Introduction

In this chapter I examine a few selected passages in which the Buddha manifests miracles that involve a display of fire. My main aim in what follows is to attempt to discern stages in the textual depiction of such miraculous performances through comparative study of these passages in the extant parallel versions. The cases I will be studying are Sakka's visit, a visit to a Brahmā, Pāṭikaputta's challenge, and the twin miracle, after which I take a brief look at the fire element in the early discourses.

---

Sakka's Visit

I begin with the Sakkapañha-sutta's depiction of the first meeting between Sakka, the ruler of the devas in the Heaven of the Thirty-three, and the Buddha. The Sakkapañha-sutta reports that on an earlier occasion Sakka had already tried in vain to visit the Buddha, not being allowed to disturb the Buddha's meditation. At the end of the discourse he receives on the present occasion, Sakka attains stream-entry.

The story of Sakka's humbled pride on initially not being given an audience with the Buddha and his eventual conversion to the Buddhist fold could be compared to a general narrative strategy of "inclusivism" in early Buddhist texts. This strategy refers to a tendency to include, although in a subordinate position and at times with significant modifications, central elements of other traditions within the framework of one's own.[n.1] On inclusivism cf., e.g., Oberhammer 1983, Mertens 2004, Kiblinger 2005, and Ruegg 2008: 97–99. In the case of Sakka, his role in early Buddhist texts involves the transformation of the ancient Indian warrior god Indra into a peaceful and devout Buddhist disciple.[n.2] Cf., e.g., Masson 1942: 46, Godage 1945: 70f, Lamotte 1966: 116, Barua 1967: 184, Arunasiri 2006: 629, Bingenheimer 2008: 153, and Anālayo 2012i: 9f. As part of this narrative strategy, the specific significance of the Sakkapañha-sutta lies in its recording his successful and complete conversion by dint of becoming a stream-enterer.

Lest I be misunderstood, my presentation in no way intends to downplay the fact that in the early discourses Sakka and other devas feature as actually existing celestial beings. In fact the Saṅgārava-sutta and its Sanskrit fragment parallel report a discussion during which the Buddha asserts the existence of devas.[n.3] MN 100 at MN II 212,26 and the Sanskrit fragment counterparts in Hartmann 1991: 260 (§ 147) and Zhang 2004: 11 (346v); on the position of this exchange within the overall discourse cf. Anālayo 2011b: 582f and note 269 in particular on the expression ṭhānaso me taṃ … viditaṃ yadidaṃ atthi devā. As pointed out by Harvey 1995: 82, in this passage the Buddha "clearly distinguished between his 'certain knowledge' that gods existed and the commonly held belief that they did". Norman 1977 (cf. also Norman 1985) suggests emending the question that begins the discussion in MN 100 from reading atthi devā to atthi adhideva, based on which Norman 1977: 336 then concludes that the Buddha, "far from conceding the existence of the brahmanical devas as Saṅgārava presumed, was merely saying that there were in the world earthly princes who were by convention called devas". This is to my mind an unconvincing suggestion and is also without support in the Sanskrit fragment parallels. I find similarly unconvincing the interpretation of the present passage by Marasinghe 1974: 128, who holds that it "does not mean anything more than that the Buddha recognized the fact that the belief in the gods was quite commonly known at that time. Hence, its very wide popularity itself would have convinced the Buddha that a direct denial was not the most expedient method of dealing with it." As far as I can see the text simply and clearly conveys an affirmation of the existence of devas; for further discussion and secondary sources on this passage cf. also Saibaba 2005: 3–6. The devas feature also as the object of one of the standard recollections described in the early discourses.[n.4] Recollection of devas is one of the traditional set of six recollections; cf., e.g., AN 6.10 at AN III 287,21 and its parallels SĀ 931 at T II 238a21 and SĀ2 156 at T II 433a22, which agree in asserting the existence of various types of devas and then direct recollection to the qualities that have led to their celestial rebirth. Thus in what follows my aim is decidedly not to attempt to strip early Buddhism entirely of its miraculous elements,[n.5] Waldschmidt 1930: 8f notes the prominence of the Buddha's magical powers already in the early tradition and advises against going so far as to turn early Buddhism into a pure philosophy, which would be in contrast to its nature, where the profound and the magic go hand in hand; "es wird im allgemeinen viel zu wenig betont, wie sehr auch im älteren Buddhismus die magischen Kräfte des Buddha in den Vordergrund treten … man sollte nicht soweit gehen, aus dem älteren Buddhismus eine reine Philosophie zu machen. Das widerspricht ganz und gar dem Wesen der Religion, bei der immer das Tiefsinnige und das Wunderbare Hand in Hand gehen." Gethin 1996: 204 explains that "there has been a tendency to play down the tradition of the 'miraculous' in Buddhism and to see it as peripheral, but it is hard to treat this tendency as anything but revisionist." Brown 1998: 50 comments that "the attempt to strip the texts of the legendary to reveal the true historical Buddha has been the focus of many scholars. How successful their attempts have been appears to me largely to rely on predetermined notions of the Buddha as a modern man." Fiordalis 2010/2011: 403 points out that "scholars have been too quick to conclude … that Buddhism rejects the miraculous wholesale in favor of some sort of rational humanism that reflects modern predilections … making this argument requires that one disregard the many Buddhist stories in which the Buddha or his eminent disciples perform acts of … displaying their superhuman powers. Scholars have suggested that such stories are merely 'popular' or represent 'later' (often a euphemism for degenerate) traditions. Yet, these are problematic conclusions." but only to discern, wherever possible, stages in their gradual growth.

The introductory section of the Sakkapañha-sutta describes Sakka getting ready for his second attempt to visit the Buddha:[n.6] DN 21 at DN II 264,10 to 264,25.

Then Sakka, the ruler of devas, surrounded by the devas of the Thirty-three and with the gandhabba Pañcasikha leading in front,[n.7] Be and Ce qualify the gandhabba Pañcasikha to be a devaputta. just as a strong man might stretch a bent arm or bend a stretched arm, in the same way he disappeared from the Heaven of the Thirty-three and reappeared in Magadha to the east of Rājagaha, on Mount Vediya, to the north of the brahmin village called Mango Grove.

At that time, due to the divine power of the devas, a very bright light manifested on Mount Vediya and in the brahmin village Mango Grove, so much so that the people in the surrounding villages said: "Today Mount Vediya is indeed on fire, today Mount Vediya is indeed burning, today Mount Vediya is indeed ablaze. How is it that today a very bright light manifests on Mount Vediya and in the brahmin village Mango Grove?", and they were excited with their hair standing on end.

In the Sakkapañha-sutta the mountain appearing to be on fire is clearly related to the divine power of the celestial visitors, devānaṃ devānubhāvena, whose arrival has caused this effect to manifest. The Sakkapañha-sutta has parallels preserved in Chinese translation as well as in Sanskrit fragments. In what follows I translate the corresponding section from the Madhyama-āgama version.[n.8] MĀ 134 at T I 633a8 to 633a13.

