It
explains
prahdna-
Poussin 43
?
Poussin 43
?
Abhidharmakosabhasyam-Vol-1-Vasubandhu-Poussin-Pruden-1991
Vinltadeva attributes to the Vibhajyavadins the following thesis:
a. The pudgala exists "absolutely"; b. the past does not exist, with the exception of the cause the result of which has not ripened; the future does not exist, with the exception of the result; the present rigs mi mthunpa (? ) does not exist; c. dhanna does not become an "immediate cause"; d. rupa does not have a "parallel cause," as the Darstantikas claim (Kosa, ii. 52).
7. More notable is the note of Hsiian-tsang (Siddhi, 179) which associates the Vibhajyavadins and the Sthaviras with belief in bhavdngavijndna.
And also: pure mind, Siddhi, 109-111; persistence of a subtle mind in nirodhasamapatti (with the Darstantikas), 207; see also 770.
8. References to Kosa-Vibhdsd:
a. Sound is of retribution (with the Vatslputrlyas) (Kosa, i. 37-38a, Vibhdsd p. 6l2cl3, Siddhi, 190).
The body of arising (janmakdya) of the Buddha is "pure" (with the Mahasarhghikas) (Vibhdsd, p. 871c2, Siddhi, 769-770).
b. Sraddha, etc. , are pure (Kosa, ii. 9, Vibhdsd, p. 7c3).
Life is cittdnuvartin (Kosa, ii. 50, Vibhdsd, p. 770c6--refuted by Vasumitra). Consequently, there is a subtle mind in asamjnisamdpatti, and in niro-
dhasamapatti (Kosa, viii. 33, Vibhdsd, p. 772c21, p. 774al4).
c. Negation of antardbhava (Kosa, iii. 10, Vibhdsd, p. 356cl5, p. 700al5. From
whence the complicated explanation of the antardparinirvdyin, Kosa, iii. 12, Vibhdsd, p. 357b9).
Pratityasamutpdda is asamskrta, like the Path (Kosa, iii. 28, Vibhdsd, p. 116c5, p. 479--like the Mahasarhghikas and the Mahlsasakas).
? d. Greed, anger, false views are "action" (Vibhdsd, p. 587a9, Kosa, iv. 65: Darstantikas).
Definition of "good by nature" (d&jfidna), "through association" (as vijndna), "in origin" (as action of the body . . . ) (Kosa, iv. 8, ix, Vibhdsd, p. 741al5).
T he mind of the Bhagavat is always absorbed (Kosa, iv. 12, Vibhdsd, p. 4l0b26).
e. The thirst for non-existence is abandoned through bhdvand (Kosa, v. 10-11, Vibhdsd, p. 138c3).
On the viparydsas {Kosa v. 9, Vibhdsd, p. 536c9).
f. Definition of the Truths (Kosa, vi. 2, Vibhdsd, p. 397b4).
Comprehension of the Truths at once (Kosa, vi. 27, Vibhdsd, p. 532a).
The Arhat does not fall (Kosa, vi. 58, Vibhdsd, p. 312b9).
Forty one bodhipdksikas (Kosa, vi. 66, Vibhdsd, p. 499a. 4).
g. Rupa in Arupyadhato (Kosa, viii. 3, Vibhdsd, p. 432a22).
Only the First Dhyana has arigas (Vibhdsd, p. 813c28).
The Aryan of the fourth drupya obtains the quality of an Arhat without the
aid of the Path (Vibhdsd, p. 929bl4). (This is thesis 12 of the Mahlsasakas in the treatise of Vasumitra).
c. Yogacarins
Oryoga-dcdryas, as the Chinese reads; we also haveyogdcdracitta (Vydkhyd ii. 49, ad Kosa, ii. 23).
67
1. People who practice yoga or the contemplation oiyogins; (see Kosa, iv. 4,
note, and the Vibhdsd, passim) they seek nirodhasatya (p. 5 34a 19), practice
sunyatdsamddhi, p. 540cll), are disgusted with vedana (Rupadhatu) and samjnd
(Arupyadhatu) (p. 775b3; also p. 35b25; p. 529b4; p. 832a22). The Ratnardiisutra 6
(Siksdsamuccaya, 55) examines the obligations of the vaiydvrtyakara bhiksu, * the intendent and the minister of the monastery, with respect to the dranyaka, to the pindacdrika, to the contemplative or yogdcdrin bhiksu, to the student or bdhusrutye'bhiyukta, to the preacher, dharmakathika.
2. Devoted to yoga, to breathing exercises, to dhyana, etc. , the yogdcdrin becomes, as the Chinese say, a "master ofyoga," ayogdcdrya: they had theories on prdndyama, on sunyatdsamddhi. . . The Kosa, iv. 18-19, mentions a thesis of this school of meditators on rupa which arises through the power of absorption.
3. [This school of meditators became a philosophical school, the Yogacara school, when, under the influence of Maitreya-Asanga, it became attached to the older formula of the Daiabhilmaka\ "The threefold world is only mind. " One can indeed see the relationship between the theories of ecstasy and idealism, and we
Poussin 41
? 42 Introduction
can see how the practice of the "mindless absorption" can lead to the affirmation of a subtle mind . . . Asanga utilizes Darstantika-Sautrantika speculations. ]
Below are the references in the Kosa to the Yogacarins:
Explaining mixeddhydna (Kofsa, vi. 42) by perfuming, (Vibhdsd, p. 879c26, as do the Darstantikas).
