Aspects of Early Buddhism

K.R. Norman

Cambridge

Panels of The VIIth World Sanskrit Conference, General Editor: Johannes Bronkhorst, Vol. II, Early Buddhism and Madhyamaka, Edited by: David Seyfort Ruegg & Lambert Schmithausen.

(C) 1990 by E.J. Brill, Leiden, The Netherlands

Cambridge - 1990

As is well known, the attā is specifically denied as a permanent entity in Theravāda Buddhism, although the word is of course widely used in Pāli 1 in the everyday sense of “oneself”. The question then arises: If there is no permenant attā, then what transmigrates in the course of rebirths in saṃsāra ? In the Mahātaṇhāsañkhayasutta of the Majjhimanikāya 2 we read of the bhikkhu Sāti, who so misunderstood the Buddha’s teaching that he thought it was viññāṇa “consciousness” which continued in saṃsāra (tad ev’ idam viññāṇaṃ sandhāvati samsārati, anaññam). 3 This would appear to be a recollection by Sāti of some such statements found in the Bṛhadāraṇyaka Upanṣad that vijñāna continues: idam mahad bhūtam anatam apāram viñānaghana eva 4, “This great being, endless, unlimited, consisting of nothing but intelligence”; sa vijñāno bhavati, sa vijñānam evānvavakrāmati 5, “He becomes one intelligence; what had intelligence departs with him”; sa vā eṣa mahān aja ātmā yo ‘yam vijñānamayah prāneṣu 6, “Verily, he is the great unborn Self who is this (person) consisting of knowledge amongst the senses.” Radhakrishan’s note on Bṛhadāraṇyaka Upaniṣad IV.4.1 states that “the principle of intelligence (vijñāna), after having absorbed all functions of consciousness, proceeds to continue in a new life” 7

When Sati’s view was made known to the Buddha, he refuted it by pointing out that he had frequently taught that aññartra paccayā n’ atthi viññannassa sambhavo 8 “Apart from condition there is no origination of consciousness”. He rejected the idea of a permanent viññāṇa which could transmigrate, by stressing the place of viññāṇa in the twelve-fold chain of the pațicca-samuppāda “dependant origination”, where viññāna is caused by sañkhāras “compounded formations” or “conditioned things”, and is itself the cause of nāmarūpa “name and form”. According to the account of the Buddha’s bodhi in the Vinaya-piṭaka, the Buddha examined the twelve-fold pațicca-samuppāda backwards and forwards immediately after bodhi 9, but elsewhere we read of the Buddha rehearsing a shorter form with only ten links, before his bodhi 10, and the longer chain is probably only a later extension of an earlier idea. The Pāli commentators analyse the twelve-fold version as being spread over three existences, 11 but it seems more likely that it was in its original formulation a simple empirical assertion, with no reference to more than one birth. There are, in fact, many other examples of chains of cause and effect mentioned in the Pāli canon, 12 and it is probable that the pațicca-samuppāda, of however many links, was simply a development of earlier, less elaborate, statements of conditionality. There is no reason to suppose that the Buddha was the first to think of a cause and its effect. 13

We may deduce that the pațicca-samuppāda chain was first reasoned out by the Buddha in what we may describe as reverse order, in the way in which the ten-fold chain mentioned above is given, starting from the end i.e. the position in which the Buddha found himself. He was alive and destined to suffer old age and death (jārāmarna) like other people who were alive. He was so destined because he had experienced birth (jāti); birth is caused 14 by existence (bhava); existence is caused by clinging (upādāna); clinging is caused by craving (tanhā); craving is caused by feeling (vedanā); feeling is caused by contact (phassa); contact is caused by the six senses (salāyatana); the six senses are caused by name-and-form (nāmarūpa); name-and-form are caused by consciousness (viññāṇa). This is the starting point of the ten-fold chain. In the twelvefold chain consciousness is caused by the compounded formations (sañkhāra); the compounded formations are caused by ignorance (avijjā). The destruction of any link in the chain would lead to the destruction of any links dependant on it. The destruction of craving would lead to the destruction of clinging, existence, birth and old age and death. The destruction of avijjā would lead to the destruction of the whole chain of conditional origination, and therefore to the end of continued existence in samsāra. The destruction of avijjā by vijjā would therefore lead to nibbāna, which was release (mokkha) from samsāra. 15

