Notes

n.1In the Pāli sources he is named Subha Todeyyaputta, “Subha, son of Todeyya.” His full name in the extant Sanskrit version of the sūtra is Śuka Taudeyaputra (śuko māṇavas taudeya­putro “The brahmin youth Śuka, son of Taudeya”). The Tibetan version has bram ze’i khye’u shu ka (equivalent to Sanskrit śuko māṇava) throughout but mentions that he is the son of the brahmin Taudeya (bram ze to’u de ya) at the end of the sūtra.

n.2In many presentations, “mixed” actions (i.e., partially good and partially bad actions) and “neutral” (Skt. avyākṛta) actions complement the pair of opposites of good and bad actions, but the Karmavibhaṅga does not contain an explicit discussion of these categories.

n.3For a canonical account of the relationship between karma, the different rebirth destinies, and Buddhist cosmology, see, for example Bruno Galasek-Hul and Lama Kunga Thartse Rinpoche, trans., The Limits of Life , Toh 307 (84000: Translating the Words of the Buddha, 2021). See also Dharmachakra Translation Committee, trans., The Application of Mindfulness of the Sacred Dharma , Toh 287 (84000: Translating the Words of the Buddha, 2021), which provides an extensive magisterial overview of the workings of moral causation in all the realms of existence.

n.4Although the Karmavibhaṅga does not make use of technical terms such as retributive, outflowing, and predominating results, as the Abhidharmakośa does (see AKK 87a,b; La Vallée Poussin/Pruden 1988–90, p. 672), the three categories of “outflowing result” (niṣyandaphala), “retributive result” (vipāka), and “predominating result” (adhipattiphala) are tacitly employed by way of example. At 1.­145-1.­154, the predominating and outflowing results are illustrated, while in most of the paragraphs up to 1.­132 the retributive result is illustrated. The retributive result is produced when a virtuous or nonvirtuous action is accumulated (upacita), i.e., carried out repeatedly and with intent, i.e., full awareness, and results in rebirth in a state of suffering in one of the lower rebirth destinations; the outflowing result is produced when the person who has performed the action is‍—due to other karmic forces‍—reborn as a human being, and it results in an experience equivalent to the action; finally, the predominating result is produced through intensively and/or repeatedly engaging in the ten virtuous actions or the ten nonvirtuous actions and results in a suitable or unsuitable environment.

n.5See Gombrich 2009, pp. 13, 127–28, and, specifically, 123: “Among the five khandha, the fourth group, volitions, includes cetanā, intention. This the Buddha declared to be what constitutes karma and therefore lends an action its ethical quality, whether good or bad.”

n.6AN III, 415,7–8: Cetanāhaṃ bhikkhave kammaṃ vadāmi; cetayitvā kammaṃ karoti kāyena vācāya manasā. Readers who are interested in more specialized descriptions and a more comprehensive treatment of the topic of karma and rebirth in ancient India and in Buddhism may consult the H-Buddhism Buddhist Studies Bibliography project on Zotero (filter tag “karma”), accessed April 9, 2020. In particular, two excellent in-depth studies are recommended: Halbfass 2000 (in German) and Timme Kragh 2006.

n.7In the extensive appendix to his edition of the Karmavibhaṅga, Kudo (2004, pp. 229–323) catalogs and comments on the known parallels of the cited texts and stories in the order of their appearance.

n.8For an excellent, exhaustive overview of the extant versions of the Karmavibhaṅga, see Maggi 1995, pp. 19–20.

n.9Lokesh Chandra 2010, p. 29.

n.10For photographs of the reliefs of the hidden base of the Borobudur accompanied by English translations from the Sanskrit of individual paragraphs of the Karmavibhaṅga, see Ānandajoti Bhikkhu, “Karma-vibhanga, Deeds and their Results,” Photo Dharma, last accessed July 24, 2020.

n.11See Dharmachakra Translation Committee, trans., The Play in Full, Toh 95 (84000: Translating the Words of the Buddha, 2013), 4.­30.

n.12See, e.g., pw, s.v. “vibhaṅga”; BHSD, s.v. “vibhaṅga”: (1) “classification”; (2) “explanation, commentary.” For the word vibhaṅga as part of titles of Indic Buddhist works, see the introduction to Bruno Galasek-Hul and Lama Kunga Thartse Rinpoche, trans., Transformation of Karma, Toh 339 (84000: Translating the Words of the Buddha, 2021), i.­5–i.­6. Sylvain Lévi in his 1932 edition of the Sanskrit text translated the title as La grande classification des actes (Mahākarma­vibhaṅga, The Great Classification of Acts).

n.13Lévi has identified this appendix, which was included in the manuscript bundle of MS[A], as constituting a commentary on the Karmavibhaṅga and appositely labeled it Karma­vibhaṅga-upadeśa. Lévi 1932, pp. 2, 167.

n.14Vasubandhu’s Abhidharma­kośabhāṣya refers to a work titled Mahā­karmavibhāga­sūtra, but it is doubtful that this sūtra is identical with the Karmavibhaṅga as we have it. See Abhidh-k-bh(P) 281,11 ad AKK V.6. The context of this passage in the Abhidh-k-bh is a discussion of different early Buddhist schools’ views on the removal of latent dispositions (Skt. anuśayas) through the realization of the four noble truths. No such passage, concerning neither the wording nor the doctrinal content, can be identified in the Karmavibhaṅga.

n.15Kudo 2004, p. 215.

n.16Looking at the first pages of several Kangyur editions, we find the following variations from the transliteration of the Sanskrit title in the Kangyurs of the Tshalpa group, karma bi b+hang ga (= karmavibhaṅga). The London manuscript Kangyur; Stok Palace manuscript Kangyur; and Shey Palace manuscript Kangyurs reads karmA bi b+hang ka (= karmavibhaṅga), and the Phukdrak manuscript Kangyur reads karma bi b+ha ga (= karmavibhāga). Complicating matters further, the Phukdrak manuscript Kangyur contains two versions of the las rnam par ’byed pa, both seemingly containing the same text, of which the second, F404 (vol. 97 [mdo sde, na], folios 333.b–356.a) has the transliterated Sanskrit title kar ma b+ha ba ka (= karma-bhavaka or karma-bhāvaka, “[the sūtra that explains] existence through actions”). One may want to readily dismiss this last variant of the title as a mere scribal error or the result of ignorance of the Sanskrit language. However, the Shelkar Kangyur records as the Tibetan title of the related sūtra Toh 339 (see Bruno Galasek-Hul and Lama Kunga Thartse Rinpoche, trans., Transformation of Karma , 2021) las kyis rnam par ’gyur ba zhes bya ba’i chos kyi gzhung (“The Dharma Scripture called Transformation through Karma”). It is, of course, possible that the latter is a scribal error (in Tibetan prints and manuscripts the case markers -kyis and -kyi are frequently mixed up), but both variant titles point to the central theme of the Karmavibhaṅga-group of texts: the coming to be or the transformation of existence through actions.