Related to the Sautrantikas (Kosa, ii. 34, Vydkhyd: "the opinion of the Sautrantikas or the Yogacarins. ").
The Yogacaracitta admits that an agglomeration can be formed from a single mahdbhuta, as a piece of dry earth; from two, the same, but wet. . . (Kola, ii. 22, Vydkhyd, ii. 49).
In the yogdcdradarsana, there is a manodhdtu which is distinct from the six vijndnas (Vydkhyd, i. 40, ad i. 17; compare the lamraparnlyas).
(The Vijnanavadin denies that the eye sees, Kosa, i. 42).
The Yogacaracitta defines adhimukti (Kosa, ii. 24, Vydkhyd, ii. 51).
According to the Yogacarins, the mindless absorptions are endowed with
mind from the fact of the alayavijnana (Vydkhyd ad ii. 44). Yas'omitra speaks here of the school of Asanga; the same in Kosa, iii. 2; Vydkhyd, ad v. 8 (the 128 klesas of the Yogacarins).
(The "ancient masters" of the Kosa, ii. 44, should be the Darstantikas).
Elsewhere, the Vydkhyd explains the "ancient masters" of the Bhdsya as being "the Yogacarins" or "the Yogacarins, Asanga, etc. " (Vydkhydad iii. 15, iv. 75, vi. 4).
vii. The Sariputrdbhidharma.
This book, Taisho 1548, is divided into four parts: saprafnaka, aprasnaka, samprayukta-safhgraha (three titles which correspond to the first four sections of the Abhidharma according to the Dharmaguptas-Haimavata [J. Przyluski, Concile, 179, 353-4] and reminds us of the Dhdtukdya-Vibhanga-Dhdtukathd, above p. 27; and "succession" (karma or ntddnal).
This is, properly speaking, a sastra, without any appearance of a sutra, with its beginning phrase, "Thus have I heard . . . " It was compiled by Sariputra, either during the lifetime of the master (according to the Ta-chih-tu-lun) or after his Nirvana, to put an end to heresy, for some have "counterfeited the Dharma"
(dharmapratirupaka).
However this may be, it is a very extensive and old treatise, much in the style of the Pali Vibhanga.
The Ta-chih-tu lun establishes some relations between the Abhidharma of 69
Sariputra and the Vatslputrlyas. But I have not encountered, in the work of Sariputra, any mention of the pudgala in the Vatsiputrlya sense of the word.
? Kyokuga Saeki (Kosa, viii. 3) mentions the fact that the Sdriputrdbhidharma admits the existence of rupa and Arupyadhatu. See in fact p. 552a, at the end of
the chapter on the skandhas. This rupa is avijnapti. The book admits the avijnapti, which is a Sarvastivadin invention.
But it is not orthodox Sarvastivadin. It believes that anusaya is disassociated from the mind (p. 690; Kosa, v. 2). It does not contain anything on the existence of
70
the past and the future,
dhdtu, "which should be abandoned" (576c; Koia, vi. 78). Its system of pratyayas, very developed (p. 679b), and its list of the dhdtus (p. 575) have nothing Sarvastivadin about them.
We can get an idea of the style of the Sdriputrdbhidharma by comparing its description of rupaskandha (p. 543) with the Kosa, i. 20 and Vibhanga, 1 and following; its definition of dharmadhdtu (p. 535) with Vibhanga, 89; and its definition of nirodhasatva (p. 553) with Vibhanya, 103.
1. Dharmadhdtu.
The dharmadhdtu is first defined as identical to the dharmdyatana; then, as
made up of vedandskandha, samjndskandha, samskdraskandha, invisible and impalpable (anidarsana, apratigha) rupa, and the asamskrtas (compare Vibhanga, 86). A third definition enumerates, after vedana and samjnd, the series of samskdras associated with the mind (beginning with cetand and ending with klesdnufaya); the series of the samskdras disassociated from the mind {viprayukta, see p. 547b): jati, jard, marana . . . nirodhasamdpatti\ finally: pratisamkhydniro- dha, apratisamkhydnirodha, niyamadharmasthiti[ta], dkdsdyatana, vijndndyatana,
11
dkimcanydyatana, naivasamjndndsamjndyatana, that is to say the list of the
asamskrtas: "this is what is called the dharmadhdtu"
On the one hand, the viprayuktas are not those of the Sarvastivadins;
although there is some doubt with respect to the equivalents of the translators (Dharmagupta and Dharmayasas, 414 A. D. ), the ndmakdya . . . are missing.
On the other hand, the asamskrtas of Sariputra recall those of the Maha- samghikas and the Mahisasakas (Siddhi, p. 78).
2. Nirodhasatya.
To the question: "What is duhkhanirodha dryasatya? ",our text answers in canonical terms: yo tassdyeva tanhdya asesavirdganirodho cdgo patinissagqo mutti andlayo (Vibhanga 103), and adds: "already cut off, not to arise anew: this is what is called duhkhanirodha dryasatya. "
The question is repeated: "What is duhkhanirodha dryasatya? Pratisamkhyd- nirodha is called duhkhanirodha dryasatya. This duhkhanirodha dryasatya is in
nor on "unconditioned" space.
