Introduction

In the present chapter I study the self-cremation of Dabba, based on a translation of the Saṃyukta-āgama (T 99) parallel to two discourses in the Udāna that record this event. Following the translation I examine the fire imagery inherent in this tale, and briefly survey the motif of auto-cremation in Buddhist literature and religious practice. I then examine the possible genesis of the Dabba narrative, after which I evaluate the implications of this tale, as well as those of the Buddha's deliberate decision to pass away, within the context of the general ancient Indian attitude towards suicide. The discourse versions reporting Dabba's autocremation are as follows:

• two consecutive discourses in the Udāna,[n.1] Ud 8.9–10 at Ud 92,24 to 93,ult.
• a discourse in the Saṃyukta-āgama (T 99),[n.2] SĀ 1076 at T II 280b25 to 280c17.
• a discourse in the 'other' Saṃyukta-āgama (T 100).[n.3] SĀ2 15 at T II 378a28 to 378b16.

In what follows I translate the version found in the Saṃyukta-āgama (T 99).

---

Translation

[Discourse on Dabba][n.4] [2] The translated text is SĀ 1076 at T II 280b25 to 280c17, which does not provide a title (the parallel SĀ2 15, translated in Bingenheimer 2011: 86f, also does not have a title). Akanuma 1929/1990: 96 suggests the name of the protagonist, 陀驃, as a title, which I have followed; the Be and Ce editions of Ud 8.9 and Ud 8.10 are entitled the "first" and the "second" version of the "Discourse on Dabba".

Thus have I heard. At one time the Buddha was staying at Rājagaha in the Bamboo Grove, the Squirrels' Feeding Ground. At that time the venerable Dabba Mallaputta approached the Buddha, paid respect with his head at the Buddha's feet, withdrew to stand to one side, and said to the Buddha:

"Blessed One, I wish to attain final Nirvāṇa in front of the Buddha."[n.5] [3] In Ud 8.9 at Ud 92,29 he says: "it is time for my final Nirvāṇa, Well Gone One", parinibbānakālo me dāni sugatā ti.

The Blesssed One remained silent. In the same way [Dabba] informed [the Buddha] three times [of his intention]. The Buddha said to Dabba Mallaputta:

"What exists is conditioned by formations, its nature is just like that!"[n.6] [4] SĀ 1076 at T II 280b29: 此有為諸行, 法應如是. I am not sure I have been successful in rendering this cryptic formulation in SĀ 1076, which must be intending some sort of permission, otherwise Dabba would not have carried out his plan. In SĀ2 15 at T II 378b2 the Buddha replies: "enter Nirvāṇa, I do not hinder you", 汝入涅槃, 我不遮汝, and in Ud 8.9 at Ud 92,30 the Buddha also gives Dabba permission to do what he wants to do, yassa dāni tvaṃ, dabba, kālaṃ maññasī ti. A reference to formations can be found in a similar context in the Mūlasarvāstivāda Vinaya, T 1451 at T XXIV 397a10, where the Buddha gives Subhadra permission to pass away: "all formations are completely impermanent, you may know yourself the [proper] time for what is to be done", 一切諸行皆悉無常, 汝於所作自可知時.

Then, in front of the Buddha, the venerable Dabba Mallaputta entered concentration of such a type that, ascending into the air towards the east, he manifested the four postures of walking, standing, sitting, and reclining.[n.7] [5] Whereas Ud 8.9 does not report any other miracle, apart from his auto-cremation (cf. note 8 below), according to Ud-a 432,25 he had displayed all the miracles that are within the purview of a disciple, sāvakāsādhāraṇāni sabbāni pāṭihāriyāni dassetvā.

He entered into the fire concentration, so that the lower parts of his body emitted fire. His entire body blazed up with a radiance spreading in the four directions, [coloured] blue, yellow, red, white, and crystal colour. The lower parts of his body emitted fire and, while this was burning his body, the upper part of his body emitted water, which sprinkled over his body. Then the upper part of his body emitted fire, which was burning the body downwards, while the lower parts of his body emitted water, which sprinkled upwards over the body.

Having manifested these various kinds of miracles in each of the ten directions, being in mid-air he emitted fire from within his body that burnt his own body and he attained Nirvāṇa without remainder,[n.8] [6] Ud 8.9 at Ud 92,33 records that once Dabba had risen into space cross-legged, he passed away after "having attained the fire element", tejodhātuṃ samāpajjitvā; Ud-a 432,12 explains that the reference to attaining the fire element intends the attainment of the fourth absorption based on the fire kasiṇa. completely disappearing and being extinguished, so that no soot was left behind. Just like a lamp, burning in mid-air, whose oil and wick are completely extinguished,[n.9] [7] The simile in Ud 8.9 at Ud 93,4 is slightly different, as it compares Dabba's passing away without leaving behind ashes or soot to burning ghee or oil, in which case also no residues are left behind – a simile also found in SĀ2 15 at T II 378b10. in the same way Dabba Mallaputta attained Nirvāṇa in mid-air with his body and mind completely extinguished.

Then the Blessed One spoke in verse: "Just as a burning iron pellet,
That is blazing up in flames.
When its heat eventually comes to be extinguished,
No one knows whither it has returned.[n.10] [8] The corresponding stanza in Ud 8.10 at Ud 93,23 seems to be about a blazing spark that comes off a hammer, presumably used by a smith who is beating a heated piece of iron on an anvil; cf. also the parallel versions listed below in note 46.

So, too, liberated ones,
Cross over the mire of defilements,
Having cut off the torrents forever,[n.11] [9] Ud 8.10 at Ud 93,25 speaks of having crossed over the flood of the bondage of sensual desire, kāmabandhoghatārinaṃ; cf. also SĀ2 15 at T II 378b14, which refers to the mire of sensual desires, 諸欲淤泥.
No one knows where they have gone.
Having attained imperturbability,
They enter Nirvāṇa without remainder."[n.12] [10] Ud 8.10 at Ud 93,26 does not refer to the imperturbable, instead describing that they have attained a happiness that cannot be shaken, pattānaṃ acalaṃ sukhaṃ.

