phug brag 45 dkon brtsegs kha 369a7-416a8 (sūtra proper: 377a1-409a); Stog 11 (14) dkon brtsegs ga 388a2-435a6 (sūtra proper: 395b1-428a3).
Tohoku 58= Tshe daṅ ldan pa dga’ bo la mṅal du ’jug pa bstan pa (= Garbhāvakrāntisūtra). Derge dkon brtsegs ga 237a1-248a7; Peking 760 (13) dkon brtsegs wi 234a3-248a7; Narthang 44 dkon brtsegs ga 408a1-426b6; Phug brag 44 dkon brtsegs kha 355b7-369a6; Stog 11 (13) dkon brtsegs ga 369a7-388a1.
Sigla for Tibetan editions

D Derge N Narthang F Phug brag P Peking S Stog Palace
Sanskrit text
Yogācārabhūmi. Ed. Vidhushekhara Bhattacharya. Calcutta: University of Calcutta, 1957. Part 1.

PDF Version: ARIRIAB XV (2012), 131-145Secondary sources Garrett, Frances. Religion,  Medicine  and  the  Human  Embryo  in Tibet. London: Routledge, 2008. de Jong, J.W. “The Bodhisattvāvadānakalpalatā and the Ṣaḍdantāvadāna.” In Buddhist Thought and Asian civilization : Essays in Honor of Herbert V. Guenther on his Sixtieth Birthday, ed. Leslie S. Kawamura and Keith Scott. Emeryville: Dharma Press, 1977. 27-38.

Kritzer, Robert. “Childbirth and the Mother’s Body in the Abhidharmakośabhāṣya and Related Texts.” In
Indo tetsugaku bukkyō shisō ron shū: Mikogami Eshō kyōju shōju kinen ronshū (インド哲学佛教
思想論集: 神子上恵生教授頌寿記念論集). Kyoto: Nagatabunshodō, 2004. 1085-1009. _______“Life in the Womb: Conception and Gestation in Buddhist Scripture and Classical Indian Medical

Literature.” In Imagining the Fetus: the Unborn in Myth, Religion, and Culture, ed. Vanessa Sasson
and Jane Marie Law. New York: Oxford University Press, 2009. 73-90.
_______ “Garbhāvakrāntau (‘In the Garbhāvakrānti’): Quotations from the Garbhāvakrāntisūtra in Abhidharma Literature and the Yogācārabhūmi.” In The Yogācārabhūmi and the Yogācāras, ed.

Ulrich Timme Kragh. Cambridge, Mass: Harvard University, Department of Sanskrit and Indian Studies, (forthcoming).
Stein, R.A. “Tibetica antiqua I: Les deux vocabulaires des traductions Indo-tibétaine et Sino-tibétaine dans
les manuscrits de Touen-huoang.” Bulletin de l’École Francaişe de Extrême-Orient 72 (1983):
149-236.

UEYAMA Daishun. “Dai bankoku daitoku sanzō hōshi shamon Hōjō no kenkyū (jō)” 大蕃國大徳三蔵法師 沙門法成の研究(上). Tōhō  gakuhō 38 (1967) 133-198.
YAMABE, Nobuyoshi. “On the School Affiliation of An Shigao: Sarvāstivāda and Yogācāra,” Unpublished draft of a paper given at the international workshop on “The Works of An Shigao” (Leiden, 19-20 Dec., 1996).


Source