It explains prahdna-
Poussin 43
? 44 Introduction
truth like that, not like that, no different, not a different thing. As the Tathagata has well spoken the truths of the Aryans, it is dryasatya. "
But, "what is pratisamkhydnirodhaT The question is repeated three times: "If a dharma is destroyed when one obtains the Aryan Path, the destruction of this dharma is called pratisamkhydnirodha" . . . "The four srdmanyaphalas are called pratisamkhydnirodha. "
"What is srotadpannaphala? If the three klesas to be cut off by Seeing are cut off; if satkayadrsti, vicikitsd and silavrata are exhausted, this is called srotadpanna-
12
phala" Sariputra takes up the question again, "What is srotadpannaphala} The
three klesas to be cut off by Seeing being cut off, satkdyadrsti-vicikitsd-silavrata being exhausted: if one obtains amrta, this is what is called srotadpannaphala"
It appears that what we are encountering here is a terminology alien to the Abhidharma and to the Sarvastivada.
73 viii. The Abhidharmasdra.
1. Before Vasubandhu, many masters undertook to summarize the doctrines of the Abhidharma. We possess notably three works: 1. The Abhidharmasdra of
74
Dharmas'rl in ten chapters, made up of kdrikds (probably in dryan stroph)
and a
commentary; 2. a second edition of this same Sara by Upasanta, to which the
Chinese give the name of Abhidharmasdra-ching [=sutra]: the same kdrikds with
a more developed commentary; and 3. a third edition of the Sara, the Tsa
[=Miscellaneous] Abhidharma-ching, by Dharmatrata, which is in fact a new
75 work, containing a new chapter and many new kdrikds.
76
2. The preface to the Vibhdsd (Taisho 1546)
Dharmasri before the Jndnaprasthdna: "After the nirodha of the Buddha, the bhiksu Dharmas'ri composed the four volumes of the Abhidharmasdra. Then Katyayanlputra composed the Abhidharma in eight books . . . "
3. The work of Dharmasri contains ten chapters: Dhdtu, Samskara, Anusaya, Arya, Jndna, Samddhi, Sutra, Tsa and Sdstravarga or Vddavarga.
Between the ninth and the tenth chapters of Dharmasri, Dharmatrata places a new chapter, the Pravicayavarga, which indeed appears to constitute an independent work.
There is a stanza of introduction: "even though many dharmas have been spoken of, their meaning remains confused . . . " and four concluding stanzas: "The author has composed this book based on the book of Dharmasri, not through pride or in order to acquire a reputation . . . ".
It begins with the dharmacakra, the Wheel of the Dharma: "The Muni said that the darsanamdrga is called dharmacakra, either because it goes into the mind
by Tao-yen places the work of
? of others . . . (Kosa, vi. 54).
There then comes the brahmacakra (vi. 54, vii. 31), the updsaka (iv. 69 ), the
four parts of sila (iv. 29), the prdtimoksa . . . Later (p. 959b), cosmology: the periods of loss, etc. (iii. 99), destruction by fire, etc. (iii. 102); and then there follows the theory of the three "fallings away" (p. 960c; Kosa, vi. 59) and the definition of the Bodhisattva (iv. 108).
Suddenly (p. 96lc): "How many types of Sarvastivada are there? " Presenta-
tion of the four doctrines (Kosa, v. 25-26) without mentioning the name of the
four masters. The second and the fourth are bad because they confuse the time
periods. The first (difference in bhdva, translated fen): "One should know that
77 this is the parindma-sarvastivada. "
There is a diversity of opinion as to whether the Truths are seen at the same time (Kosa, vi. 27), Sarvastivadins and Vatslputrlyas on the one hand, Dharma- gupta on the other; antardbhava (iii. 34); then the Sarvastivadin proof. And at the end of the paragraph, the discussion "whether the Buddha is part of the Samgha. " Finally, the concluding stanzas.
4. The Samskdravarga treats of the simultaneous arising of the citta-caittas and of atoms (Kosa, ii. 22), of the four laksanas of "conditioned things" (ii. 45), of the hetus and the pratyayas (ii. 48, 61).
The Sutravarga is a collection of notes on the three Dhatus and a calculation of the places that they contain: sixteen in Rupadhatu, but, according to some, seventeen (Kosa, iii. 2): the sattvdvdsas (iii. 6), the vijndnasthitis (iii. 5-6); the three vartman of pratityasamutpdda (iii. 20, 25), the twelve limbs; the mahdbhutas, the Truths, the fruits of the Aryans, etc.
The Tsa-varga defines the mind-mental states as samprayukta, sdsraya, etc. (ii. 34); it enumerates the viprayuktas: dsamjnika, two non-conscious absorptions, sabhdgatd, ndmakdyddayas, jivitendriya, dharmaprdpti, prthagjanatva, four laksanas (compare ii. 35-36); it concludes with half a karika on the four bhavas (in. 13) and a karika on "disgust" amd "detachment" (vi. 79).
The Sdstravaraga (or Vddavarga) is made up of ten questions in verse, followed by answers in prose, relative to samvara (iv. 13), to the results, etc. Dharmatrata adds sixteen questions.
5. In order to appreciate the character of the treatises of DharmaSrT, Upasanta, and Dharmatrata, and Vasubandhu's debt with respect to Dharmatrata, which appears to be notable, we may see how two dharmaparydyas, the chapter of the three obstacles (dvaranas) and that of avijnapti, are treated by the different masters.
a. Obstacles, Kola, iv. 95-102, Vibhdsd, p. 599.
Poussin 45
? 46 Introduction
Dharmasri, p. 815: "The Bhagavat says that there are three avaranas: karman, kiefa, and vipdka. What is their definition?
"Anantarya actions which are without remedy, developed defilements, bad aaion experienced in the painful realms of rebirth, are the dvaranas.