When the Buddha had spoken this discourse, hearing what the Buddha had said the monks were delighted and received it respectfully.[n.13] [11] Chung 2008: 211 lists Hoernle fragment 147/u.f. recto 1 as a parallel to the conclusion in SĀ 1076 at T II 280c16f. In the introduction to his edition of this fragment, Hartmann 1998: 357 points out that the case is problematic, as the fragment, Hartmann 1998: 358 R1, reads: [sya] bhāṣita[m abhi]na .[yā]numodya utthāyāsa[n] .. [r] and thus does not seem to correspond to the concluding section of SĀ 1076, according to which the monks rejoiced in what the Buddha had said, 諸比丘聞佛所說, 歡喜奉行. In fact the only protagonist of the discourse who according to the standard formulation of discourse conclusions could be portrayed as getting up from his seat – the one who has come to visit the Buddha – has self-cremated. Although the remainder of the fragment is a parallel to the next discourse in the Saṃyukta-āgama, SĀ 1077, translated above p. 123ff, in this case the sequence of discourses in the Saṃyukta-āgama preserved in Chinese translation (T 99) appears to differ from the sequence attested in the Sanskrit fragment.

---

Study

Besides a few minor variations, the main difference between the Saṃyukta-āgama discourse translated above and its parallels in the Udāna is the description of Dabba's performance of various magical feats, among them the so-called twin miracle of simultaneously producing fire and water from different parts of one's own body. Whereas the Theravāda tradition attributes the ability to perform this miracle only to the Buddha,[n.14] [12] Jā I 77,21 indicates that the Buddha performed the twin miracle four times, once right after his awakening, once when visiting his relatives in Kapilavatthu and twice in reply to being challenged by non-Buddhist wanderers. A miracle performed in reply to one of these two challenges is recorded in DN 24 at DN III 27,12, although the description only indicates that the Buddha had attained the fire element, tejodhātuṃ samāpajjitvā, rose up into space, and then displayed flames, without any reference to the twin miracle. The parallel DĀ 15 at T I 69a26, translated in Weller 1928: 134, does not even mention a magical display of fire; cf. also the discussion in Weller 1922/1987: 189f. The other challenge is situated at Sāvatthī; cf. also Dhp-a III 213,15. Paṭis I 125,2 indicates that only a Buddha can perform the twin miracle, as it is outside the purview of disciples, asādhāraṇaṃ sāvakehi. According to the Mahāvaṃsa 17.44 and 31.99, the twin miracle was also performed by the Buddha's relics. On the motif of the twin miracle cf., e.g., Foucher 1909: 10–16, Lüders 1941/ 1966: 62–73, Brown 1984, Verardi 1988: 1540f, Karetzky 1990: 72f, Rhi 1991, Schlingloff 1991 and 2000: 488–515, Skilling 1997: 303–315, Zhu 2006: 255f, Anālayo 2009j and 2015d, and Fiordalis 2010/2011: 401f. according to the Mūlasarvāstivāda tradition adept disciples are also able to perform this feat.[n.15] [14] This is stated explicitly in the Divyāvadāna, Cowell and Neil 1886: 161,13, which follows a description of the twin miracle with the indication that this is tathāgatasya sarvaśrāvakasādhāraṇā ṛddhiḥ. Skilling 1997: 315 concludes that "the Mūlasarvāstivādins, Sarvāstivādins, Lokottaravādins, Mahīśāsakas, Aśvaghoṣa, and Asaṅga along with the Ratnaguṇasaṃcaya, Ekottarikāgama, P'u yao ching, and Book of Zambasta, disagree with the Theravādins, holding that an auditor as well as a Buddha could perform the yamakaprātihārya."

Although disagreeing on some of the miracles performed by Dabba, the parallel versions agree that he magically produced fire and burnt himself in mid-air.

Dabba and Fire

In the Theravāda narrative tradition, Dabba is repeatedly associated with the element fire. The commentary on the Theragāthā reports that his birth took place when his dead mother was placed on a funeral pyre.[n.16] [15] Ñāṇananda 2010: 23 comments that "because of his power of merit, [the] flames of his mother's funeral pyre could not burn him, but by his power of determination, he cremated his own body … leaving no ash or soot." Hence, as Wilson 2003: 36 highlights, "Dabba passed away in the same igneous manner in which he was born." The motif of birth from fire recurs in the tale of Jyotiṣka, cf., e.g., Cowell and Neil 1886: 270,3, and for further versions and representations in art Härtel 1981. Because of the heat her belly burst open, whereupon the child emerged without harm.[n.17] [16] Th-a I 43,11. In this way Dabba is born without having to pass through the vagina of his mother and thus in this respect is comparable to the Buddha's birth from the side of his mother; cf., e.g., the Buddhacarita 1.9, Johnston 1936/1995: 1, the Mahāvastu, Senart 1882b: 206,4, T 184 at T III 463c13, T 185 at T III 473c1, T 188 at T III 618a17, T 189 at T III 625a24, T 193 at T IV 58b21, and Bu ston's History of Buddhism, Obermiller 1932/1986: 11.

The Pāli Vinaya records that Dabba, who had become an arahant at the age of seven, had taken on himself the task of allotting quarters to incoming monks. He would enter the fire element and, with his finger lit on fire, illuminate the path for monks who had come in at night.[n.18] [17] Vin II 76,24; cf. also Silk 2008: 160f.

His abilities in this respect are also reported in other Vinayas, some indicating that he would have one finger burning with fire,[n.19] [18] The Mahāsāṅghika Vinaya reports that he was able to produce light with the little finger of his right hand; cf. T 1425 at T XXII 394c9. others that he could do so with up to five fingers,[n.20] [19] According to the Mūlasarvāstivāda Vinaya, he could use one of his fingers, or else two, three, four or five fingers to provide illumination when assigning lodgings; cf. T 1442 at T XXIII 695c14. or with his whole hand,[n.21] [20] That his abilities involved the whole hand is recorded in the Dharmaguptaka Vinaya, T 1428 at T XXII 587b25, the Mahīśāsaka Vinaya, T 1421 at T XXII 15b4 (preceded by an indication that he entered concentration on the fire element for this purpose), and in the Sarvāstivāda Vinaya, T 1435 at T XXIII 22a14; cf. also T 203 at T IV 457a7 and the Anavataptagāthā, Hofinger 1954: 122,30 and T 1448 at T XXIV 88b22. references that show that the idea of his pyrotechnic abilities was well known in different Buddhist schools.

For a proper evaluation of the significance of the tale of his self-cremation, it needs to be kept in mind that tradition considered Dabba's dramatic way of ending his own life as something that actually happened. [n.22] [21] In relation to other stories that depict the renunciation of parts or the whole of one's body, Durt 2000: 8 notes that "la rhétorique, à la fois macabre et grandiloquente, du renoncement au corps ne doit pas être sous-estimée. Elle a eu un retentissement historique dans le bouddhisme chinois, comme le montrent notamment les biographies de moines." The Saṃyukta-āgama discourses and their Udāna parallels make a point of highlighting that Dabba's auto-cremation was so successful that no ashes or soot remained. No doubt this story is meant to be heard or read literally, as an account of what really took place.

