¹ Cf. Mutsuddi, V.C., Outlines of Buddhism and how it differs from Hinduism (a talk given at a meeting of the East and West Fraternity, Chittagong, on 25.8.1945), published by BPS Kandy (no date given), pp. 18 & 22.Google Scholar He speaks of “transmigration of souls, regarded by Hindu thinkers as the necessary complement of a belief in the essential sameness of all souls … and in their ultimate reunion with the Parameshvara, the Supreme Soul”, while “according to Buddhism there is no permanent Ego, and the soul is not transmigrating from body to body”.

Also: Malalasekera, G. P., The Truth of Anattā (Wheel 94), BPS Kandy 1966, p 16:Google Scholar “…the personality, in which other systems of thought imagine the presence of a permanent spiritual principle, a self or soul (atta), is, from the point of view of the Buddha, only a bundle of elements or forces … The individual is entirely phenomenal … without any extra-phenomenal self or soul within him”.

² See his Foreword to de Sylva, Lynn A., The Problem of the Self in Buddhism and Christianity, London 1979 (1st, Colombo 1975), p. ix.CrossRefGoogle Scholar

³ Parrinder, Geoffrey, Indestructible Soul, The Nature of Man and Life after Death in Indian Thought, London 1973.Google Scholar This book summarizes the views of several Indian systems, but completely fails to clarify the issue of soul and human personality. It uses the term “soul” throughout without any attempt to define it while it clearly denotes different shades of meaning in different contexts. The author makes no attempt to present an analysis of the different teachings on personality in Indian systems with the accompanying problems of the universal and individual core of the personality and its various components.

⁴ For more details and references see my article “The Vedic concept of the human personality and its destiny” Journal of Indian Philosophy, vol. 5 (The Hague 1978), pp. 275–89.Google Scholar

⁵ Cf. Radhakrishnan, S., The Principal Upsaniṣads, London 1953, pp. 73–5Google Scholar (the Introduction).

⁶ cf. Radhakrishnan's note to CU 8,7,4, op. cit. p. 502.

⁷ Alex Wayman wrote an exhaustive paper on the theme of nāmarāpa: “A study of the Vedantic and Buddhist theory of Nāma-rūpa”, Indological and Buddhist Studies. Volume in Honour of Professor J.W. de Jong on his Sixtieth Birthday, ed. Hercus, I. A., Kuiper, F. B. J., Rajapatirana, T., Skrzypczak, E. R., Faculty of Asian Studies (The Australian National University), Canberra 1982, pp. 617–42.Google Scholar For rūpa, which he says is matter or its appearance, he suggests the rendering “formation” (p. 620) when it occurs as a part of the compound nāmarūpa. He also lists later definitions of the two terms in contrast(nāmakāya as against rūpakāya, p. 621). In connection with his discussion of what he speaks of as “two kinds of nāmarūpa” Wayman makes a speculative suggestion when he says at the end of his paper: “A rather exciting outcome of these researches is that in the Brahmanical as well as in the Buddhist portrayal of nāmarūpa, this constitutes a kind of dividing line between our commonplace world and the superior world of the gods or of yoga-success” (p. 634). I confess that it is not entirely clear to me what Wayman means by this remark, since it seems too vague, but I do accept that there is likely to be an issue behind it which would merit a further textual and philosophical analysis.

⁸ See PTS Dictionary under manas where references can be found.

⁹ The use of the term gandharva meaning a spirit departed from a previous life is testified to in TU 2,8,1. Here we find a list of categories of beings enumerated in an ascending hierarchical order and among them are mentioned manuṣya gandharvas (human spirits) and deva gandharvas (divine spirits); these are followed by fathers in their “long enduring worlds” and then by several categories of devas (gods) above them etc. In a similar list of beings in BU 4,3,33 both the categories of gandharvas are missing. This indicates that they were not established in the hierarchy as fully fledged categories of beings comparable with the others. Therefore the human and divine gandharvas can be regarded as spirits of those who have just departed from the human and deva world, respectively, and exist in a kind of intermediary state before proceeding further to be born in another world.

A case for a longer duration of the intermediary state of discarnate beings as a feature of the teachings on rebirth as expounded in the Sutta Piṭaka has been convincingly argued also by Peter Harvey in “The between-lives state in the Pali Suttas”, Perspectives on Indian Religion, Papers in honour of Karel Werner (Bibliotheca Indo-Buddhica 30), Delhi 1986, ed. Connolly, Peter, pp. 175–89.Google Scholar

