Notes

n.1We have translated the title of this text based on the Tibetan (bcom ldan ’das kyi ye shes rgyas pa’i mdo sde rin po che mtha’ yas pa mthar phyin pa) while considering the Sanskrit title provided in the Tibetan manuscripts. In the process we have attempted our own emendation of the Sanskrit title, which we believe is the product of a back-translation from the Tibetan. The revised Sanskrit title that we suggest using for this text is: ananta­niṣṭhāga­bhagavajjñāna­vaipulya­sūtra­ratna. In arriving at this title, we have been guided by the following reflections: The Tōhoku catalog lists the title, which its compilers likewise attempted to revise, as niṣṭhāgata­bhagavajjñāna­vaipulya­sūtra­ratnānanta. However, the Tōhoku title includes a footnote for niṣṭhāgata that mentions an alternative reading of niṣṭhāgan. This has led us to believe that the original reading most probably was niṣṭhāga, and not niṣṭhāgata, since the meaning of this term is better suited in this context (see further below). The difference in meaning between these two terms is that niṣṭhāgata means “arrived at certainty” (i.e., “conclusive / definitive”), whereas niṣṭhāga can also mean “leading to certainty.” The Sanskrit title given on the title page of the Degé Kangyur reads niṣṭhatan-bhagavat-jñāna-vaipulyan-sūtra-ratna-ānanta. Other Kangyurs reflect variants, although most of them are minor. Most notably, a few Kangyurs (e.g. the Stok Palace as representative of the Thempangma line) are missing ānanta (mtha’ yas pa). Also, although the correct form might be niṣṭhāga(ta), all of the Tibetan editions that we consulted read niṣṭhatan (although ā > a is common and ga could have been elided in copying). Most importantly, however, they all place this term at the beginning of the title, and not at the end, where it appears in the Tibetan. Given this introductory placement of niṣṭhāga in the Sanskrit title, we believe that so also should ananta be moved to the beginning of the title as a qualifier of niṣṭhāga. One could of course adopt the reading of the Stok Palace Kangyur where ananta / mtha’ yas pa is missing altogether, which would also yield a straightforward title. However, by moving ananta to the beginning of the title (and thus conforming to the Tibetan where mtha’ yas pa and mthar phyin pa are placed next to each other in that sequence) one gets the compound anantaniṣṭhāga , which we believe is the better option. The phrase anantaniṣṭhā (or its synonym atyantaniṣṭhā ) actually occurs in other Buddhist texts, including the Lalitavistara, where, in a description of the Dharma wheel, we find the sentence akopyaṃ taccakram atyanta­niṣṭha­tvāt, meaning, “This wheel cannot be shaken, because of the infinite certitude [of the Dharma].” As such, it is also possible that the original Sanskrit might have read atyanta­niṣṭhāga rather than anantaniṣṭhāga. With this word order, the Sanskrit title becomes plausible and makes sense, both in terms of grammar and meaning. Significantly, in this way we also arrive at a Sanskrit title that can actually be read as a basis for the Tibetan translation, which is a feature missing from the Sanskrit as it is listed in the Tibetan manuscripts as well as the Tōhoku catalog. Thus, with our proposed emendations to the Sanskrit title, the Sanskrit and the Tibetan titles are reconciled.

n.2Csoma de Körös’s summary of the sūtra was later published in French translation by Henri Léon Feer (1881).

n.3The Denkarma (ldan dkar ma), see bibliography, was compiled by Paltsek (dpal brstegs), Lui Wangpo (lu’i dbang po), Namkhai Nyingpo (nam mkha’ snying po), and others.

n.4See Yao (2021), 2.91.

n.5See Rotman (2008), pp. 71–117.

n.6E.g., the exact role of Pūrṇa and the name of the place, see n.­8, n.­12, n.­23, and n.­25. It is noteworthy that Sarvajñādeva, who translated the Bhaiṣajyavastu, is also one of the revisers of this sūtra; that no attempt appears to have been made by the translators to reconcile these narrative differences is therefore unlikely to be because the translators of the one text were unaware of the existence of the other. They were, probably, simply staying faithful to their source texts.