Then Sakka, the ruler of the devas, the devas of the Thirtythree, and the gandhabba Pañcasikha, just as quickly as a strong man might bend or stretch his arm, they suddenly disappeared from the Heaven of the Thirty-three, being no more to be seen there, and reappeared in the country of Magadha to the east of Rājagaha, not far from the cave on Mount Vediya, to the north of the village Mango Grove. Then Mount Vediya shone with a bright light like fire. Seeing this, the people dwelling around the mountain thought: "Mount Vediya is on fire, burning everywhere."

The Madhyama-āgama discourse does not have an explicit indication that the mountain's appearance was due to the divine power of the devas, corresponding to the expression devānaṃ devānubhāvena in the Sakkapañha-sutta. Nevertheless, the same idea is clearly implicit, since the mountain shines with a bright light as soon as the celestial visitors have arrived. Similar references to the mountain manifesting a fire-like brilliance, once the devas have arrived, but without an explicit attribution of this phenomenon to their power, can be found in a parallel preserved as a discourse translated individually into Chinese, and in another parallel that forms part of a collection of avadāna-type of tales, also extant in Chinese translation.[n.9] [10] T 15 at T I 246b21 and tale no. 73 in T 203 at T IV 476a28, translated by Chavannes 1911: 53–69. A parallel extant in Sanskrit fragments, however, does have a phrase similar to the Pāli version.[n.10] [9] Although this is not the case for the relevant part of the Sanskrit fragment version which, together with translations and a study of the parallels, can be found in Waldschmidt 1932: 65–67, the phrase is found in the British Library fragment Or 15009/549 r6, Nagashima 2015: 380: devānāṃ ca devatānubhā[v]e.

The individually translated discourse just mentioned also reports that Sakka and his host disappeared from their celestial abode just as quickly as a strong man might bend or stretch an arm and appeared on the mountain, which thereupon was illuminated by a great light. Its account differs from the other versions in so far as, after describing the reaction of the people on seeing the mountain illuminated in this way, it relates this effect also to the Buddha's presence. This takes the form of an indication given by Sakka to Pañcasikha in the following way:[n.11] T 15 at T I 246b24f; the significance of this statement has already been highlighted by Waldschmidt 1932: 66f note 2.

Do you see the special appearance of this mountain in this way? This is because the Buddha, the Blessed One, dwells within.

This indication could still be read in line with the passages surveyed so far, in as much as it is the presence of the Buddha in the mountain which motivates the arrival of the devas, and their arrival is then what makes the mountain appear as if it were on fire. Taken out of context, however, the statement could alternatively give the impression that the fiery appearance of the mountain is the result of the Buddha's presence. Yet, in this version this fiery appearance manifests only once the devas have arrived.[n.12] After reporting how Sakka and his host came to the mountain, T 15 at T I 246b21 continues by introducing the great brilliance of the mountain with the phrase "at this time", 是時, thereby clearly marking the temporal relationship, also evident in the other versions, between the arrival of the celestial visitors and the appearance of the mountain. This makes it safe to assume that its presentation is still in line with the basic plot in the versions discussed so far.

The Dīrgha-āgama preserved in Chinese translation also has a version of the Sakkapañha-sutta, and its presentation foregrounds the effect of the Buddha's presence. The section corresponding to the parts translated from the Sakkapañha-sutta and its Madhyama-āgama parallel proceeds as follows:[n.13] DĀ 14 at T I 62c10 to 62c14.

Then Sakka, the ruler of the gods, the devas of the Thirtythree, and Pañcasikha disappeared from the Dharma Hall, being no more to be seen there and, just as quickly as a strong man might bend or stretch an arm, they reappeared on Mount Vediya to the north of the [village Mango Grove] in the country of Magadha. At that time the Blessed One had entered concentration on fire and Mount Vediya appeared to be completely on fire. Then the country people, on seeing this, said to each other: "Due to the power of the Tathāgata and the devas, this Mount Vediya appears to be completely on fire."

Unlike the other versions surveyed so far, in the Dīrgha-āgama discourse the motif of the mountain appearing on fire is due to the power of the Buddha, and not just the presence of the devas.[n.14] Waldschmidt 1932: 66 considers the expression 如來諸天之力 to refer to the divine powers of the Tathāgata, "Götterkräfte des Tathāgata". It seems to me more probable that the plural indicator 諸 points to the devas, wherefore I take the whole expression to refer to the "power", 力, "of", 之, "the Tathāgata", 如 來, and "the devas", 諸天. Zwalf 1996: 198 draws attention to a Tibetan biography of the Buddha, translated in Schiefner 1849/1851: 255, which in relation to the Buddha indicates that, after arrival at the venue of the present event, "nahm er in der im Magadhagebiet gelegenen Sâlahöhle von Indraçaila Feuergestalt an und erfüllte die ganze Höhle mit seinem Körper. Solche Gestalt behielt er sieben Tage." I suppose the idea of the Buddha taking on the appearance of fire implies an effect similar to that described in DĀ 14, although in this Tibetan biography the Buddha apparently does so for a period of seven days. Moreover, the Dīrgha-āgama version also furnishes an explanation for how the power of the Buddha leads to this effect by reporting that he was in meditation on the fire element (dhātu).[n.15] On meditation on the fire element cf., e.g., Dantinne 1983: 272–274.

In principle this variation could be either a case of loss in the other versions or a case of addition in the Dīrgha-āgama discourse. When evaluating these two possibilities, whereas for the Buddha to make a whole mountain appear to be in flames appears to be unique among the early discourses, the description of devas who on arrival cause a whole place to be lit up is well attested in other discourses.

One out of numerous examples of the effect associated with the arrival of a deva can be found in the Mahākaccānabhaddekaratta-sutta and its parallels in a Madhyama-āgama discourse, an individual Chinese translation, and parallels extant in Tibetan translation.[n.16] MN 133 at MN III 192,7 and its parallels MĀ 165 at T I 696c5, T 1362 at T XXI 881c8, and D 313 sa 161b3 or Q 979 shu 171b1 (the discourse recurs in the Tibetan canon as D 617 or Q 599 and again as D 974; cf. the discussion in Skilling 1997: 81–83). The parallel versions agree in describing a deva who, on visiting a monk, lights up the whole place with its radiance.

Recurrent examples of the same type of description can be found in the Devatā-saṃyutta of the Saṃyutta-nikāya. Just taking the first discourse in this collection as an example, a description of the brilliant light caused by the arrival of the deva protagonist in this discourse can similarly be found in parallels in the two Saṃyukta-āgama collections.[n.17] SN 1.1 at SN I 1,9 and its parallels SĀ 1267 at T II 348b9 and SĀ2 180 at T II 438c14. In sum, the notion that the arrival of a deva can result in lighting up a whole place is a common motif in the early discourses.

Another point to be taken into consideration is how far the Buddha's manifestation of fire fits the present narrative context. The Sakkapañha-sutta and its parallels continue with Sakka asking Pañcasikha to approach the Buddha on his behalf and request an audience. This narrative element needs to be read in the light of Sakka's earlier unsuccessful attempt to visit the Buddha. This previous attempt stands in stark contrast to the way in which according to his own report he had been received by other recluses he had visited previously, who got so excited on receiving a visit from Sakka that they wanted to become his pupils. Another element leading up to this second visit is that Sakka had witnessed the rebirth in his heaven of those who during their earlier human existence had become disciples of the Buddha.[n.18] Greene 2013: 290 comments on DĀ 14 that the "episode in the Dīrghāgama … states that Indra came to visit the Buddha because he saw the light emitted when the Buddha entered the 'fire-radiance samādhi'." As far as I can see DĀ 14 does not indicate that Sakka came to visit the Buddha because he saw any light. The decision to visit the Buddha is simply presented as being a particularly wholesome state of mind making him wish to meet the Buddha, DĀ 14 at T I 62c2: 發微妙善心, 欲來見佛. The fire and light motif comes up only later, after Sakka has assembled his entourage, left the Heaven of the Thirtythree, and arrived at Mount Vediya.