n.17For instance, the Degé version of the Karmavibhaṅga translated here differs from the extant Sanskrit versions of the Karmavibhaṅga‍—MS[A] and MS[B], respectively (Kudo 2004). These, although sufficiently different themselves, rather correspond with the version called S1 by Mauro Maggi (Maggi 1995, p. 19) and with Kudo’s “Tib–3” which is the same as the Tibetan version of the Berlin manuscript Kangyur studied by Walter Simon (1970) as well as the extensive Dunhuang fragment Pelliot tibétain 944 (see n.­31 for the latter).

n.18See also Bruno Galasek-Hul and Lama Kunga Thartse Rinpoche, trans., Transformation of Karma (84000: Translating the Words of the Buddha, 2021), i.­2.

n.19See Kudo 2004, p. xi: “However, we have no evidence for deciding whether or not this text is a sūtra.” See also p. ix, where Kudo summarizes the works on the Karmavibhaṅga done by other scholars. Namikawa Takayoshi, for example, by comparing the citations in the Karmavibhaṅga from the Cakravartisūtra with other parallel texts, supposes that the Karmavibhaṅga belonged to the canon of the Sāṃmatīya school of Buddhism. Given Namikawa’s hypothesis that the Cakravartisūtra belonged to the Abhidharma Piṭaka of a school that also possessed the Karmavibhaṅga, one might even further speculate that the Karmavibhaṅga, too, originally belonged to the Abhidharma Piṭaka of the Sāṃmatīya school. We have not been able to independently verify this hypothesis but rely solely on the excellent scholarship of these Japanese scholars. See also n.­19.

n.20See Kudo 2004, p. ix and pp. 262–63, n. 37. This sūtra is titled differently in MS[A] and MS[B]: Cakravartisūtra and Cakravarti­sūtravibhaṅga, respectively, and only MS[A] has the addition of Abhidharma before the text’s title. (We have normalized the Sanskrit spellings of the texts’ titles, following Lévi, and left out variants; for a transcription of the original spellings in the manuscripts, see Kudo 2004, pp. 262–63). For an obscure remark in the Karmavibhaṅga-upadeśa that “other schools” classified the Karmavibhaṅga as belonging to an otherwise unknown collection called Abhidharma­saṁyukta, see Lévi 1932, p. 12 (“un passage obscur qui semble indiquer que le Karmavibhaṅga était classé par d’autres écoles dans les Abhidharmasaṁyuktas, genre de division dont nous ne savons rien non plus”) and p. 167 (gotrāntarīyāṇāṃ Abhidharmasaṁyukteṣu).

n.21This is indicated by the phrase uddeśaḥ Karmavibhaṅgasya dharmaparyāyasya, “[Now follows] the exposition/ exemplification of the Dharma-discourse Karmavibhaṅga (Lévi 1932, p. 32). Cf. also Kudo 2004, p. 34 for MS[A] and p. 35 for MS[B].

n.22Lévi 1932, p. 2.

n.23Yet, the Karmavibhaṅga certainly does not exhibit the level of formality and scholasticism that one finds, for example, in Vasubandhu’s Abhidharma­kośabhāṣya. On the other hand, MS[B] contains not only the title Karmavibhaṅga­sūtra but also the expression “Dharma discourse” (dharmaparyāya), omitted by MS[A], which usually designates a sūtra-style way of teaching. See Lévi 1932, p. 32; Kudo 2004, p. 35.

n.24Resources for Kangyur and Tanjur Studies , Universität Wien, accessed October 21, 2019.

n.25Lozang Jamspal and Kaia Fischer, trans., The Hundred Deeds , Toh 340 (84000: Translating the Words of the Buddha, 2020).

n.26The manuscript that Sylvain Lévi examined and on which he based his edition has, according to the colophon, been dated between 1410 and 1411 ᴄᴇ (Lévi 1932, p. 1).

n.27For English translations of these texts, see Bhikkhu Ñāṇamoli and Bhikkhu Bodhi 2009.

n.28For details on the name, see Lévi 1932, p. 21, n. 3, where he suggests as the common origin to both variants of the name the Ardhamāgadhī Sua and furthermore references Śuka, son of Vyāsa, of the Mahābhārata.

n.29Maggi (1995, pp. 19–20) lists altogether six Chinese translations, which is correct. The information given in Lokesh Chandra is thus incomplete; see Lokesh Chandra 2010, pp. 27–28. We were unable to assess for this translation how the Chinese translations relate to the Tibetan or the Sanskrit versions of the Karmavibhaṅga.

n.30See Kudo 2004, p. viii; see Lokesh Chandra 2010, p. 90.

n.31According to Maggi 1995, p. 20, n. 2, a manuscript fragment of this Chinese translation was discovered at Dunhuang. It would perhaps be worthwhile to compare this Chinese Dunhuang fragment with Pelliot tibétain 944, the Tibetan fragment of the las rnam par ’byed pa found at Dunhuang.

n.32See Lewis R. Lancaster, The Korean Buddhist Canon: A Descriptive Catalogue, accessed November 4, 2019, K 805. This catalog furthermore states that Taishō 78, 79, and 81 are the Chinese equivalents of Toh 339, las kyi rnam par ’gyur ba zhes bya ba chos kyi gzhung (Bruno Galasek-Hul and Lama Kunga Thartse Rinpoche, trans., Transformation of Karma , 2021). A comparison of these Chinese works with the Tibetan translation Toh 339, as well as the Central Asian fragment in the Hoernle collection (reproduced in Lévi 1932, pp. 235–36) is so far lacking.