The various accounts of the Buddha’s bodhi, which led to his nibbāna, are not easy to reconcile together, since they appear in different forms in different parts of the Pāli canon, with quite large omissions and changes of emphasis in some versions. The shortest account is that found in the Ariyapariyesanasutta of the Majjhimanikāya, 16 and for this reason some scholars believe that this is the earliest account available to us. 17 It concentrates upon the gaining of nibbāna, but does not give any information about how it was attained. We read that after the Buddha had attained nibbāna, knowledge ( ñāṇa) and insight (dassana) arose in him that his release was unshakable, 18 that this was his last birth, and that there would be no renewed existence (punabbhava) for him. 19

We may assume that in the shortest account of his bodhi the Buddha would deal with the most important part of the experience, and we can therefore see that this was the gaining of nibbāna. This view is supported by the fact that when he visited other teachers, before his bodhi, he found their teachings inadequate because they did not lead to nibbāna (nāyam dhammo nibbidāya na virāgāya na nirodhāya na upasamāya na abhiññāya na sambodhāya na nibbānāya samvattati). 20 We may deduce from this that the concept of the attainment of nibbāna existed, even though the Buddha (while Bodhisatta), and his teachers, were unable to achieve it. We may also deduce that the words in the Buddha’s statement are in the order in which the various states mentioned in it are to be realised, starting with disgust with the world, and going on to sambodhi and nibbāna. This bears out the belief the Buddha’s aim to free himself from samsāra, and all aspects of his teaching were concerned with the acquisition of means to do this, either in this life or a later one, and with finding how best to dwell in samsāra until release was obtained.

The account of the Buddha’s bodhi given in the Mahāsaccakasutta of the Majjhimanikāya, 21 however, gives more information. We read there that the Buddha realised that the various efforts he had made so far were not productive, and he wondered if there was another way to bodhi (siyā nu kho añño maggo bodhāya). 22 He recalled an experience in his boyhood, when he had by chance entered into the first jhāna. He therefore entered into the first jhāna again, and form there moved into the second, third and fourth jhānas. 23 At that point he gained, in order, three knowledges (ñāṇas). The first was the knowledge of his previous existences; the second the knowledge of the arising and passing away of others, and their fates which depended upon their actions (kamma); the third was the knowledge of the destruction of the āsavas. He understood the existence, arising, stopping and path to the stopping of misery (dukkha), and then the existence, arising, stopping and the path to the stopping of the āsavas. He knew that he was released, and that birth has ended. There is no indication of how exactly release was obtained, and it may simply be that gaining knowledge of the destruction of the āsavas was the destruction itself, i.e. the knowledge was efficacious, and the bodhi was the nibbāna. It is perhaps belief in such a view that has led Collins to translate nibbāna as both “enlightenment” and “liberation”. 24

The translation “enlightenment” is normally reserved for bodhi or sambodhi, but it is somewhat misleading in that the root budh- which underlies these words has no direct connection with “light”. The root means literally “to wake up”, or metaphorically “to wake up (to a fact), to know it”, and “awakening” would be a more literal translation of bodhi. The past participle Buddha is used actively to mean “one who has awakened, one who has gained knowledge”. In the Ariyapariyesanasutta account the Buddha refers to his dhamma as being duranubodha 25 and na … susambudha, 26 and this implies that his bodhi consisted of gaining that dhamma, i.e. the knowledge of how to gain release. This accords with Buddhaghosa’s statement: uparimagga-ttayasañkhātā sambodhi 27 “sambodhi is synonymous with the three higher paths (leading to arahat-ship)”.