Against this narrative background, for Sakka not to dare to approach the Buddha directly highlights his humbled pride and throws into relief the appropriate attitude towards the Buddha adopted even by the ruler of the devas of the Thirty-three. It also demarcates the aloofness of the Buddha compared to other recluses Sakka had approached earlier. Sakka's deferential attitude moreover reflects his respect as the result of having witnessed the favourable rebirths of disciples of the Buddha, a message that in an ancient Indian setting would surely not have been lost on the audience listening to the discourse.

These elements are somewhat lost from sight once the Buddha is in fire meditation to the extent that the whole mountain appears to be in flames. This description runs the risk of giving the impression that the Buddha's attainment of fire meditation is what makes Sakka ask someone else to find out if the Buddha is willing to grant him an audience. Such an impression would result in a loss of the humour and of a considerable part of the soteriological message that seem to underlie the scene in the other versions.[n.19] This corresponds to a pattern I noted in Anālayo 2008a: 146 in relation to the tale of Aṅgulimāla where "the introduction of wonders and miracles, as well as the successive amplification of narrative details, can at times obfuscate the main message of the text."

In view of the general tendency in Buddhist texts towards an increasing apotheosis of the Buddha,[n.20] Cf., e.g., Anālayo 2010b: 130. an intentional omission of the fire motif in the other versions is highly improbable. The individually translated discourse in fact testifies to a tendency towards giving more prominence to the Buddha's superior powers, as it already highlights the effect of the Buddha's presence, even though in its presentation the arrival of the devas is still what causes the mountain to appear as if lit up by fire. In fact the individually translated discourse does not provide any indication that the Buddha either manifests fire or is immersed in meditation on it.

In sum, a comparative study of the introductory narration to the Sakkapañha-sutta and its parallels shows that the Buddha's performance of a fire miracle is only attested in the Dīrgha-āgama version. Given that the illuminating effect of devas on the surroundings is well attested elsewhere among the discourses, and that the fire miracle does not fit the narrative context of this discourse too well, it seems safe to conclude that this effect is a later addition in the Dīrgha-āgama discourse. In this way the fiery appearance of the mountain, originally seen as the result of the presence of devas, has come to be attributed to the Buddha's presence, more specifically to his dwelling in meditation on fire.

The episode described in the Sakkapañha-sutta and its parallels has also inspired ancient Indian artists,[n.21] For surveys of different artistic representations cf., e.g., Foucher 1905: 492– 497, Coomaraswamy 1928, Lamotte 1944/1981: 181 note 2, Buchthal 1945: 167f, Soper 1949: 254–259, Zhu 2009, Parlier-Renault 2014, and Rhi (forthcoming). Fǎxiǎn (法顯) and Xuánzàng (玄奘) refer to the location where the present discourse was believed to have been spoken; cf. T 2085 at T LI 862c4 and T 2087 at T LI 925a26. with examples from Gandhāra showing the Buddha seated in the cave, surrounded by spectators in respectful adoration.[n.22] Cf. below p. 668f plates 2 and 3; plate 2 has already been published in Kurita 2003: 171 fig. 331, plate 3 in Foucher 1905: 493 fig. 246.

What makes these two iconographies particularly relevant to my present discussion is that in plate 2 flames can be seen to emerge from the Buddha's shoulders, and in plate 3 a circle of flames surrounds the outlines of the cave. Such a mode of depiction would be in line with the Dīrgha-āgama's textual account. However, the depiction of flames emerging from someone's shoulders can also just express that the person in question is meditating.[n.23] Waldschmidt 1930: 4 notes the tendency to depict flames emerging from the shoulders of a Buddha or an arahant as a way of expressing their supernormal power; cf. also Schlingloff 2015: 34, who explains that fire has been a symbol for meditation since ancient times and flames emerging from the Buddha or a monk serve to represent their absorbed condition: "Feuer ist das aus alter Zeit überkommene Symbol für die Meditation; Flammen, die ein Buddha oder ein Mönch ausströmt, zeigen seinen Trancezustand an." Rhi 1991: 75 note 76 points out that, in the case of depictions of the fire miracle leading to the conversion of the Kassapa brothers, "no representations of this theme in art depict shoulder flames on the Buddha, although the fire shrine was sometimes represented as being enveloped in flames … the representation of shoulder flames seems to have been limited to the 'visit of Indra'." In relation to figure 2, Coomaraswamy 1928: 37 considers the flames to be just reflecting the "effect produced by the presence of deities, Indra and his following"; cf. also Coomaraswamy 1928: 39, who in relation to DN 21 reasons that "there we learn, not only that the representation of flaming rocks is appropriate in this scene, but that this illumination is not, as usual in Buddhist legend, to be connected with the glory of the Buddha himself, but with that of the gods … this explains the representation of flames in all the reliefs." The same metaphorical nuance is already relevant to the aniconic stage in ancient Indian Buddhist art, where for episodes like the first meditation under the Jambu tree the artists represented the presence of the Buddha-to-be simply by means of a fire,[n.24] Stache-Weiske 1990: 110; for another study of fire symbolism in Buddhist art cf. Taddei 1974. as can be seen in a specimen from Bodhgaya.[n.25] Cf. below p. 670 plate 4, which has already been published in Cunningham 1892 plate 8 figure 11; for further publications cf. the survey in Schlingloff 2000: 56.

Textual examples for a metaphorical use of the fire motif in relation to this particular episode can be found in the Lalitavistara and the Mahāvastu, where the bodhisattva's father describes the splendour of his son seated in meditation under the Jambu tree by comparing him to a sacrificial fire on a mountain top.[n.26] Lefmann 1902: 132,13 and Senart 1890: 47,14. In the Divyāvadāna the fire image then illustrates the splendour of the Buddha in general.[n.27] Cowell and Neil 1886: 158,24. A similar usage can be found in the Dhammapada and its parallels, which employ the fire motif to describe the Buddha's brilliance;[n.28] Dhp 387 indicates that the Buddha shines like fire, a comparison that has parallels in stanza 50 in the Gāndhārī Dharmapada, Brough 1962/2001: 125, stanza 39 in the Patna Dharmapada, Cone 1989: 114, and stanza 74 of chapter 33 in the Udānavarga, Bernhard 1965: 501. and the Sutta-nipāta illustrates the brilliance of the new-born bodhisattva with fire imagery.[n.29] Sn 687 describes Asita seeing the bodhisattva as a young prince who is "blazing like fire". In sum, the relationship of fire in a figurative sense to the Buddha, and in particular to his meditation practice, is well attested in art and texts.