n.7At least six are mentioned in the Kangyur: (1) Pūrṇa Maitrāyaṇīputra, a brahmin from Kapilavastu, ordained by his uncle Ājñātakauṇḍinya when the latter returned to Kapilavastu soon after the Buddha’s first teaching; this is the Pūrṇa who was “foremost in teaching” among the ten principal disciples, and is mentioned in many sūtras including The Teaching of Vimalakīrti (Toh 176) and The Sūtra of Pūrṇa’s Questions (Toh 61); (2) the Pūrṇa who was one of the second group of five monks ordained by the Buddha, the “five friends” (nye lnga sde), all Vārāṇasī merchants’ sons, headed by Yaśas; (3) the Pūrṇa of The Exemplary Tale of Pūrṇa (Pūrṇāvadāna), son of a wealthy Aparāntaka merchant and his slave girl, a successful maritime expedition leader before going forth as a monk, and almost certainly the protagonist in the present sūtra; (4) an older Pūrṇa, the “Elder Pūrṇa from Kuṇḍopadāna,” who is also mentioned in The Exemplary Tale of Pūrṇa as one of the monks in the Buddha’s airborne entourage; (5) a very rich and generous brahmin called Pūrṇa from the Mountains of the South who invites the Buddha and receives a prediction of enlightenment, but is not ordained; he is the subject of the first story in The Hundred Exemplary Tales, Beginning with That of Pūrṇa (Pūrṇapramukhāvadānaśataka, Toh 343); and (6) the sickly and short-lived Pūrṇa of Śrāvasti, attendant of Aniruddha, who became an arhat just before he died and is the subject of one of the stories in the first chapter of The Hundred Deeds (Karmaśataka, Toh 340).

n.8This is one of the differences between this version and the episode in The Exemplary Tale of Pūrṇa, according to which Pūrṇa himself is the main instigator and patron of the building project.

n.9The list, with some brief explanations, is to be found in the Mahāvyutpatti (119–129). For a list of canonical references to the ten powers, see Lamotte 2001, pp. 1234–35. As Lamotte points out, there are numerous minor variations in the order of these ten powers in the various canonical sources. See also Lamotte, pp. 1235–79, for a detailed presentation of each of the ten powers.

n.10To further complicate matters, the recurring phrase that Pūrṇa uses to describe how the ten powers and their subdivisions relate to the Buddha’s knowledge of the four truths is not entirely straightforward. The Tibetan‍—the structure of which is repeated throughout the text‍—reads (we have here used “X” to mark the changing topics): de bzhin gshegs pas X shes pa yang dag pa ji lta ba bzhin rab tu khyen / X shes pa kun ’byung shes pa yang dag pa ji lta ba bzhin rab tu khyen / X shes pa ’gog pa shes pa yang dag pa ji lta ba bzhin rab tu khyen / X shes pa ’gog par ’gyur ba’i lam shes pa yang dag pa ji lta ba bzhin rab tu khyen /. Our translation of this structure is as follows: “The Thus-Gone One correctly understands the knowledge of X. He correctly understands the knowledge of the origin as related to knowledge of X. He correctly understands the knowledge of cessation as related to knowledge of X. He correctly understands the knowledge of the path that leads to cessation as related to knowledge of X.” As we read these sentences, we take “X shes pa” to be the object of the Buddha’s understanding of the knowledge of the individual truths.

n.11In his thought-provoking article, “Buddhist Hybrid English,” Paul Griffiths discourages translators from reproducing such repetitive passages (which he finds “paralysingly boring”) on the grounds that such repetition is unlikely to yield significant insights in readers (Griffiths 1981, p. 25). While Griffiths is no doubt right that modern readers may find such extensive repetition difficult to digest and hard to appreciate in terms of literary value, it nevertheless seems important to present texts like this in their totality to allow readers of all kinds, specialist or not, to appreciate the styles and registers of Indian Buddhist literature for themselves.