Other Auto-cremations

Dabba's auto-cremation appears to be the only episode of this kind among the Pāli discourses.[n.23] [22] The suggestion by Keown 1996: 17 note 21 (cf. also Keown 1998/1999: 395 note 26) that "the aged Arhat Dabba rises in the air and disappears in a puff of smoke. There is a similar passage on Bakkula at M. iii. 124–8" does not appear to be correct. The Bakkula-sutta, MN 124 at MN III 124,24 to 128,5, and its parallel, MĀ 32 at T I 475a11 to 475c14, translated in Anālayo 2007a, make no reference to an auto-cremation of Bakkula, something only reported in the commentary; cf. Ps IV 196,23. Besides, it also does not seem that Dabba should be reckoned as "aged"; cf. below note 60. A reference to the auto-cremation of an arahant can be found in one of the Chinese parallels to the Mahāparinibbāna-sutta, which reports that Subhadra, the last disciple of the Buddha, ended his life in this way.[n.24] [23] In terminology closely reminiscent of the Dabba tale, T 7 at T I 204b25 reports that Subhadra entered concentration on the fire element and attained final Nirvāṇa in front of the Buddha, 即於佛前入火界三昧而般涅槃. The parallel discourse versions do not report that he cremated himself,[n.25] [24] DN 16 at DN II 153,11, Sanskrit fragment S 362 folio 173 V4, Waldschmidt 1950: 62, DĀ 2 at T I 25b25, T 5 at T I 172b2, T 6 at T I 188a1, SĀ 979 at T II 254b29, EĀ 42.3 at T II 752c11, and the Mūlasarvāstivāda Vinaya, T 1451 at T XXIV 397a11, with its Tibetan parallel in Waldschmidt 1951: 471, as well as the Avadānaśataka, Speyer 1906/1970: 234,6, agree that he decided to pass away before the Buddha. In none of these versions, however, does he use selfcremation for that purpose. In fact the Mūlasarvāstivāda Vinaya continues by reporting what happened afterwards in relation to the disposal of his body. which gives the impression that his auto-cremation is an element that would have been added at a later time.[n.26] [25] In a survey of scriptural models for spontaneous combustion, Benn 2007b: 103f also notes that according to EĀ 26.9 at T II 641c3 Mahāmaudgalyāyana "passed away after having gone through all the various stages of dhyāna (trance), and after entering the samādhis of fire and water and shooting both those substances from his body". His actual death is then preceded by another progressive attainment of the four absorptions, and on arising from the fourth absorption he reaches final Nirvāṇa; cf. EĀ 26.9 at T II 641c27: 從第四禪起, 尋時 取滅度, i.e., he does not pass away by self-cremation.

Turning to comparable stories in the Vinayas, the Mūlasarvāstivāda Vinaya reports the auto-cremation of the arahant Gavāṃpati on the eve of the first saṅgīti.[n.27] T 1451 at T XXIV 403b23 and D 6 da 303a5 or Q 1035 ne 287a3. References to Gavāṃpati's self-incineration are found in several works outside of the canonical Vinayas; cf. Przyluski 1926: 10 and 66 and Lamotte 1944/1981: 97 note 2. In this case again such an auto-cremation is not recorded in other Vinaya versions of the same event, making it fairly probable that the Mūlasarvāstivāda Vinaya account has in this respect incorporated a later element.

The Mahāsāṅghika Vinaya records a whole series of self-cremations by monks on hearing that the Buddha had passed away.[n.28] T 1425 at T XXII 491a2. The description of the first and last cases of autocremations indicates that these took place by rising up into the air and entering concentration on fire. Regarding the fire imagery, it is worth noting that these self-incinerations are preceded by the report that Mahākāśyapa, while on his way to the Buddha's funeral, manifested his finger being lit on fire; cf. T 1425 at T XXII 490a28. This narration is also without a counterpart in other Vinayas.

A self-incineration of Paccekabuddhas is recorded in the Ekottarika-āgama parallel to the Isigili-sutta.[n.29] EĀ 38.7 at T II 723b6 indicates that the Paccekabuddhas cremated their bodies while being up in space and attained final Nirvāṇa, 諸辟支佛即於空中燒身, 取般涅槃; for a translation of EĀ 38.7 cf. Anālayo 2010f; for a comparative study of EĀ 38.7 and its parallel MN 116 cf. Anālayo 2011b: 654–657. The circumstance that no such occurrence is mentioned in the Pāli parallel makes it probable that in this respect the Ekottarika-āgama tale, like several stories in this collection, reflects later developments.[n.30] Anālayo 2009k, 2013d, and 2013e.

Other instances of auto-cremation by a single or several Pratyekabuddhas can be found in the Mahāvastu, in the Lalitavistara, and in the Bhaiṣajyavastu as well as the Saṅghabhedavastu of the Mūlasarvāstivāda Vinaya.[n.31] Senart 1882b: 357,16, Lefmann 1902: 18,22, Dutt 1984a: 5,19, and Gnoli 1977: 92,16; cf. also Bu ston's History of Buddhism in Obermiller 1932/1986: 7. Wilson 2003: 41 comments on the self-incineration of Paccekabuddhas that, since "the appearance of the Buddha of this epoch rendering their continued existence in this world superfluous, these solitary Buddhas committed their bodies to the flames in what may be seen through the lenses of Vedic-Hindu mortuary rites as a sacrificial act of passing the torch to their successor." For a Bhārhut relief apparently depicting such an act of self-incineration cf. Lüders 1941/ 1966: 44.

In sum, within the textual corpus of the early discourses and the Vinayas, it seems that the Dabba episode is the only of its kind that is recorded in parallel versions, making it likely that we have here a fairly early, perhaps even the earliest, textual report of selfcremation in the Buddhist traditions.