n.33Maggi 1995.

n.34Hoernle 1916.

n.35Lévi 1932, pp. 243–57; reprinted in a separate publication in Lévi 1933, pp. 84–107, with a French translation.

n.36Oldenburg 1920.

n.37The third version of the text, not included in the Degé Kangyur, is represented by the Tibetan translation H343 in the Lhasa Kangyur (despite the misleading entries in the dkar chag), S287 in the Stok Palace manuscript Kangyur, B346 in the Berlin manuscript Kangyur, and N784 in the Narthang’s supplementary (kha skong) volume, as well as by a Dunhuang version (PT944). For details of its content, see Simon 1970. 84000 hopes to add an English translation to this collection in future.

n.38For detailed information on Drakpa Shedrup, see s.v. at The Treasury of Lives , accessed September 12, 2022. This short work by Drakpa Shedrup is strictly speaking not a commentary as it merely summarizes the sūtra’s main points in a concise way (don bsdus) and largely follows the structure of Toh 339. The second part of Drakpa Shedrup’s text contains a concise summary of the tshe’i mtha’i mdo (Āyuḥparyanta, Toh 307). He first gives a summary of the different karmic categories, which is then followed by a short explication of the categories and the specific actions that lead to their respective result. While Drakpa Shedrup’s commentary does not add any new information that is not available from the sūtras themselves, it stands out for its conciseness and very clear language.

n.39MS[A] reads anāthapiṇḍa.

n.40The Sanskrit adds the patronymic Taudeyaputra to Śuka’s name (Lévi 1932, p. 29).

n.41D: bram ze’i bu; H bram ze’i khye’u. The Tibetan bram ze’i bu corresponds to the Sanskrit māṇavaka (see Kudo 2004, pp. 8 and 26; Mvy [Sakaki 3846]). Bhikkhu Bodhi translates the Pāli māṇava as “student” (see Bhikkhu Ñāṇamoli and Bhikkhu Bodhi 2009, p. 1053).

n.42According to the Tibetan (las rnam par ’byed pa bstan gyis). The Sanskrit reads Karmavibhaṅgaṃ te Māṇavaka dharmaparyāyaṃ deśayiṣyāmi (“Son, I will teach you the discourse on Dharma [called] The Exposition of Karma”). (Lévi 1932, p. 29; Kudo 2004, p. 27: MS[B] only, MS[A] omits.)

n.43The terms translated here are well-known Buddhist terms in the context of the teachings on karma, and our translation follows the Sanskrit and Bhikkhu Bodhi’s translation of the Pāli Cūḷakammavibhaṅga­sutta (see Bhikkhu Ñāṇamoli and Bhikkhu Bodhi 2009, pp. 1053–54). Here the Tibetan seems to be literally saying “[beings] exist as their own actions” or “[beings] become their own actions” (sems can rnams bdag gi las las su gyur ba). The terms given in the Tibetan translation do not follow the standard translation equivalents given in Mvy (Sakaki 2313): las bdag gyir byed pa for karmasvakaḥ.

n.44The Tibetan text deviates from the standard translation of karmayoniḥ, las kyi skye gnas pa (see Mvy [Sakaki 2315]). Also, the order in which the terms are presented differs from the Sanskrit editions, where karmadāyāda (see n.­45) stands before karmayoni (Lévi 1932, p. 30; Kudo 2004, pp. 26 and 27).

n.45Again, the Tibetan differs from the standard terminology of Mvy (Sakaki 2314), which gives las kyi bgo skal la spyod pa for karmadāyādaḥ.

n.46Here, the expression “they take action as their refuge” means that action is the basis for beings’ destiny in the sense that actions determine who one is or who one will become. The Sanskrit reads karma­pratiśaraṇa (Lévi 1932, p. 30; Kudo 2004, p. 26, MS[A]; p. 27, MS[B], has a lacuna here). See Mvy (Sakaki 2316): karma­pratisaraṇam = las brten par bya ba. According to Edgerton, this is a bahuvṛhi compound (BHSD, s.v. “pratisaraṇa”). The entire phrase is well known from the Pāli Canon (MN III, 203,4–6): Kammassakā, māṇava, sattā kammadāyādā kammayonī kammabandhū kammapaṭisaraṇā. Kammaṃ satte vibhajati yad idaṃ hīnappaṇītatāyāti. See Bhikkhu Ñāṇamoli and Bhikkhu Bodhi 2009, p. 1053: “Student, beings are owners of their actions, heirs of their actions; they originate from their actions, are bound to their actions, have their actions as their refuge. It is action that distinguishes beings as inferior and superior.”

n.47The Sanskrit text first gives a list of all the categories that are to be analyzed and presented in detail in the main body of the text; see Lévi 1932, pp. 30–32. This list is absent in the Tibetan translation.

n.48The Sanskrit additionally reads “welcoming the death of an enemy” (amitramaraṇābhinandanaṃ) here.

n.49This very likely refers to the practice of abortion (Sanskrit garbhaśātana). However, we have opted for a more literal translation of the Tibetan.

n.50Our translation of this sentence largely follows the Sanskrit. The commentarial gloss on the Sanskrit reads as follows: “The children [lit. sons] and grandchildren of the originator of this sacrifice, as well as other people, hoping for a [positive, i.e., desired] result [of the ritual action of sacrifice] or filled with fear [i.e., of potential negative consequences of not performing the sacrifice], [will] kill many beings as they continue [this initial sacrifice]” (tasya yajñapravartakasya putrāḥ pautrāś cānye ca janāḥ phalārthino bhayabhītāś cānuvṛttiṃ kurvāṇāḥ sattvān nirghātayanti, Lévi 1932, p. 32; see Kudo 2004, pp. 36 and 37, with slightly different reading in MS[B] without, however, altering the meaning). The sense of repetition and thus establishing an act as a custom or tradition (anuvṛttiṃ √kṛ) is borne out by the illustrative story that follows.

n.51Following the reading of D: sgos kyis (see Jäschke 1972, s.vv. “sgos” and “phyir”). Y and K read dgos kyis?