n.12In the version of The Exemplary Tale of Pūrṇa, it is Pūrṇa himself who performs the offerings and makes the invitation.

n.13See Salomon 2018, pp. 199–228.

n.14See The Chapter on Medicines (Bhaiṣajyavastu, ch. 6 of the Vinayavastu, Toh 1), 9.­1524 ff. and n.­938.

n.15The mention, which in this translation is at 1.­192, is found in his commentary on the Mūla­madhyamaka­kārikas, p. 6 (see bibliography under Chomden Rikpai Raldri). Mistaken views are classified and enumerated in many different ways in different texts, but this enumeration of sixty-two mistaken views is probably derived from the Brahmājāla Sūtra, and is also found in the Prajñāpāramitā texts and the Lotus Sūtra.

n.16Butön, vol. 2, p. 81. The passage he cites from the sūtra is: “For example, just as [the Thus-Gone One] knows that this excellent eon is called ‘Vision of One Thousand Lotuses,’ he knows the names for each of the eons” (1.­390). This point may have been first pointed out by Sakya Jetsun Drakpa Gyaltsen (1147–1216) the third of the Five Patriarchs of the Sakya tradition.

n.17See for instance the Lhasa Kangyur dkar chag as reproduced in the Comparative Edition of the Kangyur, vol. 107, p. 93).

n.18las kyi bdud, see 3.­634. Gorampa’s citation is in Gorampa Sönam Senge, p. 11.

n.19Taktsang Lotsawa Sherab Rinchen (stag tshang lo tsa ba shes rab rin chen), p. 203. The passage in question is 3.­214.

n.20pha rol tu shes pa mi ’jug pa. Gö Lotsāwa Shönnu Pal, p. 16.

n.21Drakpa Döndrub, pp. 122–123. The passage in the text on this topic, as well as on the lifespans of the hells and other realms, starts at 1.­286.

n.22Chökyi Drakpa, p. 16.

n.23grong khyer chen po dge ba’i pha rol ’gro zhes bya ba. No place whose name has this exact Tibetan rendering is found elsewhere in the canonical texts, and it has no attested Sanskrit equivalent. There is, however, a very close match in the Gaṇḍavyūha (Degé Kangyur, vol. 38, phal chen, a, F.65.b; see also Roberts, 2021): dge ba’i pha rol tu phyin pa, Śubhapāraṃgama in Sanskrit, the southern city where the householder bodhisattva Veṣṭhila lives‍—although this may well be an allegorical rather than a geographical name. Intriguingly, Veṣṭhila worships at a sandalwood shrine. Whatever the case, the stated location in the present text differs from that of the temple in The Exemplary Tale of Pūrṇa, which is built in the coastal city Sūrpāraka (Tib. slo ma lta bu)‍—the capital of Aparānta and Pūrṇa’s native city, identified with modern Sopara, just to the north of Mumbai.

n.24The Degé and other Kangyurs of predominantly Tshalpa (tshal pa) influence read phywa mkhan, while the Stok Palace (stog pho brang) and Shelkar (shel mkhar) Kangyurs of mainly Thempangma (them spangs ma) lineage have phya mkhan. Both spellings are found in other texts in all Kangyurs, and appear to represent alternative spellings of the same term. Csoma de Körös (p. 426) and Henri Léon Feer (p. 231) both rendered phywa mkhan as “fortune teller,” a sense it may have in some contexts (although in such cases “diviner” might be a better rendering). However, the Mahā­vyutpatti (3770) lists phya mkhan as the Sanskrit stha­pati, which Monier Williams translates as “chief, governor, architect, master builder, etc.” This seems the more likely sense given the context, and in a number of Vinaya and Prajñāpāramitā texts in the Kangyur the term (in both spellings) clearly has that meaning.

n.25The role of the “master builder” therefore seems to be not merely contractual; he appears to have contributed to the costs and to have a stake in the project in terms of the merit it will create. The relationship between the master builder and Pūrṇa is not made fully clear, but the story differs in this respect from that of The Exemplary Tale of Pūrṇa, in which Pūrṇa is the instigator of the building project and pays for it with the proceeds of his brother’s expedition (which is also the source of the sandalwood).