"These three form an obstacle to the Dharma; they hinder the grasping of the Aryan dharmas\ they are thus called 'obstacles/ Which is the worst aaion?
"The action which divides the Sarhgha is said to be the worst.
"This action is the worst. One guilty of this remains a kalpa in Avici hell. Which is the best aaion?
"The cetand or 'volition' of Bhavagra is the greatest. u
Naivasamjndndsamjndyatana is Bhavagra. The volition which belongs to the realms of this sphere is the greatest and finest: its result is a life of some 80,000 kalpas in length. "
Upasanta, p. 843b-c, has the same two stanzas, but a less meager commentary:
"That which hinders the Path of the Aryans and the means (updya) to this path is said to be an obstacle. The obstacles to aaion are the five dnantaryas, namely, the killing of one's father, etc . . . He who commits such an aaion is immediately and necessarily reborn in Avici: thus the aaion is dnatarya. The killing of one's father and mother destroys goodness, hence it is Avici hell. Those guilty of the other three are led to injure a field of merit. The obstacles of the defilements are 'agitated' and 'sharp' defilements: the first is habitual defilement; the second is the overriding defilement. This refers to the 'present' defilements, not to the defilements that one 'possesses' (that one has as potential), for all beings 'possess' all the defilements . . . "
There is a variant to the second stanza: "Lying which divides the Samgha . . . ; volition in Bhavagra, among good aaions, has the greatest result," which is better. The commentary notes the differences in the two schisms (cakrabheda, karmabheda).
Dharmatrata is longer (p. 898b-899c) and very close to Vasubandhu:
According to the first stanza of Dharmasri: 1. the dvarana of klesa is the worst; the dvarana of aaion, mediocre; and the dvarana of retribution, the least; 2. the division of the Samgha, by nature, is non-concord; this is a viprayukta samskdra of the anivrta-avydkrta class; 3. the division is a thing of the Samgha; the transgression is of him who divides the Samgha; he experiences, in Avici, a retribution of kalpa\ 4. the bhiksus are divided in their opinion of who is the Master, of what is the Path: this is the division of the Samgha which was united, and he who breaks it is 'one who possesses views' {drspicarita)\ 5. in three continents, a minimum of eight persons is required for karmabheda\ in
? JambudvTpa, a minimum of nine persons is required for cakrabheda; 6. cakrabheda is impossible in six time periods: when a boundary is not delimited; at first; following; when the Muni has passed into Nirvana; when the running sore has not been produced; when the pair of chief disciples has not been established (six pddas); 7. lying which divides the Sarhgha is the worst of aaions; the volition of Bhavagra is said to bear the greatest result.
b. Avijnapti, Kosa, iv. 2, 3.
DharmasrT (p, 812c):
"Bodily action is vijnapti and avijnapti. The vijnapti of the body is the movement of the body, good, bad, or neutral: good when it arises from a good mind . . . For avijnapti: when one does an action in a firm manner, the mind can change, but the seed remains. If, for example, a person undertakes the precepts, his mind can then be bad or neutral: nevertheless the precepts continue . . . Action of the manas is solely avijnapti. . . because this action is not visible . . . vijnapti is good, bad, or neutral; the same for the avijnapti which belongs to the manas. The other vijnaptis are never neutral. "
Upasanta (p. 840) adds a bit. The hunter is regarded as free from bodily avijnapti. Mental action is called avijnapti because it does not inform others. Some say that it is called vijnapti because it is discourse (jalpa? ).
Dharmatrata (p. 888b) replaces the terms vijnapti and avijnapti with "doing" and "not doing" (karana, akarana) (Kos'ajv. 14):
"Bodily action is of two types: karanasvabhava, or akaranasvabhava. 'Doing'
78
(karana): movement of the body, exercise
'Not doing' (akarana, wu-tso): when the movement of the body has ended, the nature (good or bad) of this movement, of this action which is the movement, continues to arise, simultaneous with minds of a different nature, even as the good precepts produced by their undertaking (kusalasamddanasila) continue to arise even when bad or neutral minds are present. Like the immoral person (dauhsilya- purusa): even when good or neutral minds are present, his immorality continues to arise. "
". . . Action of the manas is cetand, volition, by nature . . .
"'Not doing' (akarana) is also called nirati (? Vyut. 21, 114), virati; upeksa, akriyd (pu-tso). Because it does not do, it is called 'not doing. ' If one says that this is not an action (karman), this is wrong, because it does. Good does not do evil, evil does not do good: this is also an action. As the upeksa part of Bodhi is not upeksa by reason of what is called upeksa; but the practice of the Path, the arresting of things, is called upeksa. The same here. Furthermore, in doing the
of the body is the 'doing of the body. '
Poussin 47
? 48 Introduction
cause one does the result: . . . 'not doing' is not rupa, but the doing of it (which is the cause of the 'not doing' or avijnapti) is rupa\ 'not doing' is thus called riipa. In
79 this same way then, 'not doing' is action.
? 1. Originally printed as a Foreword to de La Vallee Poussin's Cosmologie bouddhique: 1913, and published 1919 in the four-part Memoires of l'Acadmie royale de Belgique (Luzac, London). This contained the restoration of the kdrikds of the third chapter of the Ko/a, the Tibetan kdrikds, the Bhdsya, and the text of the Vydkhyd\ in the appendix, a summary of the Lokaprajnapti and the Karanaprajnapti.