The perhaps most famous textual instance of auto-cremation occurs in the Saddharmapuṇḍarīka-sūtra, where a bodhisattva sets himself on fire as an offering to the Buddha. The Saddharmapuṇḍarīka-sūtra episode was taken literally in subsequent times as a script to be acted out. Generations of Chinese Buddhists actually undertook self-cremation, reciting or invoking precisely this passage.[n.32] Benn 2009: 108 notes that "the Lotus Sūtra provided not only a template for auto-cremation, by showing readers how and why it might be performed, but also the liturgy: self-immolators chanted the chapter on the Medicine King as they enacted it, thus making the scripture into a kind of performative speech"; cf. also Gernet 1960: 541–544, Jan 1965: 246f, and Kieschnick 1997: 42f. At times auto-cremations also appear to have been inspired by the account of the Buddha's cremation; cf. Benn 2007a: 37. The presumably earliest reported case of self-cremation in China involves a fourth-century monk who is on record for having the constant aspiration to follow the example of the Medicine King and burn his body as an offering.[n.33] T 2059 at T L 404c13: 常欲仰軌藥王燒身供養; cf. Benn 2007a: 33f. Such re-enacting of the scriptural model provided in the Saddharmapuṇḍarīka-sūtra forms a continuum with practices in China that predate the introduction of Buddhism and which,[n.34] Benn 2004: 759 explains that "as practised in China, autocremation … developed after a particular interpretation of certain Indian texts was combined with indigenous traditions, such as burning the body to bring rain, a practice that long predated the arrival of Buddhism in China." In contrast, in India the practice of burning one's own body had according to Bronkhorst 2012 no precedent in the Vedic sacrificial tradition. needless to say, have continued up to present times.

Notably, some aspects of the Saddharmapuṇḍarīka-sūtra episode parallel elements of the Dabba tale. Both the Bodhisattva Sarvasattvapriyadarśana (the future Bhaiṣajyarāja) and Dabba undertake their feats in front of a Buddha. Whereas the bodhisattva performs the self-cremation through conventional means by wrapping himself in a cloth soaked in oil and setting this on fire, a stanza that follows the prose description of his deed indicates that at that time he was in a type of samādhi, explained in the text to be a form of concentration that displays all kinds of forms.[n.35] The stanza in Kern and Nanjio 1884/1992: 408,14 just refers to an unspecified form of samādhi, but earlier the text indicates that the bodhisattva had acquired the sarvarūpasaṃdarśanaḥ samādhiḥ; cf. Kern and Nanjio 1884/1992: 405,14. The Chinese translation by Kumārajīva uses the corresponding expression 現 一切色身三昧 or 一切現諸身三昧 on both occasions, i.e., when describing his original attainment of this concentration and in the stanza that refers back to his self-cremation; cf. T 262 at T IX 53a26 and 53b23. The translation by Dharmarakṣa describes his attainment of the concentration that displays all, 普 現三昧, T 263 at T IX 125b4, followed by a reference to just the samādhi, 三 昧, in the stanza at T IX 125c2.

The reference to such an ability brings to mind Dabba's miraculous feats described in the two Saṃyukta-āgama versions, displaying colours and different postures after he had entered the appropriate type of samādhi.

On being reborn, the bodhisattva then worships the remains of another Buddha by burning his forearms. When the audience witnessing his feat is dismayed at the deplorable condition to which his body has been reduced, he makes a vow and thereupon his forearms are restored.[n.36] Kern and Nanjio 1884/1992: 413,7, T 262 at T IX 54a5, and T 263 at T IX 126a16.

The net result of this episode is also to some extent reminiscent of Dabba, who was able to set his fingers or hand on fire without suffering damage. Notably, when the bodhisattva sets himself on fire for the first time, he illumines all directions,[n.37] Kern and Nanjio 1884/1992: 407,9, T 262 at T IX 53b10, and T 263 at T IX 125b18. just as Dabba was able to illumine the surroundings by setting his fingers or hand on fire.

Although these elements in the Saddharmapuṇḍarīka-sūtra do not correspond exactly to the narrative material related to Dabba, they do give the impression of some degree of narrative continuity.

Such narrative continuity can also be seen with other texts. The motif of burning an arm that is then restored recurs in the Samādhirāja-sūtra, where Kṣemadatta wraps his right arm in a robe, soaks it with oil, and then burns it as an offering to a Buddha, illumining all directions.[n.38] Vaidya 1961: 219,8 and T 639 at T XV 598b2; cf. also Filliozat 1963: 23. When the audience is distressed, he makes a vow and his arm becomes as it was before.

In the Karuṇāpuṇḍarīka the bodhisattva who was to become the Buddha Śākyamuni sets his arm on fire, after having wrapped it in oiled cloth, to light the way for merchants lost at sea.[n.39] Yamada 1968: 371,2.

An Avadāna collection reports that the bodhisattva destined to become the Buddha Dīpaṃkara set his head on fire to provide light for a meal offering made to the Buddha of that time.[n.40] T 152 at T III 14c27, translated in Chavannes 1910: 86.

I will stop at this point, since a more comprehensive review of textual accounts of such partial or total self-cremations would carry me beyond the confines of the present chapter. The above examples suffice for the time being to convey the impression that some form of the Dabba tale, with a self-cremation through samādhi by someone who has the ability to illuminate the surroundings by setting his hand on fire without suffering any damage, could have provided the blueprint for the Saddharmapuṇḍarīka-sūtra episode and similar tales.

Although this need not have been the original intention of these textual descriptions, it seems clear that the depiction of a deliberate decision by highly accomplished practitioners – arahants, Paccekabuddhas, and advanced bodhisattvas – to end their own life has been taken literally by subsequent tradition.

In fact texts that originated in China and were of considerable importance in the history of Chinese Buddhism clearly recommend the practice of auto-cremation. One example is the Discourse on Brahmā's Net (梵網經), according to which to be reckoned a bodhisattva who has gone forth requires burning one's body, an arm or a finger as an offering to Buddhas.[n.41] T 1484 at T XXIV 1006a19: 若不燒身臂指, 供養諸佛, 非出家菩薩. Another example is the Śūraṃgama-sūtra, which proclaims that a monk who in front of an image of the Tathāgata burns his body as a lamp or sets fire to a finger joint will forever be free from the influxes (āsava).[n.42] T 945 at T XIX 132b14: 能於如來形像之前, 身然一燈, 燒一指節 … 永脫諸漏; for a discussion of this and the above passage from the 梵網經 cf. Benn 1998.

The impact of such recommendations can still be seen in present times in the practice of burning cones of incense on one's head at the time of ordination into the monastic order.[n.43] On burning the head at ordination cf. in more detail Benn 1998: 301–310.

The Development of the Dabba Tale

Turning from later developments to what might be the precedents for the Dabba episode reveals still another dimension of the impact of literalism, an appreciation of which requires a closer examination of the Dabba narration and its relation to the stanzas that accompany it in the light of other discourses in the Udāna collection.