n.26The subheadings in the rest of this chapter are not present in the Tibetan but have been added to make the translation more navigable (see Contents list), to allow some correlation with the stories in the third chapter, and to be compared to the list of ten powers provided in the glossary and introduction (see i.­9, and also i.­10).

n.27It is unclear to us what is meant by “the eight human existences” (mi’i srid pa brgyad).

n.28From the abhidharma perspective it is not possible for a woman to manifest as a fully awakened buddha, since this would imply having male physical characteristics.

n.29The faculties mentioned in the passage that follows are the twenty-two faculties, see Glossary.

n.30The mention here of “the path” is anomalous; unlike all the other items in the list, it is missing the word “faculty,” and neither in the detailed explanation of the twenty-two faculties that follows, nor in usual list of twenty-two faculties, is there a “path” item. In the Urga Kangyur it has been left out.

n.31The chapter break, which is not present in the Tibetan, has been added here at this change in theme to make the translation more navigable.

n.32We have been unable to identify this nāga and are unsure of its translation. The Tibetan reads tshib pa’i bu. Below, an alternative spelling of this name (chib pa’i bu) is found.

n.33Here the text reads “Mahā­megha” but since this is a repeat (it appears four names before this) we have exchanged this name with “Sūrya­garbha,” which appears on the verse praises below in this position.

n.34The story that follows of the Buddha visiting the nāga king Anavatapta’s lake (also known as Lake Anavatapta) is modeled on the Anavataptagāthā (“Verses of Lake Anavatapta”), a passage‍—or rather a genre of passages‍—found in a range of early Buddhist corpora in Sanskrit, Gāndhārī, Chinese, and Tibetan. The fullest extant version is probably that in the Mūlasarvāstivāda-vinaya literature: see The Chapter on Medicines (Bhaiṣajyavastu, ch. 6 of the Vinayavastu, Toh 1), 9.­1531–9.­2505 and n.­938; see also Salomon 2008, and Salomon 2018, pp. 199–228. In the present text, the narrative starts off with an account of the contest of supernatural powers between Mahā­maudgalyāyana and Śāriputra very similar to that in The Chapter on Medicines (although that passage is not found in the Gandhārī version). The rest of the content, which continues into the following section, is very different, but reflects the Anavataptagāthā at least in the telling of events in the Buddha’s past lives that explain features of his present one. Another Mahāyāna sūtra, The Questions of the Nāga King Anavatapta (Anavatapta­nāgarāja­paripṛcchā, Toh 156), is also the record of a teaching by the Buddha given partly at Lake Anavatapta at the nāga king Anavatapta’s request, but is quite different in content.

n.36It appears that one line is missing from this verse.

n.37From here onward, some of the names of the nāga kings are not the same as they were when they were introduced earlier in the text (on folio 114.b). For instance, this one is now nag po instead of mig nag pa.

n.38bsgyings pa. This seems to be either an alternate name or mistranslation of Viraja (rdul bral).

n.39Monier Williams notes that this term, presumably onomatopoeic, is also applied to other animals including the cuckoo, ruddy goose, frog, and house-lizard.

n.40Tibetan: ’dul ba. Sanskrit: vinaya.

n.41The chapter break, which is not present in the Tibetan, has been added here at this change in theme to make the translation more navigable.

n.42The subheadings in this chapter are not present in the Tibetan, but have been added to make the translation more navigable.

n.43de bzhin gshegs pa bar snang la khrid mdzad ces bya ba ’jig rten tu byung ste / / de skye ba dang rga ba dang/ na ba dang / ’chi ba las yang dag par ’da’ ba’i chos ston to zhes de skad thos te. This translation is tentative.

n.44It appears that a line in the Tibetan was either condensed in translation into another line or left out.

n.45The text literally reads “knowledge of time.” However, we have changed this to “knowledge of karma,” since this is the topic of this chapter.