2. Cullavagga, xi. 1. 8. [For a more exact presentation,J. Przyluski, Concile de Rdjagrha, p. 311,345, 349].
3. Oldenberg, Buddha . . . , 6th edition, p. 202; Fr. trans. Foucher, 2nd edition, p. 177.
a. The pudgala exists "absolutely"; b. the past does not exist, with the exception of the cause the result of which has not ripened; the future does not exist, with the exception of the result; the present rigs mi mthunpa (? ) does not exist; c. dhanna does not become an "immediate cause"; d. rupa does not have a "parallel cause," as the Darstantikas claim (Kosa, ii. 52).
7. More notable is the note of Hsiian-tsang (Siddhi, 179) which associates the Vibhajyavadins and the Sthaviras with belief in bhavdngavijndna.
And also: pure mind, Siddhi, 109-111; persistence of a subtle mind in nirodhasamapatti (with the Darstantikas), 207; see also 770.
8. References to Kosa-Vibhdsd:
a. Sound is of retribution (with the Vatslputrlyas) (Kosa, i. 37-38a, Vibhdsd p. 6l2cl3, Siddhi, 190).
The body of arising (janmakdya) of the Buddha is "pure" (with the Mahasarhghikas) (Vibhdsd, p. 871c2, Siddhi, 769-770).
b. Sraddha, etc. , are pure (Kosa, ii. 9, Vibhdsd, p. 7c3).
Life is cittdnuvartin (Kosa, ii. 50, Vibhdsd, p. 770c6--refuted by Vasumitra). Consequently, there is a subtle mind in asamjnisamdpatti, and in niro-
dhasamapatti (Kosa, viii. 33, Vibhdsd, p. 772c21, p. 774al4).
c. Negation of antardbhava (Kosa, iii. 10, Vibhdsd, p. 356cl5, p. 700al5. From
whence the complicated explanation of the antardparinirvdyin, Kosa, iii. 12, Vibhdsd, p. 357b9).
Pratityasamutpdda is asamskrta, like the Path (Kosa, iii. 28, Vibhdsd, p. 116c5, p. 479--like the Mahasarhghikas and the Mahlsasakas).
? d. Greed, anger, false views are "action" (Vibhdsd, p. 587a9, Kosa, iv. 65: Darstantikas).
Definition of "good by nature" (d&jfidna), "through association" (as vijndna), "in origin" (as action of the body . . . ) (Kosa, iv. 8, ix, Vibhdsd, p. 741al5).
T he mind of the Bhagavat is always absorbed (Kosa, iv. 12, Vibhdsd, p. 4l0b26).
e. The thirst for non-existence is abandoned through bhdvand (Kosa, v. 10-11, Vibhdsd, p. 138c3).
On the viparydsas {Kosa v. 9, Vibhdsd, p. 536c9).
f. Definition of the Truths (Kosa, vi. 2, Vibhdsd, p. 397b4).
Comprehension of the Truths at once (Kosa, vi. 27, Vibhdsd, p. 532a).
The Arhat does not fall (Kosa, vi. 58, Vibhdsd, p. 312b9).
Forty one bodhipdksikas (Kosa, vi. 66, Vibhdsd, p. 499a. 4).
g. Rupa in Arupyadhato (Kosa, viii. 3, Vibhdsd, p. 432a22).
Only the First Dhyana has arigas (Vibhdsd, p. 813c28).
The Aryan of the fourth drupya obtains the quality of an Arhat without the
aid of the Path (Vibhdsd, p. 929bl4). (This is thesis 12 of the Mahlsasakas in the treatise of Vasumitra).
c. Yogacarins
Oryoga-dcdryas, as the Chinese reads; we also haveyogdcdracitta (Vydkhyd ii. 49, ad Kosa, ii. 23).
67
1. People who practice yoga or the contemplation oiyogins; (see Kosa, iv. 4,
note, and the Vibhdsd, passim) they seek nirodhasatya (p. 5 34a 19), practice
sunyatdsamddhi, p. 540cll), are disgusted with vedana (Rupadhatu) and samjnd
(Arupyadhatu) (p. 775b3; also p. 35b25; p. 529b4; p. 832a22). The Ratnardiisutra 6
(Siksdsamuccaya, 55) examines the obligations of the vaiydvrtyakara bhiksu, * the intendent and the minister of the monastery, with respect to the dranyaka, to the pindacdrika, to the contemplative or yogdcdrin bhiksu, to the student or bdhusrutye'bhiyukta, to the preacher, dharmakathika.
2. Devoted to yoga, to breathing exercises, to dhyana, etc. , the yogdcdrin becomes, as the Chinese say, a "master ofyoga," ayogdcdrya: they had theories on prdndyama, on sunyatdsamddhi. . . The Kosa, iv. 18-19, mentions a thesis of this school of meditators on rupa which arises through the power of absorption.
3. [This school of meditators became a philosophical school, the Yogacara school, when, under the influence of Maitreya-Asanga, it became attached to the older formula of the Daiabhilmaka\ "The threefold world is only mind. " One can indeed see the relationship between the theories of ecstasy and idealism, and we
Poussin 41
? 42 Introduction
can see how the practice of the "mindless absorption" can lead to the affirmation of a subtle mind . . . Asanga utilizes Darstantika-Sautrantika speculations. ]
Below are the references in the Kosa to the Yogacarins:
Explaining mixeddhydna (Kofsa, vi. 42) by perfuming, (Vibhdsd, p. 879c26, as do the Darstantikas).
Related to the Sautrantikas (Kosa, ii. 34, Vydkhyd: "the opinion of the Sautrantikas or the Yogacarins. ").