In contrast, Collins, Steven, in Selfless Persons, Imagery and Thought in Theravāda Buddhism, Cambridge 1982, pp. 208–15,CrossRefGoogle Scholar completely failed to recognize the evidence in the Sutta Piṭaka for a longer duration of the intermediary being. He regards the gandhabba as a “popular language” equivalent of the later Theravāda concept of the “re-linking consciousness” (p. 212) which is a far-fetched and unfounded suggestion. The expression “re-linking consciousness” is a narrowly defined technical concept of a developing sectarian doctrine on the immediate rebirth following the death of a person which did not allow any scope for an intermediary existence. Such existence, however, was advocated by other early sects (cf. Bareau quoted by Harvey and the later corroboration of this line by teachings of the schools of the Tibetan Book of the Dead). In the context of the Sutta Piṭaka texts by themselves, whose analysis Collins neglects, constantly projecting into them later Theravāda interpretations, viňňāna after death has the wider meaning of a complex being in the sense of nāmakāya before it descends into a womb to become a nāmarūpa. Collins's interpretations do, of course, reflect quite correctly the Theravāda position from whose point of view he wrote his thesis as expressed by its subtitle. Where he is wrong is when he reads the Sutta Piṭaka in the light of Theravāda orthodoxy as interpreted in the Abhidhamma and commentaries, regarding even the late Buddhaghosa's Visuddhimagga, the main compendium of Theravāda orthodoxy, as spelling out explicitly what is contained implicitly in the original Sutta Piṭaka (p. 22), a stance fully embraced in Theravāda circles of learned monks, but untenable on the level of academic scholarship, at least since Karl Seidenstücker published his Pali-Buddhismus in Übersetzungen (Breslau 1911)Google Scholar in which he “allowed the texts of the Canon themselves to speak without the rather dubious help of later commentaries” and maintained that “we have to tackle the oldest sources” with our own understanding as best we can, thus “avoiding the risk of adopting the position of a particular school by trusting later exegetical interpretations” (p. X). Which does not, of course, rule out careful consideration and evaluation of later views. Seidenstücker's stance has since been adopted by most scholars of Buddhism as well as Buddhist thinkers.

¹⁰ The nature of the tathāgata”, pp. 47–9 Buddhist Studies—Ancient and Modern, ed. Denwood, Philip and Piatigorsky, Alexander, London 1983, pp. 35–52.Google Scholar

¹¹ Harvey comes dangerously close to this position when he concludes from the quoted passage “who sees me, sees the dhamma é “that since seeing the dhamma also means seeing nibbāna, then “seeing” nibbāna is the same as “seeing” the Tathāgata. This suggests the identity of the two and that the Tathāgata is “dhamma-body” because he is nibbāna (op. cit. p. 45). However, he also says that the fact of the tathāgata's viňňāna “being ‘unsupported’, not dependent on anything, is nibbāna, before and after death”. (Ibid.) And he further contends: “When he enters samādhi to experience nibbāna, his viňňāna becomes unsupported and without object, thus being nibbāna. This unsupported viňňāna has the same characteristics as the Tathāgata and can be seen as his inner nature” (p. 50). This somewhat modifies his previous statement and it makes his idea of the tathāgata seem a little closer to the concept of the ultimate status of tathāgatas as envisaged by me, although to be sure about his position, a clearer formulation of the problem by him would be necessary.

¹² Johansson, Rune E. A., The Psychology of Nirvana, London 1969.Google Scholar Peter Harvey made a further case for the continuation of the unsupported viňňāna in nibbāna in his paper “Consciousness and nibbāna in the Pali Suttas”, Journal of Studies in Mysticism, vol 2, no. 2 (1979), pp. 70–82,Google Scholar whose slightly revised version will appear as “Consciousness and mysticism in early Buddhism” in the collection of papers The Yogi and the Mystic. Studies in Indian and Comparative Mysticism, ed. Werner, Karel, due from Curzon Press, London, in 1988.Google Scholar

¹³ Cf. Conze, Edward, Buddhist Thought in India, London 1962, pp. 122–32.Google Scholar

¹⁴ Cf. Lamotte, E., Histoire du Bouddhisme Indien I, Louvain 1958,Google Scholar p. 171 ff; Warder, A.K., Indian Buddhism, Delhi 1970, p. 7;Google ScholarConze, E., op. cit., pp. 31–2Google Scholar (although more caution has to be exercised where Conze uses later commentaries to enhance his understanding of the “archaic” Buddhist teachings). Cf. also Pérez-Remón, Joaquín, Self and Non-Self in Early Buddhism, The Hague 1980,CrossRefGoogle Scholar p. v, 2 and the whole platform from which he undertook his investigations, namely the Sutta Piṭaka with only a limited use of the Vinaya Piṭaka and almost complete rejection of Theravāda interpretations.