n.46The knowledge of adopted actions, contemplations of abodes, contemplations of causes, and the knowledge of ripening are not included.

n.47de lta bas na chos gang gis rgyu byas shes chos de brtag par bya ste / mi dge ba’i phyogs kyi tshul bzhin yid la byed pas ni brtags pa’i rgyu ces bya’o / dge ba’i phyogs la tshul bzhin yid la byed pas ni byas pa’i rgyu zhes bya ste. This translation is provisional.

n.48bya tsa go ra’i mig can, presumably Skt. cakora-dṛś, i.e., “having beautiful eyes.”

n.49The term ’phar ba is known to render a number of different Sanksrit words, such as uddhata, udvṛtta, pluta, anibhṛta, audbilya, etc., none of which presents a clear choice in the current context; thus the translation “flood” (a central meaning of pluta) remains questionable.

n.50snod kyi rgya thams cad ’dri ba zhig byung ngo. The meaning of this line is unclear. The word snod frequently translates the Sanskrit word pātra, which besides “vessel” can also carry the connotation of “a royal minister.”

n.51This may refer to a state of partial mastery of the zil gyis gnon pa’i skye mched, or “sphere of subjugation” (Skt.: abhi­bhvāyatana), one of eight stages of control over the senses.

n.52The text indicates that a stock phrase should be inserted in this sentence. The entire phrase that is commonly found in this place is: “They were all worthy ones whose defilements were exhausted. They were without afflictions and controlled. Their minds were perfectly free, their insight perfectly liberated. They were noble beings, great elephants, successful and accomplished. They had laid down their burden and fulfilled their aims. They had eliminated the bondages of existence and, thanks to their correct knowledge, their minds were perfectly liberated. They had all obtained supreme perfection in mastering all mental states.” The same applies to the passage that follows just below.

n.53Here the text also indicates that a stock phrase should be inserted in this sentence. However, we have been unable to identify which phrase that is.

n.54Here the Tibetan text indicates that a stock phrase should be inserted. However, we have been unable to determine which phrase that is.

n.55Here again, the text indicates that the following stock phrase should be inserted: “They were all worthy ones whose defilements were exhausted. They were without afflictions and controlled. Their minds were perfectly free, their insight perfectly liberated. They were noble beings, great elephants, successful and accomplished. They had laid down their burden and fulfilled their aims. They had eliminated the bondages of existence and, thanks to their correct knowledge, their minds were perfectly liberated. They had all obtained supreme perfection in mastering all mental states.”

n.56This verse has only three lines in the Tibetan.

n.57Tibetan: shing ljon pa phan pa med pa. Presumably a metaphor to describe the qualities of the monk in the passage that follows.

n.58lnga pa; Skt: pañcama. The reference here is unclear and in any case seems to appear as part of this greater list that serves as a trope in Buddhist sūtras to describe the knowledge of those of wide learning. It may imply a “Fifth Veda,” a term used at times to describe the Mahā­bhārata and several other texts, or perhaps some other topic that is commonly listed as fifth in a list of fields of knowledge.

n.59Tibetan: bged pa. We have been unable to determine the meaning of this verb with certainty.

n.60The precise meaning of this analogy is a bit unclear. It could have to do with the quality of placidness generally associated with elephants in Buddhist literature, or perhaps more specifically with the major mark of awakened beings entailed by their possession of a sexual organ that retracts like an elephant’s.

n.61Here the name of the king changes from Immaculate Consecration (dri med dbang bskur) to Consecrated by Elixir (bdud rtsis dbang bskur). However, we have chosen to continue using the name Immaculate Consecration.

n.62Translation tentative. Tibetan: ’khor de snang bar byas nas su.

n.63Translation tentative. Tibetan: smras pa khyed cag la gang gi tshe bzung ba dang bcas pa rnyed pa de’i tshe khyed cag ’jigs sam/.

n.64Translation tentative. Tibetan: tshor ba thams cad gdul ba’i don du tshor ba bdag gi don rtogs ste/.