Elsewhere I have argued that, in the case of the Buddha's descent from the Heaven of the Thirty-three, it seems fairly probable that a textual motif inspired art and artistic representation, which in turn gave rise to a literal interpretation evident in textual accounts.[n.30] Anālayo 2012i: 20; for a discussion of a comparable instance in relation to the narrative and artistic depiction of the conception of the future Buddha cf. Foucher 1949: 38, and in relation to the thirty-two marks Anālayo 2017a. In the case of my present topic I wonder if a similar process of cross-fertilization between text and art might underpin textual depictions of the meditating Buddha emanating actual fire.[n.31] Zhu 2009: 501f and 504 argues for another closely related instance of crossfertilization, where iconography depicting the scene of Sakka's visit would in turn have influenced Buddhist texts in China. In this way the metaphorical motif of the 'fire of samādhi' used in art would have supported the idea of a samādhi that results in the visible appearance of fire as a phenomenon evident to those in the vicinity, independent of their engaging in, or even having any proficiency in, meditative practice.

In view of the well-established notion that the arrival of devas can result in lighting up the whole place, the idea that in the scene depicted in the Sakkapañha-sutta the Buddha should in some way be responsible for this phenomenon is not natural. Although the fire motif does seem to be a bit out of place and not a natural product of the narrative scene, the present instance is not as unequivocal as the Buddha's descent from the Heaven of the Thirtythree, where the influence of artistic representation must have played a crucial part.

In contrast, the description in the Dīrgha-āgama parallel to the Sakkapañha-sutta could simply be a result of textual literalism, although I would surmise that such literalism, if not originating from artistic representations, would certainly have been encouraged by them. I will return to this hypothesis at the end of this chapter.

---

A Visit to Brahma

My next example continues the theme of the Buddha's superiority to ancient Indian gods, evident in all versions of the Sakkapañha-sutta. In this next example, however, the same tendency manifests in relation to Brahmā instead. The narrative plot in the relevant Saṃyutta-nikāya discourse and its parallels in the two Saṃyukta-āgama collections extant in Chinese depicts the Buddha and some of his senior disciples humbling the pride of a conceited Brahmā, who believed that nobody was able to reach him in his lofty celestial abode. The passage relevant to my discussion occurs at the beginning of the discourse, after the conceited belief of this Brahmā has been introduced. The Saṃyukta-āgama extant in the Taishō edition as entry 99 describes what happens next as follows:[n.32] SĀ 1196 at T II 324c20 to 324c23.

At that time the Blessed One knew the thought that the Brahmā had in his mind. He entered a concentration attainment of such a type that he disappeared from Sāvatthī and appeared in the Brahmā's heavenly palace, seated cross-legged, with straight body and collected mindfulness, in mid-air above the head of that Brahmā.

The Saṃyutta-nikāya version of the same event differs in so far as it brings in the fire element.[n.33] I already drew attention to this difference in Anālayo 2011a: 14 note 7; the same has also been noted by Choong 2014: 186 note 27. Here is the relevant passage:[n.34] SN 6.5 at SN I 144,13 to 144,17 (in Ee the text is partially abbreviated).

Then the Blessed One, knowing with his mind the thought in the mind of the Brahmā, just as a strong man might stretch a bent arm or bend a stretched arm, he disappeared from Jeta's Grove and appeared in that Brahmā realm. Then the Blessed One sat cross-legged in the air above that Brahmā, having attained the fire element.

According to the explanation provided in the Pāli commentary on this passage, the expression "having attained the fire element" implies that the Buddha was manifesting flames emerging from his whole body.[n.35] Spk I 212,19. Another version of the Buddha's visit to this Brahmā, found in the partially preserved Saṃyukta-āgama extant in the Taishō edition as entry 100, does not mention any fire display.[n.36] SĀ² 109 at T II 412c22 only reports that "at that time the Blessed One entered concentration and, disappearing from Jambudīpa, he appeared seated in midair above the Brahmā."

Similarly to the case of the Dīrgha-āgama parallel to the Sakkapañha-sutta, in the present case, too, the bringing in of the fire motif seems a bit out of place. According to the narrative context, the issue at stake is to humble the pride of the Brahmā who had thought himself to be in such an elevated position that nobody could reach him. As the rationale for the Buddha's visit is to dispel this illusory belief of the Brahmā, an appearance in mid air suffices to make the point. The circumstance that according to all versions the Buddha even sat above the head of the Brahmā fully drives home the message of the Buddha's superiority, visually conveying that the Buddha not only reaches this Brahmā easily, but is actually superior to him.[n.37] Conversely, as noted by Strong 2008: 117, "magical flyers of all sorts are unable to fly over the Buddha … it is as though there is a superior force field that forces them to stop, land, and pay their respects to the Blessed One." On the related motif of the impossibility for anyone to look down on the top of the Buddha's head cf. Durt 1967. The manifestation of fire seems an unnecessary element in this context.

Whereas the manifestation of fire in the Saṃyutta-nikāya version might be a later addition, for the Buddha and his disciples to commute freely to different heavenly realms is a recurrent feature in the early discourses. The three versions of the present discourse in fact agree in this respect. The three versions also agree that not only the Buddha but also some of his chief disciples joined the meeting, similarly appearing in the realm of this Brahmā. According to the Saṃyutta-nikāya account, these disciples also manifested fire.

The motif of a disciple of the Buddha manifesting fire recurs in the Udāna account of an act of actual self-cremation undertaken by the monk Dabba.[n.38] Ud 8.9 at Ud 92,33 (an event the Buddha then reports to his monks in Ud 8.10) and its parallels SĀ 1076 at T II 280c7 and SĀ2 15 at T II 378b9. In a comparative study of this episode I suggested that the depiction of self-cremation in this tale might be the result of a literal interpretation of a metaphor that illustrates the nature of an arahant with the example of a burning splinter that flies up into the air and is then extinguished.[n.39] Anālayo 2012b. Several aspects of the resultant story recur in later texts like the Saddharmapuṇḍarīka-sūtra and others. These eventually came to provide a model for the real-life undertaking of self-cremation in fourth-century China and later times,[n.40] Cf. especially the detailed study by Benn 2007 (further references can be found in Anālayo 2012b). dramatically showing the impact of literalism on the living tradition.