Unlike the Saṃyukta-āgamas, where each of the two collections has a single discourse with the Dabba tale, the Udāna has two discourse versions that report the same episode. These two versions immediately follow each other within the Udāna collection.

The prose of the first of these two versions corresponds to the Saṃyukta-āgama narration. The second Udāna discourse reports that the Buddha, who in the meantime had moved from Rājagaha to Sāvatthī, recounts the whole event to the monks dwelling at Jeta's Grove. This is rare among Pāli discourses, that is, a discourse that reports an event followed by another discourse that depicts how the Buddha then reported this event to the monks.[n.44] Another such case, noted by Kuan 2008: 137, is SN 47.43 at SN V 185,3, where the Buddha reports his original delivery of the discourse SN 47.18 at SN V 167,4, although the two discourses do not follow each other immediately. Considered in conjunction these two Udāna discourses convey the impression of a conscious effort on the side of the reciters to draw attention to this tale and perhaps also ensure that it be considered authentic.

Another difference between the two Udāna versions is to be found in the stanzas that come at their respective ends. The first Udāna discourse is followed by stanzas describing the death of an arahant, without any reference to fire.[n.45] Ud 8.9 at Ud 93,12, with Indic language parallels in Uv 26.16, Bernhard 1965a: 322, in the Mūlasarvāstivāda Vinaya, Dutt 1984a: 258,8 and Wille 1990: 142,1; cf. also the Mūlamadhyamakakārikā, de La Vallée Poussin 1913/1992: 520,4. The stanza at the end of the second discourse has a simile of a blazing spark or pellet that is extinguished as a way of illustrating the destiny of an arahant who has gone beyond the flood of sensual desires,[n.46] Ud 8.10 at Ud 93,23, with Indic language parallels in Uv 30.35f, Bernhard 1965a: 400f, and in the Mūlasarvāstivāda Vinaya, Dutt 1984b: 81,15; cf. also SHT V 1100 V1, Sander and Waldschmidt 1985: 95, Nakatani 1987: 86, and Enomoto 1994: 21. similar to the stanza found at the end of the Saṃyukta-āgama discourses.[n.47] SĀ 1076 at T II 280c11 and SĀ2 15 at T II 378b12. The motif of the blazing spark that becomes extinct occurs also in AN 7.52 at AN IV 70,18 and its parallel MĀ 6 at T I 427a28, where it serves to illustrate the attainment of final Nirvāṇa by a particular type of non-returner.

Both of the stanzas from the two Pāli Udāna discourses recur in a Chinese Udāna collection that similarly combines stanzas with prose.[n.48] T 212 at T IV 734b11 and at T IV 757c1. Another Udāna collection preserved in Chinese, T 213, as well as the Sanskrit and Tibetan Udāna collections, do not have prose sections appended to their stanzas. Neither of these two instances, however, has any reference to Dabba or an auto-cremation in their respective prose sections.

The present instance conforms to a recurrent pattern that can be observed when comparing this Chinese Udāna collection and the Pāli Udāna, where most of the Chinese parallels to the actual udānas do not have the prose narration that according to the Pāli Udāna collection records the occasion for the delivery of the respective udāna.[n.49] Cf. in more detail below p. 463ff.

Judging from this pattern, in the Theravāda tradition the udānas – stanzas that appear to be common heritage of the Buddhist schools – were at some point associated with prose narrations that eventually became canonical. Although some of these narrations could be drawing on rather ancient material, in general it is safe to conclude that the prose is later than the stanzas.

What makes this pattern of interest to my present topic is that at times a stanza in the Pāli Udāna that has a symbolic sense is accompanied by a prose narration that reflects a literal understanding of this stanza. An example that illustrates this can be found in a stanza that enquires:

What is the use of a well,
If water is there all the time?
Having cut craving at its root,
What would one go about searching for?[n.50] Ud 7.9 at Ud 79,5: kiṃ kayirā udapānena, āpā ce sabbadā siyuṃ? taṇhāya mūlato chetvā, kissa pariyesanaṃ (Ee: pariyesanañ) care ti?, with Indic language parallels in Uv 17.9, Bernhard 1965a: 236, in the Mahāparinirvāna-sūtra (§7.11), fragment TM 361 156R1f, Waldschmidt 1950: 52, and in the Divyāvadāna, Cowell and Neil 1886: 56,12.

Whereas the stanza uses the image of the well as an illustration, the prose that purports to record the original occasion when the stanza was delivered reports that brahmins of a particular village had blocked a well with chaff in order to prevent the Buddha and his monks from drinking. When the Buddha requested that Ānanda fetch some water for him, the well by itself threw up all chaff and was filled with clean water to the brim.

Pande (1957: 75) comments that "the author of the prose … seems to have grossly misunderstood the … verse, which intends 'water' in no more than a merely figurative sense." The Chinese parallel to this stanza in fact does not mention any such incident and instead provides an interpretation of the well imagery.[n.51] T 212 at T IV 707c20.

Moving from water to fire and thus coming closer to the Dabba tale, another stanza from the Udāna collection refers to moths that are burnt to death because they are attracted by a burning lamp. In the Pāli stanza and its Sanskrit and Tibetan parallel, the motif of falling into a flame is clearly a simile used for the purpose of illustration.[n.52] Ud 6.9 at Ud 72,18: patanti pajjotam ivādhipātā (Be and Ce: ivādhipātakā), which has a counterpart in Uv 29.5, Bernhard 1965a: 367: patanti hi dyotam ivāndhakārād, and in the Tibetan version, Beckh 1911: 104: me nang lhung ba’i phye ma leb dang ’dra.

According to the prose of the Pāli Udāna, however, this actually happened. On realizing that many moths were meeting with misfortune by falling into what in the prose has become actual lamps, the Buddha then speaks the inspired utterance in question.[n.53] Ud 6.9 at Ud 72,11: addasā kho bhagavā te sambahule adhipātake tesu telappadīpesu āpātaparipātaṃ anayaṃ āpajjante byasanaṃ (Ce: vyasanaṃ) āpajjante anayabyasanaṃ āpajjante (Ce does not have anayabyasanaṃ āpajjante). On realizing this, the Buddha then reacts by delivering an udāna.