In the account in the Vinaya-piṭaka, the Buddha specifically states that he gained bodhi (anuttaraṃ sammāsambodhiṃ abhisambuddho) 28 when knowledge and insight (ñāṇadassana) arose in him in respect of the four noble truths about misery, but in the account of his bodhi in the Mahāsaccakasutta the four statements about misery are not referred to as noble truths, 29 and since they appear to be subordinate to the four statements about āsavas, it is possible that the statements about misery are a later addition, which led to a parallel, but inappropriate, set of four statements being evolved about the āsavas, to provide symmetry. 30 If this is so, then something similar is probably true of the Buddha’s statement about the point at which he became Buddha. It should rather have been the point when he gained knowledge about the destruction of the āsavas.

In the Sāmañnaphalasutta of the Dīgha-nikāya, 31 the Buddha sets out the advantages of life for a samaṇa. These culminate in the practice of the four jhānas, leading to the three ñāṇas. In the account of the third ñāṇa, that of the destruction of the āsavas, we find the same insertion of the four statements about misery, once again not called noble truths. There then follow the four statements about the āsavas which lead on to the destruction of the āsavas and the attainment of arahat-ship. 32 This is therefore a repetition of the Buddha’s own experience as related in the Mahāsaccakasutta, and again we may suspect the presence of the statements about misery.

It is noteworthy that when the Buddha begins to teach, he preaches the news about the four noble truths about misery, not about the āsavas. As part of the fourth noble truth he teaches the eight-fold path leading to the destruction of misery (dukkha-nirodha). The stages of the path are: sammā-ditṭhi, -saṃkappo, -kammanto, -ājīvo, -sati, -samādhi. 33 This path is said to have been learned by the Buddha, and to lead to nibbāna (cakkhukaraṇī ñāna-karaṇī upasamāya abhiññāya sambodhāya nibbānāya saṃvattati). 34 We are therefore presumably to regard dukkhanirodha and nibbāna as synonymous. 35 The path does not include any reference to the four jhānas, although it is possible that in the final element, sammā-samādhi “right concentration”, could be interpreted as including them. 36 If it does not, then the way to nibbāna along with the eight-fold path is a means which differs somewhat from the way in which the Buddha himself gained nibbāna. 37

Elsewhere in the Pāli canon, however, there is a list of the stages of an asekha, i.e. an arahat, who has finished his training and is now an adept. This path adds two further stages, sammā-ñāṇa and sammā-vimutti, to the usual eight. 38 This is an extension, rather than a contradiction, of any other teaching. The Buddha states that the eight-fold path leads to nibbāna. Nibbāna (= vimutti) must therefore be a further stage upon the path, and the knowledge required to gain nibbāna must be the preceding stage. When one is an asekha, then one has practised the eight-fold path, gained sammā-ñāṇa, and then sammāvimutti.

And just noted, the Buddha states that the eight-fold path leads to nibbāna. He uses dukkha-nirodha and a synonym of this, but does not speak of the destruction of the āsavas, which, as we have seen, is also a synonym of nibbāna. Schmithausen suggests that the question of dukkha is prior to the question of the āsavas,39 but this is not necessarily so. It is clear that in early Buddhism the concept of the āsavas and their destruction was of great importance. The most common epithet of an arahat is khīnāsava “one whose āsavas are destroyed”, not “one whose craving or ignorance is destroyed”. The list of the āsavas which we find in the Pāli canon is: kāmāsava “lust”, bhavāsava “becoming”, and avijjā-sava “ignorance”. 40 To these diṭtāsava “wrong view” is sometimes added. 41 It is clear that if this list is correct, then the āsavas as a whole are not part of the paṭicca-samuppāda, although avijjā and bhava are there as separate items, and it might be possible to take kāma as equal to tanhā.