---

Pāṭikaputta's Challenge

The next example of fire miracles of the Buddha I have chosen for study is found in the Pāṭika-sutta and its parallel. The narrative plot of the Pāṭika-sutta involves the monk Sunakkhatta, who according to the Pāli commentarial tradition was one of the Buddha's attendants before Ānanda took this role.[n.41] Spk I 258,24; cf. also Malalasekera 1938/1998: 1206, Eimer 1987: 110f, and Eimer and Tsering 2012: 1. In the Pāṭika-sutta Sunakkhatta wants to disavow the training because the Buddha had not displayed any miracles.[n.42] DN 24 at DN III 3,10 and DĀ 15 at T I 66a27; DĀ 15 has been translated into German by Weller 1928. Sunakkhatta recurs in MN 12 at MN I 68,7 and its parallel T 757 at T XVII 591c19, where he has in the meantime disrobed. Sunakkhatta then defames the Buddha for his presumed inability to perform miracles, a passage which leads Evans 2012a: 132 to the conclusion that "one who insists that Gotama lacks superhuman states will be reborn in hell, and this seems to be the case even if the utterer is telling the truth about what he or she believes." This does not seem to reflect the situation depicted in MN 12 correctly. In DN 24 Sunakkhatta still features as a monk, showing that its narration should be considered to be an earlier episode than the one in MN 12. From this it follows that the claims he reportedly makes in MN 12 need to be read as having been made in spite of all the proofs that according to DN 24 the Buddha had given of his abilities. Thus the point made in MN 12 is about intentional defamation of the Buddha, and it is on this account that Sunakkhatta is reckoned liable to rebirth in hell. The Buddha clarifies that he had never promised to display miracles in the first place. The discourse continues with some episodes that set a contrast between Sunakkhatta being impressed by some ascetics and the Buddha's ability to predict with precision how these ascetics will soon reveal their lack of true accomplishment.[n.43] Whereas such predictions are indeed instances of divination (cf. Latin divinare: "to foresee, foretell, predict"), the same does not hold for the exercise of the divine eye, pace Fiordalis 2014: 97, who proposed that "the notion of the divine eye of knowledge of the arising and passing away of beings nicely captures … the spatial and temporal dimensions of divination." The divine eye rather seems to be conceived of as the ability to witness directly, with the mind's eye as it were, the arising and passing away of beings that take place at the moment this ability is exercised.

The last of these episodes involves the ascetic Pāṭikaputta, who publicly boasts that he will best the Buddha in performing miracles.[n.44] According to his claim, however many miracles the Buddha might perform, he was ready to perform twice as many; cf. DN 24 at DN III 13,3 and DĀ 15 at T I 67c17; part of this claim has also been preserved in SHT IV 32.7 R, Sander and Waldschmidt 1980: 113. Several sources attribute a similar claim to the six teachers as part of the events leading up to the Buddha's performance of the twin miracle at Sāvatthī; cf., e.g., the Dharmaguptaka Vinaya, T 1428 at T XXII 947a2, the Mūlasarvāstivāda Vinaya, T 1451 at T XXIV 329b16 and D 6 da 42a2 or Q 1035 ne 39b1, the Divyāvadāna, Cowell and Neil 1886: 149,4, and the Chinese counterpart to the Aṭṭhakavagga, T 198 at T IV 180c14. The Buddha goes to Pāṭikaputta's place, predicting that Pāṭikaputta will be too afraid even to meet him face to face. This is indeed what happens. Even after being repeatedly urged by spectators, who have come to witness the anticipated competition in miraculous performances, to come forward to meet the Buddha, Pāṭikaputta is too afraid to face the Buddha and fails to live up to his earlier claims. The scene ends with the Buddha giving a teaching to the crowd that has assembled at Pāṭikaputta's place. In the Dīrgha-āgama version, the Buddha concludes his own description of this episode as follows:[n.45] DĀ 15 at T I 69a25f.

I taught the Dharma to that great assembly in many ways, explaining, benefitting, and delighting them. Having in that assembly thrice roared a lion's roar, I rose into the air with my body and returned to the place where I had been before.

In the corresponding section in the Pāṭika-sutta of the Dīgha-nikāya, the Buddha reports his departure in this way:[n.46] DN 24 at DN III 27,9 to 27,15.

Having instructed, urged, roused, and gladdened that assembly with a talk on the Dharma, having made them become delivered from great bondage, having rescued eighty-four thousand beings from the great abyss, I attained the fire element, rose into the air to the height of seven palm trees, created a flame another seven palm trees high, blazing and fuming, and reappeared in the Gabled Hall in the Great Wood.

Elements specific to the presentation in the Pāṭika-sutta are the liberating effect of the teaching given by the Buddha and his manifestation of fire. Given that the Dīrgha-āgama discourse does depict the Buddha departing by way of levitation and thus by manifesting a supernormal feat, it can safely be assumed that it would also have reported the display of fire, had this idea already been around when the Dīrgha-āgama passage translated above reached its present form.

A Sanskrit fragment parallel has preserved part of this episode. In the fragment the Buddha similarly reports his "having instructed, urged, roused and gladdened that assembly with a talk on the Dharma", followed immediately by the name Sunakṣatra Lecchavīputra in the accusative.[n.47] SHT IV 165.3+4 V5, Sander and Waldschmidt 1980: 178: t(aṃ pariṣadaṃ dhārmyā katha)yā sandarśa(yitvā samādā)[pa]yitvā sa[mu](ttejayi]tvā saṃpraharṣayitvā (suna)kṣatraṃ lecchav[ī](putraṃ). Even though the fragment unfortunately stops at this point, it seems unmistakably clear that here the Buddha's report of what happened continues directly by indicating that he addressed his attendant, instead of giving any description of the way in which he departed.[n.48] Schlingloff 2015: 94 note 30 has already pointed out that the Sanskrit fragment version does not report the miracle depicted in DN 24.

This leaves open the possiblity that even the ascent into the air, described in the Dīgha-nikāya and Dīrgha-āgama versions alike, could be a later development. On this assumption, the Sanskrit fragment would have preserved an earlier version in which the Buddha, having described his delivery of a talk on the Dharma to the assembly, simply continues by reporting what he then said to his attendant Sunakkhatta. This would be in keeping with the pattern observed for the previously reported episodes involving Sunakkhatta, where each time the Buddha reports what he said to his attendant. In line with this pattern the Pāṭika-sutta continues, right after the description of the Buddha's miraculous departure, with his report of what he said to Sunakkhatta.[n.49] DN 24 at DN III 27,9: taṃ parisaṃ dhammiyā kathāya sandassetvā samādapetvā samuttejetvā sampahaṃsetvā, followed by the reference to the delivery of eighty-four thousand beings and the Buddha's miraculous departure, and then at DN III 27,18: sunakkhattaṃ licchaviputtaṃ etad avocaṃ. The two parts I have given in Pāli match the Sanskrit fragment quoted in the previous note, making it safe to restore it to taṃ pariṣadaṃ dhārmyā kathayā sandarśayitvā samādāpayitvā samuttejayitvā saṃpraharṣayitvā sunakṣatraṃ lecchavīputraṃ etad avocaṃ. DĀ 15 at T I 69a27 is of no help here, as it directly turns from the Buddha's departure to the next episode in the discourse.

Support for the assumption that the Buddha's miraculous departure is indeed a later element can be found in the Pāṭika-sutta itself, precisely in what the Buddha reportedly said to Sunakkhatta at this narrative juncture. After getting Sunakkhatta to confirm that Pāṭikaputta had acted exactly as the Buddha had predicted he would, the Buddha concludes:

What do you think, Sunakkhatta, given that this is the case, has a miracle of a nature beyond [the ability of ordinary] men been performed or not?[n.50] DN 24 at DN III 27,24 to 27,26.

Sunakkhatta confirms that this is indeed the case. This exchange relates back to the theme at the outset of the discourse and Sunakkhatta's wish to disavow the training because the Buddha did not display any "miracle", iddhipāṭihāriya, "of a nature beyond [the ability of ordinary] men", uttarimanussadhamma.[n.51] On the term uttarimanussadhamma cf. Anālayo 2008f. Now, had the Buddha at the present narrative juncture been wanting to press the point that he had performed miracles, his act of levitation would certainly have been more impressive than his prediction that Pāṭikaputta will be unable to meet him face to face, and even more impressive would have been his manifesting fire.[n.52] This has already been pointed out by Weller 1922/1987: 635f. The fact that they are not mentioned here at all gives the impression that, at the time when the present passage was formulated, the idea that the Buddha had performed an act of levitation and manifested fire had not yet arisen.