This may well be another case where what originally was a simile has been taken literally; in fact in the prose passage in the Chinese parallel the motif of a moth being burnt by a lamp is still a simile.[n.54] T 212 at T IV 748a26: "like a moth that throws itself into a flame without considering subsequent worries", 如蛾投火不顧後慮. This is particularly remarkable as the actual stanza in T 212 at T IV 748a20 (and in the other Udāna collection preserved in Chinese, T 213 at T IV 793a9, which unlike T 212 is without prose) does not refer to moths being burnt, but instead speaks of birds that throw themselves into a net (T 213: that fall into a net). As a consequence of this apparent instance of literalism, the prose of the Pāli Udāna depicts the Buddha as lacking compassion for the misfortune of these insects, since, rather than pronouncing an inspired utterance, he could have just asked one of the monks to put out the lamps and thereby stop the destruction of innocent creatures.

A similar instance of the literalization of a metaphor might stand at the background to Dabba's auto-cremation,[n.55] In a study of instances of the gift of the body, Ohnuma 1998: 356 speaks of a "literalization of a metaphor", which takes place when "a traditional Buddhist metaphor … is 'literalized' or turned into an actual, literal story." occasioned by the stanza's description, in relation to an arahant, of a splinter or pellet that is all ablaze with fire and then extinguished. A literal interpretation of this motif, along the lines of the above instances of the water in the well and the moth falling into a flame, could indeed result in imagining that this refers to the cremation of an arahant's body.

In other words, the first stage of the textual development of the Dabba tale could have been the existence of a mere stanza without an explicit mention of any auto-cremation, such as still nowadays found in several of the Udāna collections extant in Chinese, Sanskrit and Tibetan.[n.56] Besides the Sanskrit version mentioned above in note 46, the simile of the glowing spark occurs on its own, without any accompanying prose, in the Tibetan Udāna collection, Beckh 1911: 118 stanza 30.36 (= stanza 30.37 in Zongtse 1990: 341), and in one of the two Chinese Udāna collections, T 213 at T IV 795a8, translated in Willemen 1978: 144 (stanza 30.36).

Over the course of oral transmission, this stanza would have been accompanied by an explanation of its significance, such as now found in the prose section that accompanies this stanza in the Chinese Udāna collection and in the Pāli commentary on the Udāna of the Pāli canon. Alongside such explanations, however, a literal understanding of the stanza would have related it to the tale of an arahant who cremates himself.

The powerful attraction of the resulting tale would then have inspired further developments, resulting in the various miracles Dabba performs prior to his self-cremation in the two Saṃyukta-āgama versions on the one hand and in the doubling of the discourse in the Pāli Udāna on the other hand. In both cases, the inclusion of this tale as part of a canonical discourse corresponds to tendencies in the respective collections to include material of a more commentarial type.[n.57] In the case of the Saṃyukta-āgama, an example is SĀ 506 at T II 134a7, a discourse that reports Mahāmoggallāna visiting the Buddha who is spending the rainy season in the Heaven of the Thirty-three, paralleling Dhp-a III 216,13; cf. below p. 415ff. For more cases of inclusion of material in the Saṃyukta-āgama that in the Pāli tradition is found in commentarial literature cf. below p. 441ff; for another case found in the Udāna and in the two Saṃyukta-āgamas that also appears to reflect later influences cf. Anālayo 2012g: 392 note 66. Nevertheless, given that this tale is found in parallel discourse versions, it must have come into being at a comparatively early stage in the evolution of Buddhist literature.[n.58] For a critical reply to the suggestion by Schopen 1985/1997 that parallelism should be seen rather as the result of later levelling cf. Anālayo 2012f.

The doubling of the tale in the Pāli Udāna collection is in fact strange, since it results in the same tale leading to two different inspired utterances by the Buddha. Once the Buddha was on record as having uttered a particular stanza at the time of Dabba's passing away, it would have been more natural to employ the same stanza when reporting what had happened. This gives the impression that the doubling of the narrative was perhaps accommodating the stanzas already in existence. As a result, one of the two versions came to be associated with a different stanza.

Be that as it may, the tale of Dabba's auto-cremation, itself perhaps the outcome of literalism, in turn seems to stand at the outset of textual descriptions that inspired further literalism. In the transition from India to China such literalism resulted in a long series of actual instance of self-cremation by monastic and lay Buddhists, male and female.

Without thereby intending to deny the complex interweaving of various cultural and religious factors that must have contributed to this development, it nevertheless seems to me that if my hypothetical reconstruction should be correct – and I need to stress that this is just a hypothesis – then a poetic spark set fires burning all the way down the history of Buddhism up to modern times.

Arahants and Suicide

In as much as the term suicide stands for the deliberate decision to end one's life, the Dabba tale and its successors, be they only textual imagination or actual happenings, would fall into this category. Regarding the nature of the "suicide" undertaken by Dabba, it is noteworthy that the different versions agree on beginning their account with Dabba requesting the Buddha's permission. This introductory narration only makes sense as part of a story that depicts an arahant who deliberately ends his life. In fact, in the two Saṃyukta-āgama discourses the Buddha at first remains silent and only replies when Dabba has made his request for a third time, in line with a recurrent trope in the early discourses where a request made for a third time signals the importance the petitioner attaches to the request and his keenness on having it granted.

In line with the difficulties that the notion of an arahant deliberately ending his or her life has for some later traditions – evident in the cases of Channa and Vakkali I studied in earlier chapters – the Pāli commentary reasons that Dabba was on the verge of passing away anyway, a condition of which he had just become aware.[n.59] Ud-a 431,4 explains that on emerging from a meditative attainment he reviewed his life-formation, which were exhausted and only going to last a few more moments, samāpattito vuṭṭhahitvā, attano āyusaṅkhāre olokesi. tattha te parikkhīṇā katipayamuhuttakālā upaṭṭhahiṃsu. Viewed in this light, Dabba's request and the Buddha's permission are only concerned with how Dabba should pass away, that is, if he should do so in the spectacular manner described in the discourses.