It is to be noted that the etymological meaning of the word āsava “influx”, the use of the terms āsava and anhaya by the Jains, and the use of the related word āsinava by Aśoka, suggest, as was proposed by Alsdorf, 42 that the usual use of the word by the Buddhists is probably not the original usage. This view is also supported by the fact that the four āsavas in this list are identical with the four oghas, 43 suggesting that substitution has taken place at some time. Schmithausen points out 44 that the āsavas are glossed in pāli as vighāta-pariḷāha, which would give a meaning something like “afflictions”. In the Sammādiṭthisutta of the Majjhima-nikāya 45 the eight-fold path is said, unusually, to lead to the destruction of the āsavas.46 This might be a relic of an earlier theory, but it is said in such a way that āsava might almost be taken as the equivalent of dukkha, which perhaps further explains the parallelism between the āsavas and dukkha in the account of the Buddha’s bodhi.

It is possible that, whatever the original meaning of the word āsava, it was noted that the destruction of the āsavas led to the destruction of kāma, bhāva and avijjā, which might have suggested the identification of the āsavas with the things which were destroyed at the same time. If, however, we believe that the āsavas are to be identified with any one link of the twelve-fold paṭicca-samuppāda, then the most likely candidate for identification would be the sañkhāras. It is not impossible that there was some earlier meaning of āsava which was approximately synonymous with that of sañkhāra. It is to be noted that the cause of the āsavas is avijjā.47 which is also the cause of the sañkhāras. We read that the stopping of avijjā leads to the stopping of the āsavas. This does not prove that the āsavas and the sañkhāras are the same, but it does show that the āsavas cannot be avijjā, for they can scarcely be their own cause, as is pointed out by Schmithausen. 48 It may be that avijjā and the sañkhāras were originally a separate cause-and-effect, which were prefixed to the chain of causation beginning with viññāṇa. It seems possible that at some early stage of Buddhist thought there was a view that the āsavas were very similar in effect to the sañkhāras, the active “formulating factors”, or “formative influences” or “karmic formulations”, as Nyanatiloka translates. 49 In the individual there was the passive version of the sañkhāras, the “formed factors”, as one of the group of khandhas. The idea of the active sañkhāras as the karmic formations, and then the passive formed sañkhāras as part of the individual, would not be inconsistent with the idea of āsava in Jainism as the process by which kamma flowed into the soul.

If this was so, then it is possible that the older meaning of āsava was forgotten in Buddhism 50 when the emphasis in the Buddha’s teaching was placed upon the idea that the world was dukkha. This may have been the result of the change of emphasis from what has been called the jhānic side of Buddhism, where stress was on jhāna “meditation” as a means of gaining nibbāna, i.e. the destruction of the āsavas, to the kammic side of Buddhism, whereby the emphasis was on the entry into the stream, whereby the entrant could hope, by successfully following the teaching, to rise higher and higher in successive rebirths towards the goal of arahat-ship. In these circumstances, the main need was to convince followers that the world was dukkha, but that there was a way of release from it which did not demand special ability in meditation. The fact that nibbāna of mokkha could be attained in various ways led to a situation where there was different terminology employed to denote what was basically the same concept. So one who had gained arahat-ship could be described as khīnāsava, nibbuta, or dukkhassa antakara “one who has put an end to misery”. We also find references in the Pāli canon 51 to those who have put an end to misery by breaking the seven fetters (saṃyojanāni). Since these fetters include ignorance (avijjā) and lust for existence (bhavarāga, which is perhaps a synonym for tanhā), it may be that there is no inherent contradiction between this teaching and the idea of breaking the chain of dependant origination by destroying one of the links.

There is an interesting point which arises in connection with the four jhānas which the Buddha practiced at the time of his bodhi. As noted above, one version of the occurrence relates that the Buddha recalled a boyhood experience in which he had entered upon the first jhāna. Repeating his boyhood experience, he entered the second, third and fourth jhānas. We have, however, an account of the Buddha’s pre-bodhi visits to two teachers, Āḷāra Kālāma and Uddaka Rāmaputta. With these teachers he practiced meditation and reached with them the states of ākiñcaññāyatana52 “the state of nothingness” and nevasañña-nāsaññāyatana 53 “the state of neither perception nor nonperception”, respectively. As already stated, he rejected both of these as not leading to nibbāna, but in his own teaching after his bodhi he included them as stages on the way to nibbāna. If, as taught by the Buddha, they are the third and fourth of the arūpa-jhānas, which are the seventh and eighth of the samāpattis “attainments”, since they come after the four rūpa-jhānas, then the Buddha had already attained the first four samāpattis with those teachers before he gained the seventh and eighth. We have the statement of the commentator Buddhaghosa to this effect. 54 This would make the story of his boyhood memory seem very strange, and we should perhaps follow the view that the four rūpajhānas and the four arūpa-jhānas were originally two quite separate sets of states of meditation. 55