n.52According to the Sanskrit, all editions of which read yuddhadarśanaṃ (Lévi 1932, p. 33; Kudo 2004, pp. 40 and 41). Tibetan reads dmag (“army” or “host”), the expected translation equivalent of which, according to Mvy, is the Sanskrit senā, etc.

n.53C reads mang por dog pa. If dog pa could be read as synonymous with nyam nga (“fear, despair; harm”), C should be read and the Degé reading corrected: “[Seeing] a sacrificial ground such as this is like a battle, seeing the fear and despair in the many beings, humans and horses, and so forth [that are going to be killed].”

n.54Lévi translates, “and furthermore congratulating the men engaged in battle on their equipment” (et aussi les félicitations à propos des armes, adressées à des gens en train de se batter; Lévi 1932, p. 112).

n.55See Kudo 2004, pp. 229–30, n. 2.

n.56The Sanskrit editions (Lévi 1932, p. 33; Kudo 2004, pp. 40 and 41) contain an additional passage that mentions ten evil consequences (“disadvantages”) of killing living beings (without listing them) and refer to the Nandikasūtra for the full list; for the complete passage in a surviving Skt. witness of a version of the Nandikasūtra , the Ārya­nandika­pari­pṛcchā­sūtra, see Vinīta 2010, pp. 109–11.

n.57Translated according to Lévi’s conjecture of the Tibetan, which makes good sense; see Lévi 1932, p. 34, n. 2: dmag la sogs par sngar smos pa thams cad bzlog pa ste (“and having rejected/done the opposite of all that was said above regarding armies and so forth”). However, his conjecture (smon pa to smos pa) seems not to be backed by any edition of the Kangyur that we have consulted, i.e., Narthang (in Lévi 1932, p. 185), K, D, and the critical apparatus of the Comparative Edition.

n.58The Tibetan here, and throughout, has only two terms, mchod rten and gtsug lag khang, where the Sanskrit specifies three objects: stūpa , caitya, and vihāra. According to Mvy (Sakaki 6999, 7000), the Tibetan mchod rten should be used to translate both the Sanskrit stūpa and caitya. The Sanskrit word caitya can refer to a stūpa , but also to a shrine, a sacred place, or any sacred object. Thus, in addition to the place where the Buddha Śākyamuni attained awakening (the bodhimaṇḍa) and other sacred sites associated with the life of the Buddha, it can also designate non-Buddhist shrines or sacred sites (see Drewes 2007, p. 103).

n.59Tibetan zhig cing ral ba literally means “that which has dilapidated and/or was torn down.”

n.60A, D: tshe ’phrog bdud (literally, perhaps, “life-stealing demon”). This seems to be a gloss of the variations in Y, K, J, and N: ye ’grog and C: ye ’brog. According to Rolf A. Stein (see McKeown 2010, pp. 21, 34, 267, and specifically, 61–62), ye ’brog/ye ’drog/yen ’brog (the latter seemingly a phonetically influenced variant spelling) is Chinese vocabulary for ’dre gdon, a kind of evil spirit, demon, or demonic force that causes premature death and misery. (For the translations “contagious disease,” “injury inflicted on the soul,” and “harm done to the mind,” see Jäschke [1972, s.v. “ye ’brog”], who gives Milarepa’s songs and oral explanation as his sources.) The Sanskrit reads akālamṛtyur, “premature death” (Lévi 1932, p. 34; Kudo 2004, pp. 42 and 43). Thus it seems likely that the explanatory gloss tshe ’phrog bdud refers to more or less specific demonic forces (the ’dre) traditionally believed to steal the life force and cause sudden accidents and untimely death.

n.61According to the Tibetan. The Sanskrit does not mention anger.

n.62Reading Y and K: de ni thal mo dang khu tshur gyis mi rteg pa dang; D and other versions omit mi.

n.63This translation follows Lévi (1932, p. 37: “donner des remèdes et aussi des aliments digestibles”) and the Sanskrit parijīrṇabhojanaṃ. The Tibetan kha zas zhu nas stobs pa (literally “Having digested food, [they regain their] strength”) is not clear.

n.64The core meaning of the Sanskrit term varṇa (Tibetan kha dog) is “color.” In a particular sense (from as early as the Ṛgveda), it can mean the color of skin or complexion, thus the derived meanings of “species,” “class of people,” and “caste” (Mayrhofer 1976, p. 154, s.v. “várṇaḥ”). It seems that it can also refer to the general appearance or figure of a person, and it is this latter sense that is probably intended here. Lévi interprets the Sanskrit durvarṇa (Kudo 2004, p. 52) as “disgracieux” (Lévi 1932, p. 115). The following paragraph makes it clear that the Tibetan mdzes (“beautiful,” for suvarṇa) refers to a general physical feature or appearance. See also Lévi 1932, p. 37, n. 2, where he refers to plate 21 of Borobudur’s hidden base, which is inscribed with virūpa and shows representations of people with deformations.

n.65According to Mvy (Sakaki 1964), the translation equivalent of the Tibetan ’tshig pa should be the Sanskrit pradāsa. Lévi prints the variant pradāśaḥ (Lévi 1932, pp. 37–38 and n. 2), but MS[A] has paridādya, which according to Kudo should be read as paridāgha (Kudo 2004, p. 52, n. 5). Edgerton prefers the reading paridāgha over pradāsa (see BHSD, s.v. “pradāsa”). Other Abhidharma lists of the upakleśas read pradāśa/pradāsa (see, e.g., Abhidh-k-bh 57,4 [= AKK II.27b] and 313,12 [= AKK V.50a]). Pradāśa (and its variants pradāsa/paridāgha) is a Buddhist Sanskrit word. Perhaps its etymological meaning, as reflected in the Tibetan translation ’tshig pa, is “burnt” or “smolder(ing),” which may hint at an underlying feeling of suppressed or concealed anger. Kenjiu Kasawara’s translation of the Dharmasaṃgraha has “contentiousness” for pradāśa (see Müller and Wenzel 1885, no. 69: Sanskrit p. 14 [pradāśa]; English translation p. 50). Someone whose anger is “smoldering” beneath the surface may consequently be more quarrelsome, i.e., more easily given to animosity, malice, and spite. The word “contentiousness,” however, does not seem to hit the appropriate linguistic register.