The impression that the episode in the Pāṭika-sutta translated above is late is reinforced by its reference to rescuing eighty-four thousand beings from the great abyss and delivering them from great bondage. According to the commentarial explanation,[n.53] Sv III 829,22. the great bondage mentioned here is the great bondage of defilements, kilesa, and the abyss stands for the four floods, ogha, which are the floods of sensuality, becoming, views, and ignorance.[n.54] For a listing of the four floods cf., e.g., DN 33 at DN III 230,11 and its Sanskrit fragment parallel, Stache-Rosen 1968: 117. Being delivered from the bondage of defilements and rescued from these four floods would entail that these eighty-four thousand beings all attained full awakening during the talk given by the Buddha.

According to the preceding section in the discourse, Sunakkhatta had called together various Licchavīs to witness the contest in miraculous abilities between the Buddha and Pāṭikaputta. It follows that the assembled crowd should be understood to have been various spectators from the nearby location. The presentation in the Pāṭika-sutta thereby implies that a single talk on the Dharma turned this whole crowd of spectators into arahants. Although the early discourses do recognize the possibility that lay people can become arahants,[n.55] On laity and the attainment of arahant-ship cf. Anālayo 2010g: 61f note 2. this usually is conceived of as the outcome of considerable practice and acquaintance with the teachings, not as something that such a large group of chance spectators could attain during a single meeting with the Buddha. Such a hyperbolic depiction of the effects of the Buddha's teaching is uncommon at least among the early discourses.[n.56] Another such reference to attainment by eighty-four thousand beings occurs in DN 14 at DN II 44,3 as part of a description of the followers of the former Buddha Vipassī. Having on an early occasion attained stream-entry and gone forth under him, these eighty-four thousand attain full awakening on being "instructed, urged, roused, and gladdened with a talk on the Dharma" by the Buddha Vipassī. In the parallels the attainment of full awakening of those monks is not just the result of a talk on the Dharma by the Buddha Vipassī; it rather happens after the performance of the three miracles. According to the Sanskrit fragment version, Waldschmidt 1956: 154f, the two chief disciples of Vipaśyī, Khaṇḍa, and Tiṣya, performed the miracle of psychic power, ṛddhiprātihārya, and the miracle of telepathy, ādeśanāprātihārya, respectively, and the Buddha himself performed the miracle of instruction, anuśāsanāprātihārya. DĀ 1 at T I 9c18 and T 3 at T I 157b18 report that the Buddha himself performed all of these three miracles, leading to the attainment of arahant-ship by the eighty-four thousand (or else eighty thousand) monks. None of these versions comes close to the indication in DN 24 that a similarly sized crowd could reach full awakening on the spot during a single discourse by the Buddha. The apparent lateness of this description further confirms the impression that this part of the Pāṭika-sutta has gone through some development.

Given that Pāṭikaputta has been thoroughly defeated, and the whole crowd of spectators has been successfully converted, at the present narrative juncture the performance of any miracle is quite superfluous, be it an act of levitation or the manifestation of fire.[n.57] Walshe 1987: 598 note 749 comments: "could this peculiarly unnecessary miracle have been inserted later?"; cf. also Rhys Davids and Rhys Davids 1921: 2f.

I will return to this part of the Pāṭika-sutta and its parallels in the next chapter in relation to the topic of levitation.

---

The Twin Miracle

Continuing with my main theme of fire miracles, a well-known instance of the Buddha manifesting fire is the famous twin miracle. Although not reported in Pāli discourse literature, descriptions of this miracle can be found in the Pāli commentaries. According to the Jātaka commentary, one of the four instances when (according to the Theravāda tradition) the Buddha performed the twin miracle was precisely on the occasion described in the Pāṭika-sutta.[n.58] Jā I 77,23; in fact Neumann 1912/2004: 778 note 760 sees the present passage in DN 14 as the starting point for a development eventually leading to the depiction of the twin miracle at Sāvatthī (cf. also the parallelism between DN 24 and several depictions of the Śrāvastī miracle in the claim to double the Buddha's abilities, mentioned above p. 265 note 44). On the motif of the twin miracle cf., e.g., Foucher 1909: 10–16, Waldschmidt 1930, Lüders 1941/1966: 62–73, Brown 1984, Verardi 1988: 1540f, Karetzky 1990: 72f, Rhi 1991, Schlingloff 1991, Skilling 1997: 303–315, Schlingloff 2000: 488–515, Zhu 2006: 255f, Anālayo 2009l, Fiordalis 2010/2011: 401f, and Anālayo 2015i. Thus by the time this commentarial gloss came into existence, the miracle believed to have been performed by the Buddha after Pāṭika's discomfiture had developed from being a manifestation of fire alone to a miracle that combines this with the simultaneous manifestation of water.[n.59] This seems to be comparable in kind to the apparent development of the miracle related to Sakka's visit where, as mentioned above p. 255 note 14, in a Tibetan version translated in Schiefner 1849/1851: 255 the Buddha apparently manifested fire for seven days.

A reference to the Buddha performing the twin miracle can also be found in the Saṃyukta-āgama parallel to the Āditta-sutta (or Ādittapariyāya-sutta). According to the traditional reckoning, the Āditta-sutta is one of the chief discourses delivered by the Buddha soon after his awakening. The Āditta-sutta has the three Kassapa brothers and their Jaṭila followers as its audience, whom the Buddha had earlier impressed by performing various miracles.[n.60] For a survey of textual accounts and artistic representations cf., e.g., Zin 2006: 136–166. Having in the meantime become Buddhist monks, at the present narrative juncture they receive a penetrating instruction that leads them to full awakening.

According to the Āditta-sutta and its Theravāda Vinaya counterpart, the talk delivered by the Buddha on this occasion presented all aspects of sense-experience as being "on fire",[n.61] SN 35.28 at SN IV 19,24 (= Vin I 34,16). For comparative studies that take into account some of the parallels cf. Bareau 1963: 317–320 and Fiordalis 2010/ 2011: 398f. a way of teaching apparently adjusted to the interest of the discourse's audience in fire worship.[n.62] On the fire imagery cf. Gombrich 2009: 111ff. The Saṃyukta-āgama parallel reports that on this occasion the Buddha displayed the three miracles (the miracle of psychic power, the miracle of telepathy, and the miracle of instruction). The Saṃyukta-āgama discourse's depiction of the Buddha's display of the first miracle of psychic power is as follows:[n.63] SĀ 197 at T II 50b18 to 50b23.

The Blessed One entered into an attainment of concentration appropriate for the manifestation of his ascent into the air towards the east to perform [the miracle of psychic power] in the four postures of walking, standing, sitting, and reclining. He entered into concentration on fire, emitting flames of various types in blue, yellow, red, white, crimson, and crystal colours.[n.64] A description of the same miracle in the Saṅghabhedavastu of the Mūlasarvāstivāda Vinaya also reports that the Buddha manifested flames of various colours, having attained meditation on fire or the fire element; cf. Gnoli 1977: 230,18: tejodhātum samāpannasya, T 1450 at T XXIV 134b9: 入火光定, Waldschmidt 1962: 319,9 (§26.6): me’i ting nge ’dzin la snyoms par zhugs pa (the transliteration style has been adjusted). He manifested fire and water concurrently. The lower part of his body emitted fire and the upper part of his body emitted water, or else the upper part of his body emitted fire and the lower part of his body emitted water. In the same way he kept going around the four directions. Then, having performed various miracles, the Blessed One sat among the assembly.