In the Buddha’s accounts of the eight samāpattis, however, we read of a ninth state, that of saññāvedayitanirodha 56 “cessation of feelings and perceptions” or “cessation of feeling 57 of perceptions”. In this state, for one with seeing with perceptive knowledge, the āsavas are destroyed (paññāya c’ assa disvā āsavā parikkhīnā honti). 58 This would seem to imply that, if we equate āsavakkhaya with nibbāna, this was another way of attaining nibbāna, and Schmithausen quotes Nagasaki 59 as believing that saññā̄vedayitanirodha and nibbāna were originally identical. It is not entirely clear, however, how one could see by knowledge when in such a state, and it is possible that the seeing with paññā refers to something which happens after attaining this ninth state, not while one is in it. Without further information about the nature of pañña it is difficult to come to any firm conclusions about this, but if we equate pañña with ñāna, then this could be another reference to bodhi leading to nibbāna.

We must, however, note that there is no reference to the four arūpa-jhānas in the accounts of the Buddha’s own attainment of nibbāna at the time of his bodhi. In the story of his death, in the Mahāparinibbānasuttanta of the Dīgha-nikāya, 60 we read that the Buddha went through all the stages of the rūpa-jhānas and the arūpa-jhānas, and then entered saññāvedayitanirodha. He was then thought by Ānanda to have attained nibbāna. 61 Anuruddha, however, pointed out that he had only attained saññāvedayitanirodha. 62 From there the Buddha went back, in due order, to the first jhāna, and then up to the fourth jhāna, from which he died, and presumably attained nibbāna. It is therefore noteworthy that it was also from the fourth jhāna that the Buddha gained bodhi and nibbāna on the earlier occasion, and it may be relevant that in the Saṃyutta-nikāya 63 we read of a bhikkhu going from the fourth jhāna to saññāvedayitanirodha, 64 passing beyond nevasaññā-nāsaññāyatana, without any mention of the other arūpa-jhānas.

It would seem from the account of the Buddha’s death that saññāvedayitanirodha was probably some sort of death-like trance, and we may wonder how Anuruddha, seeing the Buddha in this condition, nevertheless know that he was not in nibbāna. It is clear that as far as Anuruddha was concerned the state was not identical with nibbāna, but it does not reveal how anyone in this death-like trance could make use of pañña to attain nibbāna. It may well be that the statement that the Buddha was in saññāvedayitanirodha was merely the result of later theorising. This perhaps supports the suggestion that it was after gaining saññāvedayitanirodha, not while one was in the state, that one was able to use pañña and gain the destruction of the āsavas.

The object of the Buddha’s teaching was to gain release from the beginningless and endless samsāra. There is a reference in the Pāli canon to two varieties of release in nibbāna. 65 One is attained in life and is called the element of nibbāna with a remnant of clinging (sa-upādisesā nibbānadhātu). In this the defilements (kilesas) are destroyed, and lust, hatred and delusion (rāga, dosa, moha) are annihilated. The remainder of physical life is perfect bliss and peace. The second form of nibbāna is that without a remnant of clinging (anupādisesā nibbānadhātu). It coincides with death, and is not followed by rebirth, for the elements of existence (khandhas) have been destroyed.