The Yogacaracitta admits that an agglomeration can be formed from a single mahdbhuta, as a piece of dry earth; from two, the same, but wet. . . (Kola, ii. 22, Vydkhyd, ii. 49).
In the yogdcdradarsana, there is a manodhdtu which is distinct from the six vijndnas (Vydkhyd, i. 40, ad i. 17; compare the lamraparnlyas).
(The Vijnanavadin denies that the eye sees, Kosa, i. 42).
The Yogacaracitta defines adhimukti (Kosa, ii. 24, Vydkhyd, ii. 51).
According to the Yogacarins, the mindless absorptions are endowed with
mind from the fact of the alayavijnana (Vydkhyd ad ii. 44). Yas'omitra speaks here of the school of Asanga; the same in Kosa, iii. 2; Vydkhyd, ad v. 8 (the 128 klesas of the Yogacarins).
(The "ancient masters" of the Kosa, ii. 44, should be the Darstantikas).
Elsewhere, the Vydkhyd explains the "ancient masters" of the Bhdsya as being "the Yogacarins" or "the Yogacarins, Asanga, etc. " (Vydkhydad iii. 15, iv. 75, vi. 4).
vii. The Sariputrdbhidharma.
This book, Taisho 1548, is divided into four parts: saprafnaka, aprasnaka, samprayukta-safhgraha (three titles which correspond to the first four sections of the Abhidharma according to the Dharmaguptas-Haimavata [J. Przyluski, Concile, 179, 353-4] and reminds us of the Dhdtukdya-Vibhanga-Dhdtukathd, above p. 27; and "succession" (karma or ntddnal).
This is, properly speaking, a sastra, without any appearance of a sutra, with its beginning phrase, "Thus have I heard . . . " It was compiled by Sariputra, either during the lifetime of the master (according to the Ta-chih-tu-lun) or after his Nirvana, to put an end to heresy, for some have "counterfeited the Dharma"
(dharmapratirupaka).
However this may be, it is a very extensive and old treatise, much in the style of the Pali Vibhanga.
The Ta-chih-tu lun establishes some relations between the Abhidharma of 69
Sariputra and the Vatslputrlyas. But I have not encountered, in the work of Sariputra, any mention of the pudgala in the Vatsiputrlya sense of the word.
? Kyokuga Saeki (Kosa, viii. 3) mentions the fact that the Sdriputrdbhidharma admits the existence of rupa and Arupyadhatu. See in fact p. 552a, at the end of
the chapter on the skandhas. This rupa is avijnapti. The book admits the avijnapti, which is a Sarvastivadin invention.
But it is not orthodox Sarvastivadin. It believes that anusaya is disassociated from the mind (p. 690; Kosa, v. 2). It does not contain anything on the existence of
70
the past and the future,
dhdtu, "which should be abandoned" (576c; Koia, vi. 78). Its system of pratyayas, very developed (p. 679b), and its list of the dhdtus (p. 575) have nothing Sarvastivadin about them.
We can get an idea of the style of the Sdriputrdbhidharma by comparing its description of rupaskandha (p. 543) with the Kosa, i. 20 and Vibhanga, 1 and following; its definition of dharmadhdtu (p. 535) with Vibhanga, 89; and its definition of nirodhasatva (p. 553) with Vibhanya, 103.
1. Dharmadhdtu.
The dharmadhdtu is first defined as identical to the dharmdyatana; then, as
made up of vedandskandha, samjndskandha, samskdraskandha, invisible and impalpable (anidarsana, apratigha) rupa, and the asamskrtas (compare Vibhanga, 86). A third definition enumerates, after vedana and samjnd, the series of samskdras associated with the mind (beginning with cetand and ending with klesdnufaya); the series of the samskdras disassociated from the mind {viprayukta, see p. 547b): jati, jard, marana . . . nirodhasamdpatti\ finally: pratisamkhydniro- dha, apratisamkhydnirodha, niyamadharmasthiti[ta], dkdsdyatana, vijndndyatana,
11
dkimcanydyatana, naivasamjndndsamjndyatana, that is to say the list of the
asamskrtas: "this is what is called the dharmadhdtu"
On the one hand, the viprayuktas are not those of the Sarvastivadins;
although there is some doubt with respect to the equivalents of the translators (Dharmagupta and Dharmayasas, 414 A. D. ), the ndmakdya . . . are missing.
On the other hand, the asamskrtas of Sariputra recall those of the Maha- samghikas and the Mahisasakas (Siddhi, p. 78).
2. Nirodhasatya.
To the question: "What is duhkhanirodha dryasatya? ",our text answers in canonical terms: yo tassdyeva tanhdya asesavirdganirodho cdgo patinissagqo mutti andlayo (Vibhanga 103), and adds: "already cut off, not to arise anew: this is what is called duhkhanirodha dryasatya. "
The question is repeated: "What is duhkhanirodha dryasatya? Pratisamkhyd- nirodha is called duhkhanirodha dryasatya. This duhkhanirodha dryasatya is in
nor on "unconditioned" space.
It explains prahdna-
Poussin 43
? 44 Introduction
truth like that, not like that, no different, not a different thing. As the Tathagata has well spoken the truths of the Aryans, it is dryasatya. "
But, "what is pratisamkhydnirodhaT The question is repeated three times: "If a dharma is destroyed when one obtains the Aryan Path, the destruction of this dharma is called pratisamkhydnirodha" . . . "The four srdmanyaphalas are called pratisamkhydnirodha. "
"What is srotadpannaphala? If the three klesas to be cut off by Seeing are cut off; if satkayadrsti, vicikitsd and silavrata are exhausted, this is called srotadpanna-
12
phala" Sariputra takes up the question again, "What is srotadpannaphala} The
three klesas to be cut off by Seeing being cut off, satkdyadrsti-vicikitsd-silavrata being exhausted: if one obtains amrta, this is what is called srotadpannaphala"
It appears that what we are encountering here is a terminology alien to the Abhidharma and to the Sarvastivada.