The commentary then records another explanation as well, according to which Dabba was not on the verge of passing away.[n.60] Ud-a 431,21, an explanation that begins by indicating that he had been neither old nor sick, na tāva thero jiṇṇo, na ca gilāno. Regarding his age, from the viewpoint of the above-mentioned Vinaya account that he had become an arahant at seven and then decided to take up the duty of allotting quarters and meal invitations to other monks, Vin II 74,25, it would follow that he was seven at a time when the Buddhist monastic order had already grown to such a size that it needed someone to ensure the proper distribution of living quarters and meal invitations. Since the account in Ud 8.9 and Ud 8.10 implies that the Buddha was still alive at the time of Dabba's death, it would follow that at that time Dabba could not have been older than in his forties. In fact Malalasekera 1937/1995: 1060 concludes that "Dabba evidently died young." His decision was rather motivated by the grudge some monks bore against him, which had motivated them to stage a wrong accusation that he had committed a pārājika offence by engaging in sex.[n.61] On the wrong accusation levied against Dabba by the nun Mettiyā cf. the discussion in Horner 1930/1990: 266f, von Hinüber 1995: 37 and 1997, Hüsken 1997a: 96–98, Nolot 1999: 66–68, Heirman 2000, and the solution of the conundrum discovered by Clarke 2008. This line of reasoning is also recorded in a version of the Dabba episode found in a collection of tales preserved in Chinese translation.[n.62] After referring to his being slandered, T 203 at T IV 457a13 continues by reporting that, being disgusted, he went up into the sky, performed eighteen transformations and entered concentration on bright fire. As the blazing fire in the sky was extinguished, no bones remained, 駝驃厭惡, 即昇虛空, 作十八 變, 入火光三昧, 於虛空中, 如火焰滅, 無有屍骸. According to the Pāli commentary such an interpretation should be rejected, however, since arahants will not be motivated by praise or blame to prolong or shorten their lives.[n.63] Ud-a 431,29: taṃ akāraṇaṃ, na hi khīṇāsavā … paresaṃ apavādādibhayena parinibbānāya cetenti ghaṭanti vāyamanti, na ca paresaṃ pasaṃsādihetu ciraṃ tiṭṭhanti.

The expression used by Dabba to communicate his intention to the Buddha in the Udāna discourse, indicating that for him the time for final Nirvāṇa had come, is parinibbānakālo.[n.64] Ud 8.9 at Ud 92,29 (this part is abbreviated in Ud 8.10): "it is time for my final Nirvāṇa, Well Gone One", parinibbānakālo me dāni sugatā ti. The same expression recurs in the Mahāparinibbāna-sutta to indicate that the time for the Buddha's final Nirvāṇa has come, preceded by a reference making it clear that he could have lived much longer.[n.65] DN 16 at DN II 104,17: "venerable sir, it is time for your final Nirvāṇa, Blessed One", parinibbānakālo dāni bhante bhagavato; cf. also AN 8.70 at AN IV 310,13 and Ud 6.1 at Ud 63,17. This remark is preceded in DN 16 at DN II 103,1 by the indication that one who has developed the four roads to (spiritual) power can, if he so wishes, remain alive for a kappa or its remainder. Whatever the term kappa means, it definitely entails an extra time period. This indication is then followed by the declaration that the Tathāgata had developed the four roads to [spiritual] power and thus would have been able to remain alive, if he so wished. The parallels to DN 16 agree that the Buddha could have continued to live for a kalpa or its remainder; cf. Sanskrit fragment S 360 folio 173 R4–6, Waldschmidt 1950: 19, DĀ 2 at T I 15b20, T 5 at T I 165a10 (which mentions the potential of the ṛddhipāda to enable one to remain alive for a kalpa or its remainder, but does not explicitly attribute the ṛddhipāda to the Buddha, although this much is clearly implied), T 6 at T I 180b15, T 7 at T I 191b16, and the Tibetan version in Waldschmidt 1951: 205,16; cf. also the Divyāvadāna, Cowell and Neil 1886: 201,7. Judging from this passage, it would be natural if the same phrase used by Dabba had similar implications. That is, the phrasing of the Udāna discourse reads as if he decided to enter Nirvāṇa even though he could have lived longer.

Be that as it may, regarding suicide by fully awakened ones it is significant that the Mahāparinibbāna-sutta and its parallels agree that the Buddha gave up his life deliberately.[n.66] DN 16 at DN II 106,22: āyusaṅkhāraṃ ossaji, Sanskrit fragment TM 361 folio 166 V4, Waldschmidt 1950: 53: āyuḥsaṃskārān-uts[ṛ]jati, DĀ 2 at T I 15c20: 捨命住壽, T 5 at T I 165a23: 放棄壽命, T 6 at T I 180c6: 棄餘壽行, T 7 at T I 191c8: 捨壽, and the Tibetan version, Waldschmidt 1951: 213,4: tshe’i ’du byed ni spangs so; cf. also the reference in the Divyāvadāna, Cowell and Neil 1886: 203,7 to āyuḥsaṃskārān utsraṣṭum; cf. also below p. 471 note 31. The description of this act of deliberately letting his life end cannot have been intended to portray him acting with an unwholesome state of mind.

This in turn indicates that tradition did envisage that such a decision can be taken without being motivated by defilements. In other words, this passage portrays a fully awakened one taking the deliberate decision to let life end.[n.67] Blum 2004: 206 comments that "almost akin to a pronouncement of suicide, the sūtra reads, 'and now, Ānanda, the Tathāgata has today at Chāpāla's shrine consciously and deliberately rejected the rest of his allotted time.'" Delhey 2006: 36 observes that "even the death of the Buddha can be considered as a kind of suicide"; cf. also the Abhidharmakośabhāṣya, Pradhan 1967: 74,28f, discussed in Berglie and Suneson 1986: 29f and 38, and in Lamotte 1965: 158f.

Unlike the other cases discussed so far – Channa, Vakkali, and Dabba – the report of the Buddha deliberately letting his life end does not appear to have led to ambivalence in later tradition regarding the possibility that an awakened one could take the decision to end his or her own life.[n.68] Needless to say, the famous last meal of the Buddha takes place only after he has decided to renounce his life principle; cf. DN 16 at DN II 127,2. Moreover, except for DĀ 2 at T I 18b5 (which refers to some sort of mushroom), the other parallels do not report anything unusual about this meal, the point of relating the episode instead being that one of the monks present commits a theft (an event also known in the Theravāda tradition; cf. the narration in Pj II 159,20 in relation to Sn 83–90); cf. Sanskrit fragment 360 folio 187 V3, Waldschmidt 1950: 25, T 5 at T I 167c17, T 6 at T I 183b5 (T 7 at T I 197a29 does not mention either the nature of the meal or the theft), and the Tibetan version, Waldschmidt 1951: 257,31. On the Buddha's last meal cf., e.g., Waley 1931/ 1932, Waldschmidt 1939/1967, Fa Chow 1942, Bareau 1968, Martini 1972, Wasson and O'Flaherty 1982, Mettanando and von Hinüber 2000, An 2006, and Strong 2012.