n.66This translates the (lexicalized) variant reading skyang nul in Y and H (D: rkyang nul). See Jäschke 1972, s.v. “skyang nul.” The Sanskrit text reads sudhādānaṃ, “donating” (Kudo 2004, p. 52; Lévi 1932, p. 38; see also n.­67).

n.67Tibetan rdo thal (“limestone,” “chalk,” or “ashes”). Sanskrit reads stūpa­caityagṛheṣu ca sudhādānaṃ, which may suggest chalk or lime (sudhā). Lévi (1932, p. 115) translated, “donner du plâtre pour les stūpas, les maisons à caitya.” This sentence may refer to the practice of whitewashing a stūpa, a ceremony that can still be witnessed today in the Kathmandu valley at the stūpas of Svayambhūnāth and Bodhanāth.

n.68The Tibetan sdug gu here translates the Sanskrit suvarṇa, which can also mean “golden.”

n.69Tibetan mthu chung ba, Sanskrit alpeśākhya, Pāli appesakkha(ttaṃ). See Bhikkhu Bodhi’s translation “uninfluential” (Bhikkhu Ñāṇamoli and Bhikkhu Bodhi 2009, p. 1056). The Buddhist Sanskrit alpeśākhya (Pāli appesakkha) (see CPD, s.v. “appesakkha”; BHSD, s.v. “alpeśākhya”) originally means “unrespected, insignificant, of little esteem.” The scholarly consensus is that the Middle Indic form of the term is derived from an original Sanskrit form alpa-/mahā-yaśas-ka. Two forms are attested in the early Buddhist Sanskrit texts: alpa-/mahāśakya and alpe-/maheśākhya. The occurrence of either term can serve as an indicator of a text’s affiliation with the Sarvāstivādin or the Mūlasarvāstivādin school. The form alpa-/mahāśakya is surmised by Edgerton to have originated as a folk-etymological alteration of alpe-/maheśākhya (see BHSD, s.v. “mahāśakya”). The traditional analysis of the term, however, is alpa-īśa-ākhya (maheśākhya being an analogous formation), literally perhaps “named after an insignificant chief or master, of low origin” (Apte, s.v. “alpa-”). This meaning is reflected in the standardized Tibetan translation given in Mvy (Sakaki 6412): dbang chung bar grags pa.

n.70Sanskrit dharmahīne, Tibetan chos ngan pa. The Tibetan expression is usually translated as “inferior doctrines.” Here, the Sanskrit equivalent seems rather to imply that something or someone outside of Dharma is referred to (see pw, s.vv, “a-dharma,” “dharma-hīna”).

n.71The Sanskrit (Kudo 2004, p. 56) reads parasya yaśovarṇṇaśabda­ślokaśravanena [sic] tuṣṭiḥ (“being gratified when hearing about others’ glory, praise, renown, and good reputation”). The Tibetan gzhan gyi grags dang sgra dang tshig su bcad pa snyan pa thos na seems not to have translated Sanskrit varṇa (“praise, renown, glory”; Tibetan brjod pa) here, while in later paragraphs we find the list extended by brjod pa (“praise”). See 1.­165 (grags pa dang brjod pa dang sgra dang tshigs su bcad pa), 1.­172 (brjod pa dang / sgra dang tshigs su bcad pa), and 1.­173 (grags pa dang / brjod pa dang / sgra dang tshigs su bcad pa). This variation in translating a stock phrase supports the assumption that the extant text of the Karmavibhaṅga is a text that grew over time, i.e., passages and paragraphs were added on at different times. Nonetheless, grags pa can translate both the Sanskrit yaśas and varṇa, which indeed share the same sense (“praise, renown, glory”) in Sanskrit.

n.72We have supplied the word “commemoration” here to form an intelligible English sentence. The Tibetan reads, “Erecting a stūpa of/for the Bhagavān” (bcom ldan ’das kyi mchod rten … brtsigs pa). The Sanskrit expresses a causative sense (-kārāpaṇa, “causing to be built”): “commissioning the building of shrines and stūpas [in commemoration] of the Bhagavān” (bhagavānaś caityastūpa­kārāpaṇaṃ).

n.73According to the Tibetan reading of D. Neither the Tibetan (dge ba’i rtsa ba thams cad kyis mthu chen por sems bskyed pa ste) nor the corresponding Sanskrit (sarva­maheśākhyakuśalamūlaṃ ­bodhicittotpādanaṃ; Kudo 2004, p. 56‍—we have normalized Kudo’s transcription of the Sanskrit manuscript MS[A]) are clear. The Tibetan has only sems bskyed pa (cittotpāda), instead of byang chub kyi sems bskyed pa for bodhicittotpādana, and the syntax is parallel to the immediately preceding sentence (i.e., mthu chen po + la-don followed by sems bskyed pa). The Tibetan of D here seems to have preserved the better reading. The Sanskrit bodhicittotpādana does not make much sense in this context, and we think that bodhi- is a later addition. Lévi (1932, p. 39) edited the Sanskrit to read sarva­maheśākhyakuśalamūle bodhicittotpādanaṃ (Lévi’s copy seems to have read sarva­maheśākhyakuśalamūle instead of -mūlaṃ) and translated as “produire la Pensée de toutes les Racines-de-bien qui font les grands personnages” (to produce the thought [of awakening] from all the roots of good which make (the) great persons). However, Lévi (1932, p. 39, n. 4) refers to bas-relief 43 of the hidden base of the Borobudur, which bears the inscription maheśākhya (“distinguished, exalted, eminent, powerful”) and portrays a life of luxury. For a similar idea, see verse 18, chapter 31 of the Prajñāpāramitāratna­guṇasaṃcayagāthā (Yuyama 1976, p. 128: tatu vardhate kuśala–mūla mahānubhāvo candro v’ an-abhru prabha-maṇḍala śukla-pakṣo, and the Tibetan translation of it in the Dunhuang recension: de-las dge-ba’i rtsa-ba mthu-chen ’phel ’gyur-te/ /sprin med zla ba yar ngo’i ’od kyi dkyil ’khor bzhin (Yuyama 1976, p. 190) (“From that his wholesome root grows into something of great might; As the moon, in the absence of cloud, is a circle of radiant light in the bright half of the lunar month”; translation Conze 1975, p. 70). Z reads dge ba’i rtsa ba chung ngu rgyun mi gcod pa dang / dge ba’i rtsa ba chen po yang dag par ’dzin du ’jug pa’o (“Not interrupting the continuous accumulation of lesser (or small? chung ngu) roots of merit; causing [someone] to take up/engage in the [accumulation of the] greater roots of merit”). It is not clear to us, however, what the lesser and greater roots of merit would be. See the extant Central Asian Sanskrit fragment from Eastern Turkestan (the Śukasūtra; Lévi 1932, pp. 235–36): [mahā-]śakyāt kuśalamūlād vicchandanam alpaśakyānāṁ pudgalānāṁ paribhavaḥ (“discouraging [vicchandana; for this Buddhist Sanskrit word, see Wogihara 1971, p. 37 and Mvy (Sakaki 6527): vicchandyati] [someone] from [accumulating] powerful roots of merit; having contempt for persons with little power”). In any case, the passage seems to point to the idea of dedicating one’s accumulated merit to the attainment of worldly happiness, wealth, and greatness with which comes worldly power and authority.