73 viii. The Abhidharmasdra.
1. Before Vasubandhu, many masters undertook to summarize the doctrines of the Abhidharma. We possess notably three works: 1. The Abhidharmasdra of
74
Dharmas'rl in ten chapters, made up of kdrikds (probably in dryan stroph)
and a
commentary; 2. a second edition of this same Sara by Upasanta, to which the
Chinese give the name of Abhidharmasdra-ching [=sutra]: the same kdrikds with
a more developed commentary; and 3. a third edition of the Sara, the Tsa
[=Miscellaneous] Abhidharma-ching, by Dharmatrata, which is in fact a new
75 work, containing a new chapter and many new kdrikds.
76
2. The preface to the Vibhdsd (Taisho 1546)
Dharmasri before the Jndnaprasthdna: "After the nirodha of the Buddha, the bhiksu Dharmas'ri composed the four volumes of the Abhidharmasdra. Then Katyayanlputra composed the Abhidharma in eight books . . . "
3. The work of Dharmasri contains ten chapters: Dhdtu, Samskara, Anusaya, Arya, Jndna, Samddhi, Sutra, Tsa and Sdstravarga or Vddavarga.
Between the ninth and the tenth chapters of Dharmasri, Dharmatrata places a new chapter, the Pravicayavarga, which indeed appears to constitute an independent work.
There is a stanza of introduction: "even though many dharmas have been spoken of, their meaning remains confused . . . " and four concluding stanzas: "The author has composed this book based on the book of Dharmasri, not through pride or in order to acquire a reputation . . . ".
It begins with the dharmacakra, the Wheel of the Dharma: "The Muni said that the darsanamdrga is called dharmacakra, either because it goes into the mind
by Tao-yen places the work of
? of others . . . (Kosa, vi. 54).
There then comes the brahmacakra (vi. 54, vii. 31), the updsaka (iv. 69 ), the
four parts of sila (iv. 29), the prdtimoksa . . . Later (p. 959b), cosmology: the periods of loss, etc. (iii. 99), destruction by fire, etc. (iii. 102); and then there follows the theory of the three "fallings away" (p. 960c; Kosa, vi. 59) and the definition of the Bodhisattva (iv. 108).
Suddenly (p. 96lc): "How many types of Sarvastivada are there? " Presenta-
tion of the four doctrines (Kosa, v. 25-26) without mentioning the name of the
four masters. The second and the fourth are bad because they confuse the time
periods. The first (difference in bhdva, translated fen): "One should know that
77 this is the parindma-sarvastivada. "
There is a diversity of opinion as to whether the Truths are seen at the same time (Kosa, vi. 27), Sarvastivadins and Vatslputrlyas on the one hand, Dharma- gupta on the other; antardbhava (iii. 34); then the Sarvastivadin proof. And at the end of the paragraph, the discussion "whether the Buddha is part of the Samgha. " Finally, the concluding stanzas.
4. The Samskdravarga treats of the simultaneous arising of the citta-caittas and of atoms (Kosa, ii. 22), of the four laksanas of "conditioned things" (ii. 45), of the hetus and the pratyayas (ii. 48, 61).
The Sutravarga is a collection of notes on the three Dhatus and a calculation of the places that they contain: sixteen in Rupadhatu, but, according to some, seventeen (Kosa, iii. 2): the sattvdvdsas (iii. 6), the vijndnasthitis (iii. 5-6); the three vartman of pratityasamutpdda (iii. 20, 25), the twelve limbs; the mahdbhutas, the Truths, the fruits of the Aryans, etc.
The Tsa-varga defines the mind-mental states as samprayukta, sdsraya, etc. (ii. 34); it enumerates the viprayuktas: dsamjnika, two non-conscious absorptions, sabhdgatd, ndmakdyddayas, jivitendriya, dharmaprdpti, prthagjanatva, four laksanas (compare ii. 35-36); it concludes with half a karika on the four bhavas (in. 13) and a karika on "disgust" amd "detachment" (vi. 79).
The Sdstravaraga (or Vddavarga) is made up of ten questions in verse, followed by answers in prose, relative to samvara (iv. 13), to the results, etc. Dharmatrata adds sixteen questions.
5. In order to appreciate the character of the treatises of DharmaSrT, Upasanta, and Dharmatrata, and Vasubandhu's debt with respect to Dharmatrata, which appears to be notable, we may see how two dharmaparydyas, the chapter of the three obstacles (dvaranas) and that of avijnapti, are treated by the different masters.
a. Obstacles, Kola, iv. 95-102, Vibhdsd, p. 599.
Poussin 45
? 46 Introduction
Dharmasri, p. 815: "The Bhagavat says that there are three avaranas: karman, kiefa, and vipdka. What is their definition?
"Anantarya actions which are without remedy, developed defilements, bad aaion experienced in the painful realms of rebirth, are the dvaranas.
"These three form an obstacle to the Dharma; they hinder the grasping of the Aryan dharmas\ they are thus called 'obstacles/ Which is the worst aaion?