After having displayed the miracle of telepathy as well, the Buddha gives his talk on all aspects of sense-experience being on fire. The Buddha's display of all three miracles on this occasion is also recorded in a discourse in the Ekottarika-āgama,[n.65] EĀ 24.5 at T II 622b10. in the Catuṣpariṣat-sūtra, [n.66] Waldschmidt 1962: 316,9 (§26.3). in the Dharmaguptaka, Mahīśāsaka, and Mūlasarvāstivāda Vinayas,[n.67] T 1428 at T XXII 797a13, T 1421 at XXII 109b25, and Gnoli 1977: 230,12, with its Chinese and Tibetan counterparts in T 1450 at T XXIV 134b6 and Waldschmidt 1962: 317,12 (§26.3); cf. also Dhammadinnā 2015: 42f. and in some biographies of the Buddha preserved in Chinese translation.[n.68] T 185 at T III 483a11, T 191 at T III 962a14, and T 196 at T IV 152a1; cf. also the brief reference in T 189 at T III 650a23, which appears to mean the same, and the comparative survey in Waldschmidt 1951/1967: 193.

Although the events leading up to the conversion of the Jaṭilas are clearly pervaded by the topic of miracle performance, when it comes to the actual instruction at the present juncture in the narrative it seems unnecessary for the Buddha to perform further miracles. Once the Jaṭilas have gained sufficient faith to go forth under the Buddha, there would be no need to continue to improve on his earlier performance of miraculous feats by undertaking the twin miracle. Instead, what fits the present occasion well is the "miracle of instruction", the Buddha building his presentation skilfully on a theme of central importance to these Jaṭilas before their conversion, putting their concern with fire to use for cultivating liberating insight.

Judging from the narrative context it seems probable that the Āditta-sutta and its Theravāda Vinaya counterpart have preserved an earlier version of the Buddha's third sermon in this respect,[n.69] This has already been suggested by Bareau 1963: 320. when the theme of displaying miracles, leading to the conversion of the three Kassapa brothers and their following, had not yet spilled over into the occasion of giving them the teaching that led to their liberation.

Another miraculous event relating the Buddha to fire is reported in the Mahāparinibbāna-sutta and a range of parallels. When the Buddha had passed away, his corpse could not be burned until Mahākassapa arrived. Once Mahākassapa had come and paid his respects, the pyre spontaneously ignited.[n.70] DN 16 at DN II 164,2, a Sanskrit fragment parallel, Waldschmidt 1951: 430,5 (§49.21), DĀ 2 at T I 29a26, T 6 at T I 190a12, T 7 at T I 207a11, the Dharmaguptaka Vinaya, T 1428 at T XXII 966c11, the (Haimavata?) Vinayamātṛkā, T 1463 at T XXIV 818a4, the Mūlasarvāstivāda Vinaya, T 1451 at T XXIV 401b19, and its Tibetan counterpart in Waldschmidt 1951: 431,12 (§49.21). According to Mahāvaṃsa 17.44 and 31.99, Geiger 1958: 137,13 and 254,11, even the Buddha's relics manifested fire as part of performances of the twin miracle in Sri Lanka, during which the relics rose up into the air to the height of seven palm trees; on miracles in the Mahāvaṃsa cf. also Scheible 2010/2011. Halkias 2015: 178 holds that "self-immolations are intimately related to Buddha Śākyamuni, who is reported by some influential recountings to have ended his own life by auto-cremation." This is not correct. The episode in the Mahāparinirvāṇa narrative of the auto-combustion of the Buddha's corpse took place after he had passed away and thus has no direct relationship to the ending of his life. An 'intimate relationship to self-immolation' emerges only much later in the history of Chinese Buddhism, where according to Benn 2007: 37 "the imitation of the Buddha's parinirvāṇa and subsequent cremation is suggested in many accounts of auto-cremation in China." In this case, too, some versions do not report such a miracle and instead indicate that either lay people or else Mahākassapa himself ignited the pyre.[n.71] According to T 5 at T I 174b11 and the Sarvāstivāda Vinaya, T 1435 at T XXIII 446a20, the pyre was lit by householders, and according to the Mahāsāṅghika Vinaya, T 1425 at T XXII 490b20, by Mahākāśyapa; cf. also Waldschmidt 1948: 305.

---

The Fire Element

Among Pāli discourses found in the four Nikāyas, the visit to Brahmā (2) and the Pāṭika-sutta (3) are the only references to an attaining of the fire element that I have been able to identify.[n.72] Instances found outside the four Nikāyas are Ud 8.9 at Ud 92,33 (= Ud 8.10), mentioned above p. 263 note 38, and performances of miracles in the Vinaya; cf. Vin I 25,5, Vin II 76,4, (=Vin III 159,21), and Vin IV 109,8. These report the Buddha's subduing of a serpent (as part of the conversion story of the Kassapa brothers), the monk Dabba's ability to set fire to his finger and use this to show the way to incoming monks late at night, and the monk Sāgata's subduing of a serpent. In both cases comparison with the parallel versions makes it probable that these references are the results of later developments.

Elsewhere in the four Pāli Nikāyas the fire element occurs regularly in meditative contexts, but in such occurrences the fire element comes together with other elements and is moreover something to be experienced instead of being attained.[n.73] An occurrence of the fire element on its own that is related to meditation can be found in AN 1.14.4 at AN I 25,14, which lists the monk Sāgata as foremost in (meditative) ability regarding the fire element, obviously a reference related to Vin IV 109,8, mentioned above in the previous note. In such contexts the fire element can be part of a set of four elements or else part of a set of six elements, which in addition to earth, water, and wind also comprises space and consciousness.[n.74] An example would be DN 33 at DN III 247,19 (§6.16) and its parallels in Sanskrit fragments, Stache-Rosen 1968: 165 (§6.15), and DĀ 9 at T I 52a6 (§6.10).

The implications of the fire element in such listings are spelled out in the Mahāhatthipadopama-sutta and the Dhātuvibhaṅga-sutta, together with their parallels. The two discourses differ only in so far as the Mahāhatthipadopama-sutta takes up the fire element as part of the set of four, whereas the Dhātuvibhaṅga-sutta presents it as part of the set of six elements. The two discourses and their parallels agree, however, that the fire element stands for physical warmth and bodily manifestations of heat.[n.75] MN 28 at MN I 188,5 and its parallel MĀ 30 at T I 465c16; MN 140 at MN III 241,13 and its parallels MĀ 162 at T I 691a5, T 511 at T XIV 780a23, and D 4094 ju 37b3 or Q 5595 tu 41a2. Here is the relevant passage from the Madhyama-āgama parallel to the Dhātuvibhaṅga-sutta:

Now in this body of mine there is the internal fire element, which was received at birth. What is it? That is, it is bodily heat, bodily warmth, bodily discomfort, warmth from bodily nourishment, that is, from digesting food and drink, and whatever else is of this nature and exists in this body internally, is contained in it internally, and is fiery, of a fiery nature, and hot internally.[n.76] MĀ 162 at T I 691a5 to 691a8.