n.74Tibetan rim gro is generally translated as “respect” or “honor.” Here, however, the meaning “to serve” or “to attend on” can be confirmed by the corresponding Sanskrit (a-)pratyupaṣṭhānaṃ in MS[A] (Kudo 2004, p. 58), used here in its meaning as found in the Pāli suttas: paccupaṭṭhāna (“tending to”). See PED, s.v.; BHSD, s.v. “pratyupasthāna (2).”

n.75This sūtra has not been identified. A similar passage, however, can be found in the Pāli Aṅguttara Nikāya of the Sutta Piṭaka (AN III, 244–45). See Kudo 2004, p. 240, n. 10; Lévi 1932, p. 40, n. 3.

n.76According to the Tibetan, which uses grong (“village”) in the first sentence and rigs (= Sanskrit kula) in the following sentences. The Sanskrit (Lévi 1932, p. 40; Kudo 2004, p. 60) reads kula (“family, household, house community”). In Old and Classical Sanskrit, the term refers to all the people with whom one shares meals (German Speisegemeinschaft), which in traditional societies may exceed the modern, so-called (nuclear) family; thus we have opted for “community.”

n.77The Tibetan longs spyod chung ba literally means “(having) few possessions/little wealth.”

n.78The Tibetan ma byin par len pa (Sanskrit adattādāna) literally means “taking what was not given.”

n.79The Sanskrit has only tadabhyanumodanam, “taking pleasure in stealing” (Kudo 2004, p. 60). An alternative translation of the Tibetan may be “taking pleasure in stolen goods” (ma byin par blangs pas dga’ ba).

n.80The Tibetan reads shes rab chung ba. The Tibetan translation equivalent of the Sanskrit duṣprajña prescribed by the Mvy is shes rab ’chal ba “misconception, wrong understanding” (see Mvy [Sakaki 2470]).

n.81According to the Sanskrit, which adds paṇḍitān (Kudo 2004, pp. 64 and 65, MS[A] and MS[B], respectively; Lévi 1932, p. 41).

n.82This sentence (chos ma yin pa ni gang yin) is missing in Y, J, K, N, and C.

n.83The Sanskrit reads, “What is Dharma/right? Practicing which Dharma promotes happiness?” (Lévi 1932, p. 41: ko dharmaḥ kiṃ dharmaṃ kurvataḥ śreyaskaram iti; Kudo 2004, pp. 64 and 65).

n.84This translates the variant reading sten in Y, J, K, and N. D has bston, which seems to be corroborated by the Sanskrit (both MS[A] and MS[B]; Kudo 2004, pp. 64 and 65, respectively) sevati (“to stay/remain with someone”; “to associate with someone”).

n.85This is a free translation of the Tibetan ’jungs pa’i sems yod pas, corresponding to the Sanskrit abhiniveśa in the extant Sanskrit editions.

n.86D reads glegs bam ’dri ba. The better reading is perhaps that of Y and K: bri ba.

n.87According to the Tibetan (yongs su ’dri phod pa). The Sanskrit term pari­pṛcchakajātīyaḥ “inclined to inquire/ a questioner” occurs in the Aṣṭasāhasrikā Prajñāpāramitā , published on the website Bibliotheca Polyglotta by the University of Oslo, accessed December 6, 2019. See also Gareth Sparham, trans., The Perfection of Wisdom in Eighteen Thousand Lines (84000: Translating the Words of the Buddha, 2022).

n.88According to the Sanskrit of MS[B] (Kudo 2004, p. 69): dharma­bhāṇakānāṃ vaiśāradyaṃ varṇṇayati. MS[A] (Kudo 2004, p. 68) reads varddhaya{ṃ}ti, which according to Kudo (2004, p. 68, n. 10) is a scribal error for varṇṇayanti. The Tibetan chos smra ba rnams la mi ’jigs pa nye bar sgrub pa dang does seems to be a standard translation of the Sanskrit; nye bar sgrub suggests some form of the Sanskrit upa + √hṛ (see Mvy [Sakaki 6395]: upasaṃhāra).

n.89Sanskrit sahita literally means “connected, sensible, coherent, reasonable” (see BHSD, s.v. “sahita”: “of speech, connected, coherent, sensible”), which seems to be the sense chosen by the Tibetan translators. The Sanskrit could also be interpreted as “one applauds/acclaims those who speak what is beneficial/good [for others] (sa-hita), and one avoids those who speak what is unbeneficial/not good (a-hita) [for others]” (sahitabhāṣiṇāṃ sādhukāraṃ dadāti | ahitabhāṣiṇaḥ pariharati according to the emended Sanskrit version in Kudo 2004, p. 69, MS[B]). Lévi (1932, p. 44, n. 6) records saṁhitabhāṣitānām; according to Kudo’s new transliteration of MS[A], however, the part up to hita- is illegible in the manuscript (Kudo 2004, p. 68).

n.90Translating Y and K smad pa instead of dmod pa (D and other Kangyur versions).