"The action which divides the Sarhgha is said to be the worst.
"This action is the worst. One guilty of this remains a kalpa in Avici hell. Which is the best aaion?
"The cetand or 'volition' of Bhavagra is the greatest. u
Naivasamjndndsamjndyatana is Bhavagra. The volition which belongs to the realms of this sphere is the greatest and finest: its result is a life of some 80,000 kalpas in length. "
Upasanta, p. 843b-c, has the same two stanzas, but a less meager commentary:
"That which hinders the Path of the Aryans and the means (updya) to this path is said to be an obstacle. The obstacles to aaion are the five dnantaryas, namely, the killing of one's father, etc . . . He who commits such an aaion is immediately and necessarily reborn in Avici: thus the aaion is dnatarya. The killing of one's father and mother destroys goodness, hence it is Avici hell. Those guilty of the other three are led to injure a field of merit. The obstacles of the defilements are 'agitated' and 'sharp' defilements: the first is habitual defilement; the second is the overriding defilement. This refers to the 'present' defilements, not to the defilements that one 'possesses' (that one has as potential), for all beings 'possess' all the defilements . . . "
There is a variant to the second stanza: "Lying which divides the Samgha . . . ; volition in Bhavagra, among good aaions, has the greatest result," which is better. The commentary notes the differences in the two schisms (cakrabheda, karmabheda).
Dharmatrata is longer (p. 898b-899c) and very close to Vasubandhu:
According to the first stanza of Dharmasri: 1. the dvarana of klesa is the worst; the dvarana of aaion, mediocre; and the dvarana of retribution, the least; 2. the division of the Samgha, by nature, is non-concord; this is a viprayukta samskdra of the anivrta-avydkrta class; 3. the division is a thing of the Samgha; the transgression is of him who divides the Samgha; he experiences, in Avici, a retribution of kalpa\ 4. the bhiksus are divided in their opinion of who is the Master, of what is the Path: this is the division of the Samgha which was united, and he who breaks it is 'one who possesses views' {drspicarita)\ 5. in three continents, a minimum of eight persons is required for karmabheda\ in
? JambudvTpa, a minimum of nine persons is required for cakrabheda; 6. cakrabheda is impossible in six time periods: when a boundary is not delimited; at first; following; when the Muni has passed into Nirvana; when the running sore has not been produced; when the pair of chief disciples has not been established (six pddas); 7. lying which divides the Sarhgha is the worst of aaions; the volition of Bhavagra is said to bear the greatest result.
b. Avijnapti, Kosa, iv. 2, 3.
DharmasrT (p, 812c):
"Bodily action is vijnapti and avijnapti. The vijnapti of the body is the movement of the body, good, bad, or neutral: good when it arises from a good mind . . . For avijnapti: when one does an action in a firm manner, the mind can change, but the seed remains. If, for example, a person undertakes the precepts, his mind can then be bad or neutral: nevertheless the precepts continue . . . Action of the manas is solely avijnapti. . . because this action is not visible . . . vijnapti is good, bad, or neutral; the same for the avijnapti which belongs to the manas. The other vijnaptis are never neutral. "
Upasanta (p. 840) adds a bit. The hunter is regarded as free from bodily avijnapti. Mental action is called avijnapti because it does not inform others. Some say that it is called vijnapti because it is discourse (jalpa? ).
Dharmatrata (p. 888b) replaces the terms vijnapti and avijnapti with "doing" and "not doing" (karana, akarana) (Kos'ajv. 14):
"Bodily action is of two types: karanasvabhava, or akaranasvabhava. 'Doing'
78
(karana): movement of the body, exercise
'Not doing' (akarana, wu-tso): when the movement of the body has ended, the nature (good or bad) of this movement, of this action which is the movement, continues to arise, simultaneous with minds of a different nature, even as the good precepts produced by their undertaking (kusalasamddanasila) continue to arise even when bad or neutral minds are present. Like the immoral person (dauhsilya- purusa): even when good or neutral minds are present, his immorality continues to arise. "
". . . Action of the manas is cetand, volition, by nature . . .
"'Not doing' (akarana) is also called nirati (? Vyut. 21, 114), virati; upeksa, akriyd (pu-tso). Because it does not do, it is called 'not doing. ' If one says that this is not an action (karman), this is wrong, because it does. Good does not do evil, evil does not do good: this is also an action. As the upeksa part of Bodhi is not upeksa by reason of what is called upeksa; but the practice of the Path, the arresting of things, is called upeksa. The same here. Furthermore, in doing the
of the body is the 'doing of the body. '
Poussin 47
? 48 Introduction
cause one does the result: . . . 'not doing' is not rupa, but the doing of it (which is the cause of the 'not doing' or avijnapti) is rupa\ 'not doing' is thus called riipa. In
79 this same way then, 'not doing' is action.
? 1. Originally printed as a Foreword to de La Vallee Poussin's Cosmologie bouddhique: 1913, and published 1919 in the four-part Memoires of l'Acadmie royale de Belgique (Luzac, London). This contained the restoration of the kdrikds of the third chapter of the Ko/a, the Tibetan kdrikds, the Bhdsya, and the text of the Vydkhyd\ in the appendix, a summary of the Lokaprajnapti and the Karanaprajnapti.
2. Cullavagga, xi. 1. 8. [For a more exact presentation,J. Przyluski, Concile de Rdjagrha, p. 311,345, 349].
3. Oldenberg, Buddha . . . , 6th edition, p. 202; Fr. trans. Foucher, 2nd edition, p. 177.