The descriptions of nibbāna in the Pāli canon are set out in very general terms, and it is often defined in terms of negatives of opposites. It is “blissful” (siva) or “happy” (sukha) as opposed to the dukkha of existence. It is “unmoving” (acala) as opposed to the endless movement of samsāra. It is “undying”66 (amata) as opposed to the repeated deaths of saṃsāra. It is “unborn” (ajāta), “unoriginated” (abhūta), “uncreated” (akata), and “unformed” (asainkhata) as opposed to the world, which is born, originated, created and formed. The last named epithet is the most important, for in Theravāda Buddhism nibbāna is the only asańkhata thing. 67

Buddhism denied the existence both of a permanent soul and a permanent individuality. An individual is merely a group of five “elements of existence” (khandha), 68 “form” (rūpa), “feeling” (vedanā), “perception” (saññā), “mentalformations” (sañkhāra) and “consciousness” (viññāna). If the “compounded formations” (sañkhāra), the second link in the chain of dependant origination, are destroyed because their “ignorance” (avijjā), is destroyed by vijjā, then all compounded formations, including the passive “mental formations” (sañkhāra) and other khandhas which go to make up the individual are destroyed and we are left only with the “uncompounded” (asañkhata), i.e. nibbāna, which is outside saṃsāra.

In these circumstances it is not surprising that the condition of being nibbuta or in nibbāna cannot be defined. The word nibbuta is also used of a fire which has gone out. Schrader long ago pointed out the Indian belief that an expiring flame does not really go out: vahner yathā yonigatasya mūrtir na dṛśyate naiva ca lingganāśah 69 “as the form of a fire … is not seen nor its seed destroyed”. So it is with an individual who has gained nibbāna. His state cannot be described any more than the state of a fire which has gone out can be described. The only thing that is certain is that, because nibbāna is “not-self” (anatta), it cannot be reconciled with the views of those who think that the object of religious exertion is to re-unite the individual soul with Brahman or Ātman.

Footnotes

  1. Abbreviations of title of Pāli texts are as in the Epilegomena to the Critical Pāli Dictionary, Vol. I, Copenhagen, 1924-48. References are to the editions of the Pali Text Society.

  2. M I 256-71.

  3. M I 256, 19-20.

  4. II.4.12

  5. IV.4.2

  6. IV.4.22

  7. S. Radhakrishan, The principle Upaniṣads, London 1953, p. 270.

  8. M I 258, 20.

  9. Vin. I 1.

  10. S II 104.

  11. See S. Collins, Selfless Persons, Cambridge 1982, p. 108.

  12. The paṭicca-samuppāda is described in the Pāli Canon with varying lengths and starting from various points. This probably represents its use in different contexts of teaching or instruction, i.e. in his sermons the Buddha would sometimes start from a specific cause or arrive at a specific effect in the chain, for various doctrinal reasons.

  13. If the Sn does really contain pre-Buddhist material, then it is possible that some of the statements of cause-and-effect which occur there are older than the Buddha. Despite Aramaki (N. Aramaki, “On the formation of a short prose Pratityasamutpāda sūtra”, in Buddhism and its relation to other religions (Festschrift for Dr. Shozen Kumoi), Kyoto 1985, 87-121), there is no reason for thinking that the earliest presentation of the paṭicca-samuppāda was in verse, although it may be that the oldest literary version we have is in a verse text.

  14. Or “because of” (-paccayā). The meaning is that one link in the chain is a pre-requisite for the next, e.g. birth is a pre-requisite for death: a person cannot die unless he has been born.

  15. These are simply different analyses of the same problem, i.e. suffering, old age, death, etc. which involves tracing them back to a cause. The Buddha then seeks a way to destroy that cause. Since the method of destruction depends upon knowing how to do it, the starting point for destruction must be knowledge (vijjā). It therefore follows that the starting point of saṃsāra must be avijjā.

  16. M I 160-75.

  17. See L. Schmithausen. “Liberating Insight and Enlightment in Early Buddhism”, in Studien zum Jainismus und Buddhismus (Gedenkschrift für Ludwig Alsdorf), Hamburg 1981, p. 207, quoting A. Bareau.

