Introduction

i.1The Sūtra of the Moon (2) is a short sūtra explaining how a lunar eclipse involves the covering of the moon by the asura Rāhu. While the Buddha Śākyamuni is staying at the Traveler’s Pond in Campā, the moon is covered by Rāhu, lord of the asuras. Candramas, the god of the moon, asks the Buddha for refuge, after which the Buddha urges Rāhu to release the moon. Seeing that Rāhu has set the moon free, Bali, another lord of the asuras, asks Rāhu why he did so. The asura explains that if he had not released the moon, his head would have split into seven pieces. The text concludes with a verse praising the emergence of buddhas.

i.2The Sūtra of the Moon (2) is one of a few closely related texts in the Kangyur, versions of which are also found in the Chinese Āgamas and the Pali Nikāyas of the Theravāda canon. Different versions of this text are also included in the Paritta and Rakṣā collections, which have the function of providing both spiritual and worldly protection through their recitation.

i.3In the Kangyur there are two texts with the title The Sūtra of the Moon, both of which have the same basic content. There is the text translated here, designated The Sūtra of the Moon (2) (Toh 331), which was translated from Sanskrit in the time of the “early diffusion,” and The Sūtra of the Moon (1) (Toh 42), which was probably translated from Pali in the fourteenth century. An almost identical narrative is also found in The Sūtra of the Sun (Toh 41) concerning a solar eclipse. The Sūtra of the Moon (1) and The Sūtra of the Sun remain popular among Tibetans today, appearing in recent Tibetan collections of mantras and incantations for recitation. Six of seven verses from The Sūtra of the Moon (2) are also quoted verbatim in Great Upholder of the Secret Mantra, Toh 563, 1.33, one of the Tibetan Pañcarakṣā texts.

i.4Neither the colophons, nor the imperial catalogs, nor Tibetan historical works mention the translators of The Sūtra of the Moon (2). However, the fact that The Sūtra of the Moon (2) is listed in the Denkarma catalog from the early ninth century confirms that the translation was probably produced either during or prior to that period, probably from Sanskrit.

i.5The Sūtra of the Moon (2) is available in the Kangyurs of the Tshalpa, Thempangma, and mixed lines, in independent collections such as the Phukdrak Kangyur, the Langdo collection, and the Namgyal Kangyur, and in some collections from Western Tibet, namely Gondhla and Tholing. It is also included in Hemis I Kangyur from Ladakh. All versions of this text are to be found in the Sūtra section of the respective collections.

i.6The colophons of the majority of the Kangyurs only indicate the conclusion of the sūtra, but the Thengpangma texts have the following observation at the colophon: “It is evident that this and the sūtra translated by Tharpa Lotsāwa are the same,” referring to The Sūtra of the Moon (1). In the same way, the colophon of The Sūtra of the Moon (1) in the Narthang edition states that “there is also an early translation,” referring to The Sūtra of the Moon (2). Although the Tibetan texts present the Sanskrit title Candrasūtra, this title is not attested in Sanskrit works, but the title Candra­maṇḍala­sūtra is documented in a Sanskrit fragment from Central Asia that lists different sūtras.

i.7No complete Sanskrit version of The Sūtra of the Moon (2) exists, but Waldschmidt and Klaus Wille prepared annotated editions based on Sanskrit fragments that correspond to this text. It is also extant in two Chinese translations of the Saṃyuktāgama. It is likewise quoted in the *Mahā­prajñā­pāramitā­śāstra (大智度論).

i.8In the Pali Saṃyutta Nikāya we find the Candimasutta (SN 2.9), which is the probably close to the source text for The Sūtra of the Moon (1), but differs slightly from The Sūtra of the Moon (2). Some passages from the Chinese Āgamas were translated into Old Uyghur, and there are two different Old Uyghur renditions of the verses of The Sūtra of the Moon (2) from Taishō 99, both edited by Zieme. Feer presents the legend of Rāhu according to Hindu and Buddhist texts, and his translations of both versions of The Sūtra of the Moon include extensive notes.

i.9The Sūtra of the Moon (2), the Sanskrit fragments edited by Klaus Wille, and the Chinese canonical translations in the Saṃyuktāgama would seem to belong to related textual traditions, while The Sūtra of the Moon (1) and the Pali Candimasutta can be grouped together as representing the Theravāda tradition. They all share the same basic content, but the texts differ in wording and in the number of verses and passages in prose. For example, The Sūtra of the Moon (2) has seven verses, while The Sūtra of the Moon (1) has only four. Among them the setting also differs in that only The Sūtra of the Moon (2) is set in Campā, while all the other versions are set in Śrāvastī. The characters differ too. For instance, in The Sūtra of the Moon (1) and the Pali Candimasutta, it is Vemacitra who speaks to Rāhu, not Bali, as in the other versions. Furthermore, Bali is mentioned by name in Sanskrit and Chinese, but in The Sūtra of the Moon (2) he is addressed as the great son of Virocana. The Sūtra of the Sun is almost identical in wording to The Sūtra of the Moon (1) but of course reads “sun” instead of “moon” and includes one additional verse, as is the case with the Pali canonical versions. This extra verse, in which the Buddha explains how the sun dispels darkness, corresponds to a verse in The Sūtra of the Moon (2).

i.10We have based our translation on the Degé edition while also consulting the Comparative Edition of the Kangyur, and the Stok Palace, Phukdrak, Hemis I, and Gondhla manuscripts. We compared Wille’s edition of the Sanskrit with the Tibetan. When significant variants from Tibetan Kangyur manuscripts are supported by Sanskrit, they were preferred. Our translation benefited from Feer’s notes on The Sūtra of the Moon (1) and The Sūtra of the Moon (2), as well as from his translations. We sometimes referred in the notes to The Sūtra of the Moon (1) and The Sūtra of the Sun. We also consulted passages from The Sūtra of the Moon (2) quoted in the English translation of Great Upholder of the Secret Mantra, translated by the Dharmachakra Translation Committee for 84000. We have on occasion referred to variants found in this text and its commentary.

i.11Apart from its role as a text for protection that shows the Buddha’s power to bring eclipses to an end, it is noteworthy that linguistic expressions relating to Indian views on eclipses can be heard in regions far beyond the influence of Buddhism. In present-day Turkey, for instance, Jens Peter Laut has shown the connection between Old Turkish Buddhist culture from Central Asia with Modern Turkish expressions such as “the moon was seized,” meaning that the moon is eclipsed. This same expression is found in The Sūtra of the Moon (1) and the Pali Candimasutta, though here in The Sūtra of the Moon (2) we find the expression that the moon “was covered.”