The fact that the parallel versions of the Mahāparinibbāna-sutta unanimously attribute the Buddha's passing away to a conscious decision taken by him, even though he would have been able to live longer, provides a strong argument in favour of the notion that, according to early Buddhist canonical literature, a fully awakened one can deliberately end his or her own life.

This certainly in no way exonerates the cases of actual autocremation mentioned above. Paraphrasing the words of the Channa-sutta and its Saṃyukta-āgama parallel, killing oneself is blameworthy if after death one is reborn.[n.69] SN 35.87 at SN IV 60,1 and SĀ 1266 at T II 348a25; cf. above p. 264f. In other words, suicide is blameworthy if one does not pass away as an arahant.

According to this dictum, the conscious decision to let one's life end is only acceptable when the person in question will not be reborn.[n.70] Olivelle 1978: 20 explains that in the ancient Indian setting "suicide in the various ascetic traditions … has one characteristic in common. It is resorted to at the end of a life of asceticism when the ascetic has already reached perfection." In the case of early Buddhism, as pointed out by Harran 1987: 129, "while suicide … may be appropriate for the person who is an arhat, one who has attained enlightenment, it is still very much the exception to the rule." Florida 1993: 41 explains the rationale behind this position to be that "since the body is by nature mortal and corruptible it is only of instrumental value, there is no absolute duty to hold on to it when it has lost its utility." It is important to keep in mind this distinction, in that the examples I have been discussing in this and previous chapters are of a special kind, different from the suicide of those who are not fully awakened or else on the brink of such attainment.

Ancient Indian religious traditions did consider suicide appropriate in such circumstances,[n.71] Cf., e.g., Kane 1941: 924–928 and 1953: 604–614, Filliozat 1967, Sircar 1971, and Oberlies 2006, and on suicide as a form of threat to obtain some objective Hopkins 1900. particularly famous in this respect being the Jain tradition's advocating of a practice often referred to as sallekhanā, where the accomplished saint fasts to death.[n.72] Cf., e.g., Tatia 1968, Tukol 1976, Caillat 1977, Settar 1990, Bronkhorst 1993/ 2000: 31–36; and on suicide by Ājīvikas cf. Basham 1951: 63f, 84–90, and 127–131.

The Jain tradition sees the undertaking of such fasting to death as quite different from ordinary suicide, which is in fact not allowable.[n.73] Cf. the contrast made in the Viyāhapaṇṇatti 2.1, Lalwani 1973: 163f, between the death of a fool by entering fire, taking poison, etc., and the death of the wise by remaining motionless and refusing food. For a case study of a foolish death by auto-cremation cf. Settar 1990: 81f. The aspirant has to get permission for this undertaking from his teacher, who needs to ascertain that the fast to death is undertaken with the proper attitude. This proper attitude is described in the Āyāraṅga as requiring that the aspirant longs neither for life nor for death.[n.74] A description of the preparation for fasting to death in the Āyāraṅga 1.7.8.4, Jacobi 1882: 38,22, indicates: jīviyaṃ nā 'bhikaṃkhejjā maraṇaṃ no vi patthae, duhato vi na sajjejjā jīvite maraṇe tahā; translated in Jacobi 1884/1996: 75 as "he should not long for life nor wish for death, he should yearn after neither, life or death." Tukol 1976: 10 points out that wishing for death to occur more quickly or else for it to come later are considered forms of transgression. Skoog 2003: 298 indicates that "sallekhanā requires dispassion on the part of the person who is allowed to engage in the process. Part of the screening process by the religious supervisor involves deciding whether the practitioner is truly ready to take on the long process of fasting to death." Thus, as already noted by de La Vallée Poussin 1922: 25, "while practising starvation, the Jain must avoid any desire for death." This is precisely the attitude that Buddhist texts associate with an arahant.[n.75] Cf., e.g., the stanza attributed to Sāriputta at Th 1002f: nābhinandāmi maraṇaṃ nābhinandāmi jīvitaṃ, according to which he delighted neither in death nor in life. This phrase occurs several times in the Theragāthā collection, suggesting that it conveys an attitude held in general esteem; cf., e.g., Th 196, Th 606, Th 654, and Th 685.

Strabo reports that Alexander the Great was able to witness an auto-cremation, undertaken by the Indian Kalanos who had accompanied him during part of his journey.[n.76] Geography of Strabo 15.1.68, Jones 1930: 118,3; cf. also Majumdar 1960: 187, for a critical examination of this tale Fick 1938, and for a recent discussion of Kalanos Halkias 2015. Thus already in ancient times the Indian attitude to self-inflicted death had its impact on a western audience. Behind this attitude stands to some degree a different value system. The emphasis in the Indian context is often not on valuing life as such,[n.77] Pace Keown 1998/1999: 387, who affirms that "to choose death over life is to affirm all that Buddhism regards as negative", it "is to negate in the most fundamental way the values and final goal of Buddhism". but rather on the overarching importance of non-violence, ahiṃsā, in order to avoid the karmic consequences that will be incurred when harming others. As the case of the Jains shows, ahiṃsā was evidently not seen as being in conflict with suicide by an advanced practitioner.[n.78] As noted by Young 1989: 88 it is quite telling that "a religion which has, as a cardinal doctrine, ahiṃsā or non-injury to any living creature – including never killing an insect intentionally or unintentionally – yet has as an ideal: religious, self-willed death." Skoog 2003: 298 explains the reasoning to be that "sallekhanā does not harm another being; in fact it can and should be considered a supreme act of nonviolence, as one is no longer harming another being through the process of harvesting, storing, cooking, and consuming food." Delhey 2006: 56f then concludes that "suicide does not fulfil all the requirements that are needed to judge it in the same way as killing other living beings" as can be seen with the Jains, whose creed "combines a much stricter interpretation of the doctrine of ahiṃsā [compared to the Buddhists] with a remarkably positive attitude to a certain form of religious suicide: voluntary fasting unto death." In the case of Buddhism, then, taking "a position according to which life is considered as sacred and as a basic value in Buddhism is not in accordance with most of the ways in which suicide is dealt with in Buddhist texts". As pointed out by Kitagawa 1989: 29 "the 'dignity of life' has very different implications for the Buddhist tradition than for the Western world."

This decisive difference explains how at least part of the early Buddhist tradition could have had an approving attitude towards the suicide of a fully accomplished one. As Schmithausen (2000: 36 and 38) sums up, "in contrast to ordinary living beings, saints (i.e. fully liberated persons, arhats) do not regard their biological life as something valuable", since for them "the ultimate value is obviously not (biological) life but nirvāṇa."