  18. akuppā me vimutti. In other versions we find the phrase vimuttasmiṃ vimuttam iti ñānaṃ hoti (for references see Schmithausen, op. cit. (in n. 17), p. 219 n. 69). In some contexts this can be taken as referring to the word cittaṃ, which precedes it, and the phrase can be translated, “In (it) released there is the knowledge ‘(I am) released’”. In some contexts, however, it is not cittaṃ but ariyasāvako which precedes vimuttam. Various explanations of this are possible: it is perhaps a pericope, with the standard phrase being used in a context where it is not appropriate; since there is no verb to indicate who or what is released, it is possible to take vimuttam impersonally, meaning “(it is) released” (see T. Vetter, “The most ancient form of Buddhism”, in the Festschrift for Dr. Shozen Kumoi (see n. 13), pp. 67-85, (p. 70)). In such a context, however, it might seem preferable to take vimuttam as an example of a past participle being used as the equivalent of an action noun: “(there is) release”. Another explanation would be to assume that in an earlier form of this phrase in an eastern dialect the reading was vimutte, which could be either masculine or neuter. When this was converted into a Western dialect form the redactors had a choice between vimutto and vimuttam, and having decided in the majority of occurrences to have vimuttam, to agree with cittaṃ, this was then extended to all occurrences, even though it was not appropriate with ariyasāvako. There is an extended version of the phrase including a verb: vimuttam vimutt’ amhī ti ñānam hoti, and the fact that this is not merely a scribal error in the Pāli tradition is proved by the (Mūla-?)Sarvāstivādin reading vimukto ‘smīti. This form of the past participle with the case ending elided, was applicable to both a masculine and a neuter subject. Schmithausen (ibid.) thinks that the form with amhī is a later reading, but the sandhi formation whereby -e or -o is elided before a-, rather than a- being elided after -e or -o, is not common in Pāli and seems to be early. I would suggest that this is, in fact, the earlier form of the phrase. I suspect that in some branches of the scribal tradition vimutt’ amhī ti was misunderstood as vimuttaṃ hī ti, form which hi was dropped as an unnecessary particle, leaving a neuter form vimuttam iti even when the context demanded a masculine.

  19. M I 167, 27-29

  20. M I 165,10-12=166,29-31.

  21. M I 237-51.

  22. M I 246, 35 .

  23. It is debatable whether the Buddha actually went through four stages of meditation as set out in the texts. It seems more likely that there was a single developing state of meditation, which (when he came describe it to his followers) could conveniently be broken down into four states. The problem which Vetter raises (Vetter, op. cit. (in n. 18), p. 80), is best explained by saying that the Buddha was trying to put into words something ineffable which had happened to him, and his words are really a later rationalisation (perhaps by his followers) of the irrational. All his views are simply ways of describing different aspects of the same experience, and are complementary, not contradictory. These various rationalisations are not necessarily of different dates, since they may be products of different environments and (preaching) needs.

  24. Collins, op. cit. (in n. 11), index, p. 319, s.v. nibbāna.

  25. M I 167, 31 .

  26. M I 168, 6 .

  27. Sv 313, 4.

  28. Vin I 11, 28.

  29. For a discussion of the formulation of the four noble truths see K.R. Norman, “The four noble truths: a problem of Pāli syntax”, in Indological and Buddhist Studies (Festschrift for J.W. de jong), Canberra 1982, pp. 377-91.

  30. See Schmithausen, op. cit. (in n. 17), p. 205.

  31. D I 47-86.

  32. D I 85

  33. Vin I 10, 20-23.

  34. Vin I 10, 24-25.

  35. See Schmithausen, op. cit. (in n. 17), p. 214.

  36. We have to assume that at the very beginning of Buddhism, i.e. at the level of the Buddha’s own personal experiences, the meaning of technical terms was not rigorously fixed, and so samādhi could be used in the sense of jhāna, etc.

  37. In the exposition of the four noble truths, it does not seem to make sense to say, as Vetter does (op. cit. (in n. 18), p. 77), that the Buddha had actually practiced the fourth noble truth. Surly the Buddha had gained nibbāna by means of jhāna, but was holding out to others the possibility of following a different path leading to the destruction of dukkha?