n.91For “ink,” reading N, U, and H snag tsha instead of D snag tsa. D reads glegs bam dang snag tsa dang smyug gu la sogs pa’i sbyin pa byed pa, which faithfully (except that it has glegs bam, translating pustaka, first) renders the Sanskrit masīpustakalekhanīpradādāni dadāti (according to Lévi 1932, p. 44; MS[A] and MS[B] [Kudo 2004, pp. 68 and 69], give the same reading), “making gifts of ink, books (or leaves of birch bark?), and reed pens.” For the loanwords masi and pustaka, see Falk 1993, pp. 241 and 305–6 (pustaka [“skin”] perhaps meant tree bark).

n.92For identification of the Nandikasūtra (Toh 334), see Kudo 2004, p. 230, n. 3; p. 233, n. 6; and p. 240, n. 13. For an English translation of the Tibetan version of the Nandikasūtra , see The Sūtra of Nandika (Toh 334). For an edition and English translation of the extant Skt. witness, the Ārya­nandika­pari­pṛcchā­sūtra, see Vinīta 2010, pp. 97–114.

n.93MS[A] instead mentions forty faults of drinking alcohol (catvārīṃśad ādīnavāḥ madyadoṣāḥ; Kudo 2004, p. 70). The Tibetan follows MS[B].

n.94According to the Sanskrit akuśalapakṣeṇa (Lévi 1932, p. 44; Kudo 2004, pp. 70 and 71: MS[A] akuśalapakṣena; MS[B] akuśalapathe). The Tibetan reads dge ba’i phyogs su, which corresponds to kuśalapakṣeṇa (or -pakṣe?), instead of akuśalapakṣeṇa in the Sanskrit. This refers to the section below which starts with the sentence, “Regarding the loss of mindfulness induced by drinking beer made from fermented barley and other intoxicating liquors…”

n.95According to the Tibetan, which omits the Sanskrit akriyādṛṣṭiḥ (Lévi 1932, p. 44; Kudo 2004, pp. 70 and 71: MS[B] furthermore adds asatkriyāvādaḥ).

n.96According to the Sanskrit matsarivādaḥ (“one who speaks with hostility”). The Tibetan reads ser sna che ba, which one might literally translate as “(having) strong jealousy” or “great with regard to jealousy,” although that does not fit the context of the hell realms here. Edgerton’s definition in BSHD, s.v. “matsarin,” does not seem to apply here: “(Skt. id., Pali maccharin), in deśanā-°riṇaś ca Mv i.90.3, of backsliding would-be Bodhisattvas, prob. resentful of religious instruction.”

n.97This well-established “Buddhist Hybrid English” term translates Tibetan (dud ’gro’i) skye gnas su, which renders Sanskrit (tiryag-)yoni, “mode/realm of existence” (German Daseinsform; see pw, s.v. “yoni”). As an alternative, one may perhaps translate this as “rebirth as an animal” or “rebirth in the animal kingdom.”

n.98Sanskrit akalpika­pradānam, Tibetan tshul dang mi ’dra (the translation equivalent of the antonym tshul dang ’dra ba according to Mvy (Sakaki 7073) is Sanskrit sārūpya): a gift that would not be appropriate with respect to the status and/or vows of its recipient.

n.99According to the Sanskrit avahasana, which means to make fun of someone by either laughing about/ridiculing or mocking a person’s ailment, condition, or disability. The Tibetan here reads phyas byed pa “to reproach, blame.”

n.100According to the Sanskrit kukkurravratika (Kudo 2004, p. 73). The Tibetan, strangely, reads bya’i brtul zhugs can (“observing a bird vow”), which is most probably a mistake in the manuscript or due to a misreading on the part of the translators, who read kukkuṭa (“cock”) instead of kukkura (“dog”) in the Sanskrit original. The “ox ascetics” and “dog ascetics” are well known from the Pāli Canon (see for example the Kukkuravatikasutta of the Majjhima Nikāya, no. 57; Bhikkhu Ñāṇamoli and Bhikkhu Bodhi 2009, pp. 493–97).

n.101According to the Tibetan. The Sanskrit reads atra, probably in the sense of “there” or “in that direction” (see Apte, s.v. “atra”; Lévi 1932, p. 44): atropapadyeyam iti. This appears to be a (deliberately?) warped understanding of these penance practices, the goal of which, according to the Kukkurravatikasutta (MN 57; Bhikkhu Ñāṇamoli and Bhikkhu Bodhi 2009, pp. 493–97), seems to have been rebirth in heaven.

n.102According to the Sanskrit, which reads avadānaṃ. The Tibetan has ched du brjod instead (= Sanskrit uddāna; see BHSD, s.v.). All available Sanskrit editions read avadānaṃ, Tibetan rtogs par brjod pa (see Lévi 1932, p. 44; Kudo 2004, pp. 72 and 73). The avadāna s are illustrative moral stories of the former rebirths of the Bodhisattva Siddhārtha Gautama before becoming a buddha, in addition to the stories of other bodhisattvas.

n.103According to the Sanskrit. The translation of the following story follows the Tibetan. According to Kudo’s and Lévi’s notes, this story is not found in any of the known jātaka or avadāna collections (see Lèvi 1932, p. 44, n. 13; Kudo 2004, p. 247, n. 17). However, some version of it seems to be contained in Buddhaghosa’s commentary on the Gopakamoggallānasutta (MN 108) of the Majjhima Nikāya, Majjhimanikāyāṭṭhakathā (Majjhimanikāya-atthakathā II, 854; see also DPPN, s.v. “Vassakāra”). Here, the minister Vassakāra sees the elder Mahākassapa walking down the slope of Vulture Peak, when he makes the derogatory remark that the elder Mahākassapa looked like a monkey climbing down the hill. The Buddha reprimands him for having said this and adds that he may be reborn as a monkey if he does not apologize to the elder.

n.104According to the Tibetan, which leaves out ṛṣigiriparśvaṃ parvvataṃ (MS[A], MS[B]: girmakaparvataṃ) but otherwise corresponds almost literally to the Sanskrit (see Kudo 2004, pp. 72 and 73, respectively): “The brahmin Varṣākāra saw the elder Mahākāśyapa in the sky above the city of Rājagṛha, flying from Vulture Peak to the Mount of the Seers.”

n.105The Sanskrit literally says, “he committed a wrongdoing with his speech” (vāgduścaritaṃ kṛtam).