n.65Translation tentative. Tibetan: grogs po dag tshor ba rnams gdul ba’i don du yun ring por brjod du med par sangs rgyas kyi zhing gi sems can rnams byang chub tu yongs su smin par byas so/.

n.66It is unclear to us what this line refers to.

n.67Here the text reverts to a presentation of knowledge of the world.

n.68It seems that some text is missing here as this paragraph ends abruptly with no apparent relation to the following text.

n.69Up to this point the significance of the present paragraph remains unclear to us.

n.70Reading mchog gi bstan pa based on the Choné, Lithang, Narthang, Kangxi, and Lhasa Kangyurs. The Degé reads mchog gis bstan pa.

n.71Translation tentative. Tibetan: yon gnas la dga’ ba / sbyin bdag sgra thams cad la mkhas pa zhig byung ngo/.

n.72This paragraph remains obscure to us and our translation is therefore only tentative.

n.73The following verses are obscure and our translation is therefore only tentative.

n.74Translation tentative. Tibetan: bkug shud kyis rtse.

n.75The speaker of these last four lines appears to be the sea gods although the text does not indicate so.

n.76Following le lo med pa yang rnyed according to the Choné, Narthang, Kangxi, Yongle, and Lhasa Kangyurs. The Degé reads le lo med pa yang rnyid.

n.77Tibetan: shes pa bar mas.

n.78It is not clear to us where this statement concerning Dīpaṃkara is found.

n.79Reading bgod pa according to the Kangxi and Yongle Kangyurs. The Degé Kangyur instead has bged pa, and the Choné ’god pa.

n.80Translation tentative. Tibetan: de bzhin gshegs pa de dag la srid pa’i ’gro bar ’chi ba dang / skye ba de dang de dag tu mdud par ’dzin pa rjes su dran te / de bzhin gshegs pa de dag bzhugs pa dang / yongs su mya ngan las ’das pa de dag la bdud kyi las rnams byas so/.

n.81Reading sdang according to Narthang and Lhasa Kangyurs. The Degé reads snang.

n.82Tibetan: bu med gang dag gi lus she pa nyam chung.

n.83Literally “gods of the grass” (Tibetan: rtswa’i lha).

n.84Translation tentative. Tibetan: ji ltar las byas dag gi rnam par smin pa ’grub pa ’dir skye bo thams cad la mngon sum yin ba mi rigs pas yongs su bsgyur nas bdag gi las kyi ’bras bu chud gzon to/

n.85Reading bgog par ’gyur instead of bgog par mi ’gyur/.

n.86khyod ni mngon par mkhyen pa chos ’di’i tshul lags so/.

n.87Tibetan: kha dog tsam gnyis pa tsam mo/. The translation here remains tentative.

n.88We have been unable to determine the name of this sage. In Tibetan the name is spelled: ud dza+ya dad ta.

n.89Following the bsten that appears in the Choné, Lithang, Narthang, Kangxi, Yongle, and Lhasa Kangyurs. The Degé reads bstan.

n.90Tibetan: ’du shes yod pa ma yin par smra ba. Given the information that is supplied by the subsequent specification of the various views, the Tibetan may be corrupt here.

n.91It is not quite clear how the four verbal confusions are enumerated.

n.92Considering the earlier reference to “four views that teach that there is no origin” and the enumeration of a set of eighteen views that follows immediately below this passage, it appears that two “views that teach that there is no origin” remain unaccounted for here.

n.93Tentative translation. Tibetan: gang gi tshe bdag gzugs can de ’jig rten pha rol nas shin nas mi’i bzang po yongs su spyad de chad par ’gyur/.

n.94Tentative translation. Tibetan: de da ltar de dag gzhan kyis kyang nges par ’byung bar blta ste/ de dag phal cher ni nges par ’byung bar mi blta’o/.

n.95Tibetan: byung ba chen po bzhi ’byung ba tha mi dad pa dus gsum mnyam pa nyid la gnas pa’i gsang tshig ’don to/.

n.96The translation of this analogy is tentative.