The description of the fire element is not concerned with a visual apperception of flames, but rather with the physical experience of heat.[n.77] Soper 1950: 73 reasons, in relation to fire miracles and kasiṇa meditation, that "what seems to have begun in the Pāli tradition as a part of the technique of meditation – the adept visualizing fire as he would the other elements – developed in an age of miracles into the exteriorization of the fire element." It seems to me that the starting point for a process of exteriorization of the fire element in the Pāli discourses, however, did not yet involve a meditative form of visualization. This implication then informs meditative approaches to the fire element, which are about the experience of warmth as one of the characteristics of matter and which do not require any form of visualization. In this sense the fire element features among the objects of satipaṭṭhāna meditation,[n.78] MN 10 at MN I 57,37 (= DN 22 at DN II 294,16) and its parallels MĀ 98 at T I 583b18 and EĀ 12.1 at T II 568a24. and can lead to insight into the absence of a self.[n.79] AN 4.177 at AN II 165,3 and its parallel SĀ 465 at T II 119a3. For arahants it is in turn characteristic that they will be free from any notion of a self in regard to the fire element (as well as the other elements).[n.80] MN 112 at MN III 31,28 and its parallel MĀ 187 at T I 733a5.

This well-attested use of the fire element in the sense of warmth as part of a set of elements contrasts to the fire element in the Pāṭika-sutta and in the Saṃyutta-nikāya report of the visit to Brahmā, where the fire element is something to be attained and that attainment then results in the visual manifestation of fire. Since these two instances appear to be late, it seems safe to conclude that this alternative use of the term "fire element" reflects a later development.

Fire also features in a list of ten "totalities", kasiṇa, where it similarly occurs preceded by the earth and water kasiṇas, and is followed by the kasiṇas of wind, space, and consciousness. In addition to these six, the remaining four that make up the full list of ten kasiṇas are colours.[n.81] Cf., e.g., AN 10.26 at AN V 47,11 and its parallel SĀ 549 at T II 143a23 or AN 10.29 at AN V 60,19 and its parallel MĀ 215 at T I 800b5.

Regarding these remaining four kasiṇas, the Visuddhimagga describes how one fashions an object of the corresponding colour as the basis for meditation practice.[n.82] In order to cultivate the meditative vision of the blue kasiṇa, for example, Vism 173,2 suggests arranging blue flowers on a tray, covering the mouth of a bucket with a blue cloth, or fashioning a disk coloured blue as the basis. In fact the commentaries employ the term kasiṇa for such objects rather than for the "totality" of experience that is to result from their use.[n.83] Vetter 1988: 66f comments that "in the Visuddhimagga we no longer find that a Kasina sphere is considered as being immeasurable. It is even characteristic of the technique that one first concentrates on a limited" object; on the kasiṇas cf. also Karunaratna 1996, Wynne 2007: 31–34, and Chapple 2014. Although in the case of colours the situation is quite straightforward, with the other six kasiṇas the idea of contemplating them as visual objects is not without difficulties.

Already the kasiṇas of wind and space are not easily experienced as visual objects, as wind can be observed mainly by its effects and space becomes visible only in the form of an absence of visual objects. In the case of the consciousness kasiṇa, it seems hardly possible to think of a way of turning this into a visual object. In fact the Visuddhimagga drops consciousness from the list in its exposition of kasiṇa practice and replaces it with the light kasiṇa.[n.84] Vism 174,19 (although Vism 609,14 does refer to the consciousness kasiṇa as one of various objects for the cultivation of penetrative insight); cf. also Anālayo 2009j: 668 and 2011b: 592f note 33.

As far as the early discourses are concerned, it seems improbable that the whole set of ten kasiṇas was meant to refer to visual experiences, pace later exegesis. Besides, even in the Visuddhimagga the fire kasiṇa is something experienced internally and there is no indication that others, on seeing a meditator engaged in this practice, will also be able to apperceive fire. In fact, had the fire kasiṇa been the starting point for the idea of seeing someone emanate fire, one would expect the Pāṭika-sutta and the Saṃyutta-nikāya discourse on the visit to Brahmā to speak of attaining the "fire kasiṇa", instead of referring to the attainment of the "fire element".

In view of this it seems to me quite possible that the notion of fire manifesting outwardly as a result of someone having attained the fire element would have been facilitated by some form of pictorial depiction. This could have been a sort of canvas taken around to aid oral performance with some visual stimulation, a usage already attested in a discourse in the Saṃyutta-nikāya.[n.85] SN 22.100 at SN III 151,23 (the formulation in the parallel SĀ 267 at T II 69c18 could be due to a misunderstanding of the translator; cf. Anālayo 2013c: 45 note 119) seems to reflect the employment of pictures as aids in oral teaching. According to Spk II 327,18 it refers to a canvas with paintings taken around by wandering brahmins to illustrate teachings on karma and its fruits; cf. also Mair 1988: 17–37, Dehejia 1990: 377, and Brown 1997: 81, as well as on a similar custom in medieval China Teiser 1988: 446. The use of such a type of visual depiction could have influenced the texts already at a comparatively early stage in their transmission.

---

Conclusion

The selected examples of fire miracles performed by the Buddha surveyed in this chapter seem to be for the most part identifiable as later developments, probably the result of literal interpretations of metaphorical usages of the fire motif attested in text and art.[n.86] As pointed out by Gombrich 1996: 21, "unintentional literalism has been a major force for change in the early doctrinal history of Buddhism." At the same time, examining these instances clearly testifies to other type of supernormal abilities as integral to the early Buddhist world. Celestial travels by the Buddha and his disciples, a topic to which I return in the next chapter, or devas lighting up a place on arrival form a common heritage among the early discourses and it would not be doing justice to their testimony if one were to consider all such elements as the result of later influences.

Taken together this suggests to my mind the appropriateness of a middle-way approach to the topic of supernormal feats and miracles in early Buddhist thought. This middle-way approach steers clear of ignoring supernormal elements and according the status of genuine teachings only to the type of textual material that conforms to modern-day Western expectations and values.

At the same time this middle-way approach also avoids reading the early discourses through the lenses of later tradition and ignoring the development in the depiction of miracles,[n.87] An example of this tendency would be McClintock 2011, who on the basis of some miracle tales in the commentary on the Dhammapada comes to the conclusion that in early Indian Buddhist narratives the Buddha functions as a trickster. Such uncritical employment of one Pāli commentary as if it were to reflect the whole of early Buddhist narrative traditions needs to be counterbalanced by taking into account literature reflecting the commentarial period that has been preserved outside the Pāli canon and, even more importantly, historical contextualization of such tales by way of comparison with what can be gathered about the Buddha's role in the material common to the discourses in the four Pāli Nikāyas and their Chinese Āgama parallels, where I for one am not aware of evidence that would corroborate an assessment of the Buddha's role as that of a trickster. the historical stages of which can be detected with the help of comparative study of versions of a text transmitted by different reciter traditions.