¹⁵ Although a lot has been written on the theme of atta/anatta, the problem of its relation to the question of personality, personal identity and continuity has not usually been understood in its full context. Most polemics centre around the question of the existence or non-existence of the atta as the self without fully discussing the question of its individuality and/or universality and the implications of one or the other position for personhood and its continuity within and beyond samsāra. Basically, there are three main positions assumed by the participants in the creation of the vast literature on the subject. Two of them could conveniently be called anattavāda and attavāda respectively. The first is that of the Theravāda tradition which emphatically denies the existence of atta, and its advocates interpret relevant Sutta Piṭaka passages in a way that suits their stance. As we have seen, they are followed by some modern scholars who think that since the oldest texts have been preserved by the Theravāda tradition, this tradition understands them best. The second is that of some modern scholars worried by the discrepancy between denying a permanent self and asserting transmigration and salvation who in turn interpret and translate the same passages to suit this opposite view. This can be demonstrated on a passage from the Alagaddūpama Sutta (M 22, PTS I, p. 138) or the Discourse on the parable of the water snake:

Attani vā bhikkhave sati attaniyam-me ti assâti.—Evam-bhante.—Attaniye vā bhikkhave sati attā me ti assati.—Evam bhante.—Attani ca bhikkhave attaniye ca anupalabbhamāne yam-p’ idem, diṭṭhiṭṭhanam: so loko so attā, so pecca bhavissāmi nicco dhuvo sassato avipariṇāmadhammo, sassatisamaṁ thath’ eva thassāmîti, nanāyaṁ bhikkhave kevalo paripūro bāladhammo ti.

Nyānapoṇika Thera, a leading Theravāda scholar-monk, uses the passage to show that there is a direct denial of atta in the discourses of the Buddha (Anattā and Nibbāna, Wheel Publications 11, BPS Kandy 1952, p. 20Google Scholar) and translates it as follows: “If, bhikkhus, there is a self, will there also be something belonging to a self?”—“Certainly, Lord.”—“If there is something belonging to a self, will there also be (the view) ‘My self?”—“Certainly, Lord.”—“But since, bhikkhus, a self and anything belonging to a self cannot be found, is it not a perfectly foolish doctrine to hold the point of view “This is the world. This is the self. Permanent, abiding, eternal and immutable shall I be after death, in eternal identity shall I persist’?”

Horner, I. B. is well known for her view that “atta is not denied. … anywhere … in the Pali Canon; it is accepted” (The Middle Length Sayings III, PTS 1959, p. XII)Google Scholar and translates the passage accordingly (Ibid. I. PTS 1954, p. 177): “If, monks, there were Self, could it be said: ‘It belongs to my self’?” “Yes, Lord.” “Or, monks, if there were what belongs to Self, could it be said: ‘It is my self’?” “Yes, Lord.” “But if Self, monks, and what belongs to Self, are incomprehensible, is not the view and the causal relation that: ‘This is the world, this is the self, after dying I will become permanent, lasting, eternal, not liable to change, I will stand fast like unto the eternal’—is not this, monks, absolute complete folly?”

Here is my own translation: “Were there, monks, a self, could one say: ȘThere is something selflike in me’?”—“Just so, Lord.”—“Were there something selflike in me, could one say ‘My self’?”—“Just so, Lord.”—“As self and something selflike cannot be made out, is not the point of view: ‘This is the world, this is the self; that, having passed away, I shall be; permanent, stable, eternal, a changeless entity; I shall stand the same in eternity,’ monks, a wholly and completely foolish doctrine?”

The remarkable difference in the translation by I.B. Horner of the underlined phrase shows to what length the modern attavādins are sometimes prepared to go to force their point. But there is nothing in the passage in support of Nyānapoṇika's interpretation, either. To avoid the two pitfalls, the passage has to be understood in relation to the five khandhas, none of which can be pinpointed as being ultimately the self, my self or selflike or belonging to my self, although each one of them may be referred to, on occasions, as oneself, one's own or belonging to oneself. This evasive and shifting nature of the experience of oneself simply proves that the self cannot be truly and reliably made out (anupalabhamāne). By taking this phrase as a proclamation of the non-existence of atta Nyānapoṇika contradicts himself, since virtually the same phrase is used with respect to the tathagata. In fact, he himself quotes it later (p. 22): “…the Perfect One, Anurādha, cannot truly and really be found … .” (tathāgate anupalabbhyamāne—S XLIV,10,2, PTS IV, p. 384). To admit the non-existence of the tathāgata would be, of course, heresy.

The third stance is represented by a variety of views whose common denominator is the endeavour to reconcile in some way the apparent contradiction of the other two. The most comprehensive attempt along these lines has so far been that of Pérez-Remón (op. cit.) whose thesis is that “The doctrine of anatta can co-exist with the reality of atta” (p. 153). While most of his arguments for the reality of atta successfully refute the interpretation of the anatta statements in the Sutta Piṭaka as an absolute denial of the self, they still fail to convince that the Suttas did teach its absolute reality as a metaphysical entity. In fact, almost all of them could be modified to show the inevitable continuity of the personality of one who has overcome saṁsāric conditioning and reached nirvāṇic status irrespective of whether he is still using the phenomenal shell of his personality or not, i.e. whether he is “alive” or has passed away.

All this suggests that the problem of the ultimate existence of the ātman/atta in the Upaniṣads and early Buddhism may be only a question of semantics. It can no doubt be regarded as an important and interesting philosophical question worth a thorough analysis in its own right. That would, however, be a separate undertaking beyond the scope of this paper.