  38. See D III 271, 5-9.

  39. See Schmithausen, op. cit. (in n. 17), p. 205.

  40. See M I 249, 14-15.

  41. See The Pali Text Society’s Pali-English Dictionary, s.v. āsava.

  42. See L. Alsdorf, Les études jaina: état present et táches futures, Paris 1965, p. 4.

  43. See Pali-English Dictionary, s.v. ogha.

  44. See Schmithausen, op. cit. (in n. 17), p. 248 n. ad 23.

  45. M I 46-55.

  46. M I 55.

  47. M I 55.

  48. See Schmithausen, op. cit. (in n. 17), p. 205.

  49. See Nyanatiloka, Buddhist Dictionary, s.v. saṅkhāra.

  50. The way in which the Buddha retained the use of the word āsava with a changed meaning was quite in keeping with his practice of taking over technical terms from other religions, but giving them new meaning, e.g. the term tevijja was explained as referring to three (special) types of Buddhist knowledge, not the brahmanical vedas. Having taken over the idea of the āsavas from the Jains, or some other source, he kept the term, with a changed meaning, even when it was not longer a matter of great doctrinal importance.

  51. A IV 7-8.

  52. M I 164,15

  53. M I 165, 35

  54. ākiñcaññāyatanapariyosānā satta samāpattiyo maṃ jānāpesi, Ps III, 22-23.

  55. The four arūpa-jhānas were not originally Buddhist, and that is why they were included in the nonBuddhist teachers’ views. If Bareau is correct in stating that the story of the Buddha being taught by these teachers has no historical basis, we must conclude that the inclusion of a mention of the arūpa-jhānas in the Buddha’s life history was intended to show that they were inadequate when compared with the Buddha’s method. They do, however, lead to a state which seems to be equal to nibbāna, which presumably means that some, at least, of these non-Buddhist teachers had succeeded in finding a way out of samsāra. It was presumably because the arūpa-jhānas were successful in gaining the desired end that they were incorporated into the Buddhist scheme of jhānas, not as simultaneous means (which would have been better, because they are really an alternative) but as consecutive.

  56. M I 165, 35.

  57. Although saññāvedayita is usually translated as a dvandva compound, this is not necessarily correct. Grammatically, it could as well be taken as a tatpuruṣa compound with the past participle vedayita being used as an action noun. This interpretation would depend upon the occurrence of sañña with the verb vedayati. This combination seems to occur in the Pāli canon as we have it now, but it is possible that it existed at an earlier date, when the precise signification of technical terms had not yet been fixed.

  58. M I 175, 3-4.

  59. See Schmithausen, op. cit. (in n. 17), p. 259 ch. H.

  60. D II 72-168.

  61. D II 156, 17.

  62. D II 156, 18-19.

  63. S V 215.

  64. See Schmithausen, op. cit. (in n. 17), p. 215 n. 54.

  65. See Nyanatiloka, Buddhist Dictionary, s.v. nibbāna.

  66. Vetter (op. cit. (in n. 18), p. 74) may not be correct in translating amata as “immortality”. This translation perhaps gives the wrong impression, since the Buddha was presumably trying to gain release from samsāra, i.e. he was trying to find a state where there was no rebirth, and therefore no dying leading to rebirth. For this reason nibbāna is described as being without birth, without death, without gati, etc.

  67. The reference in the Milinda-pañha (268,14 ; 271,11) to ākāsa being akammaja is taken to be due to Sarvāstivādin influence. See I.B. Horner, Milinda’s Questions, Vol. I, London, 1963, p. xviii.

  68. Cf. evaṃ khandhesu santesu, hoti satto ti sammuti, S I 135, 21.

  69. Śvetāṣvatara Upaniṣad I.13, quoted by F. Otto Schrader, “On the problem of nirvāṇa”, in JPTS 19041905, p. 167 n. 2.