n.106According to the Tibetan. The Sanskrit reads, “When Varṣākāra learned that the Bhagavān had said that during five hundred lives he would certainly be reborn as a monkey, he became upset and developed faith [in the Buddha]” (tatas tena Varṣākāreṇa śrutaṃ Bhagavānā nirdiṣṭas tvaṃ kila pañca janmāntaraśatāni markaṭo bhaviṣyatīti, sa saṃvignacittam prasāditavān; Sanskrit text according to Lévi’s emendation and MS[A]: Lévi 1932, p. 45; Kudo 2004, pp. 72 and 74).

n.107This specification is unclear. No known story of the Buddha’s parinirvāṇa contains this episode.

n.108According to the Tibetan, which reads “where” (gang du). The Sanskrit reads “when” (kadā) in all editions.

n.109The Sanskrit text of MS[B] adds “karmic ripening (of this deed)” (vipākaḥ; Kudo 2004, p. 75). Degé reads bcom ldan ’das sug las de gang du bas par ’gyur zhes zhus pa dang. The compound sug las (literally “manual work/labor,” “toil,” etc., in the sense of making or carrying out something with one’s own hands) does not seem to fit this context. We have translated it here as “deed” in accordance with its context (see the term karmabhūmi, Negi, s.v. “sug las kyi dog sa”). For bas par ’gyur (= Sanskrit kṣīyate) as an old Tibetan expression for zad pa, see Martin 2003, s.v. “bas pa.”

n.110According to the Sanskrit of MS[B]. The Tibetan interpretation of this passage is not entirely clear. This sentence is possibly corrupt. See Lévi 1932, p. 45, n. 16. The Sanskrit reads bhagavān āha. tāny eva pañca janmāntaraśatāni kiṃ tu rājagṛhe utpatsyase yathā [MS[B]: yayā] jambvā jambudvīpe jāyate [MS[A], MS[B]: jñāyate] yatroṣṭrikāmātrāṇi phalāni, yathā kṣaudramadhv aneḍakam evamāsvādāni. tatropapattir bhaviṣyati (Lévi 1932, p. 45; Kudo 2004, pp. 74 and 75). That the Rose-Apple Continent (i.e., the known human world or, more specifically, the Indian subcontinent) derives its name from the rose-apple tree(s) growing there is a well-known trope. The Tibetan here seems to say, “The Bhagavān said, ‘Through/after these five hundred lives [your karma] will be exhausted. For a short time, however, when [the jambu–?] tree in Jambudvīpa carries approximately five hundred fruits called jambū, which are sweet like honey and irresistibly tasty, you will be reborn there.” This Tibetan interpretation perhaps followed a Sanskrit version that read yadā instead of yathā, and jāyate instead of jñāyate.

n.111According to the Sanskrit. The Tibetan reads, “Then, having been reborn in Rājagṛha and died there, you will be liberated and reborn in heaven.” Lévi’s translation (1932, p. 120) of the whole section reads, “During five hundred births, you will be born in Rājagṛha; as the jambū is born only in the Jambudvīpa, where the fruit is the size of an uṣṭrikā pot, and its flavor is like that of clarified bee honey, this is where you will be born. Then you will part from there and have a good rebirth [in heaven]. (This is how the impiety of thought arises among lions.) And it is on this subject that the Most Holy pronounced this stanza.” (Pendant cinq cents naissances, tu naîtras a Rājagṛha; comme le jambū ne naît que dans le Jambudvīpa; là où les fruits ont la taille d’une cruche uṣṭrikā, et que leur saveur est comme celle du miel d’abeilles clarifié, c’est là que tu naîtras. Ensuite tu te relèveras et tu auras une bonne Destination. [C’est ainsi que l’impiété de la pensée fait naître parmi les lions.] Et c’est à ce sujet que le Très Saint a prononce cette stance.)

n.112This is our tentative interpretation of the Tibetan. It is unclear what exactly this sentence refers to, but it may allude to the Siṃhajātaka mentioned above, in which the Bodhisattva is born as a lion (Haribhaṭṭa’s Jātakamālā contains a Siṃhajātaka , but it does not contain this verse). The Sanskrit editions of the Karmavibhaṅga print this sentence as belonging to the story of Varṣākāra’s rebirth as a monkey and thus to the Bhagavān’s speech. However, we fail to see the logical connection between this (or the following stanza) and the foregoing narrative. Furthermore, in the Tibetan the sentence starts with dper na (“for example”), which is regularly employed in the Karmavibhaṅga to introduce a scriptural quotation, an illustrative story, or a simile, and the Tibetan sentence is clearly signaled to end before dper na with zhes gsungs pa lta bu (“like that it was/is said”). Even Lévi’s translation is unfortunately not very helpful here: “C’est ainsi que l’impiété de la pensée fait naître parmi les lions” (Lévi 1932, p. 120; see also n.­111).

n.113The Sanskrit of this verse (Lévi 1932, p. 46; Kudo 2004, pp. 74 and 75) is identical with a verse from the Udānavarga: dīrghā jāgarato rātrir dīrghaṃ śrāntasya yojanam / dīrgho bālasya saṃsāraḥ saddharmam avijānataḥ (Uv I.19 = Patna-Dharmapada 185 = Dhammapada 60). The Tibetan translated here, however, differs from the Tibetan Udānavarga (ched du brjod pa’i tshoms) translation, Chapters of Utterances on Specific Topics : /mel tse byed la mtshan mo ring / /lam gyis dub la rgyang grags ring / /dam chos rnam par mi shes pa’i/ /byis pa rnams la ’khor ba ring /

n.114According to the Sanskrit. The Tibetan reading gshin rje’i ’jig rten is confirmed by MS[A], which reads yamalokaḥ. Lévi (1932, p. 46) has adopted the reading of MS[A]. The Sanskrit of MS[B], however, reads pretaloka, which seems to be an alternative term for yamaloka (see also Tom Tillemans, trans., Questions Regarding Death and Transmigration, Toh 308 [84000: Translating the Words of the Buddha, 2019], g.­39). The inhabitants of the realm of the dead are also often designated “ghosts”; however, this particular translation is possibly more specific to a Chinese Buddhist cultural context. We have here opted for the translation “ghosts” because we see a cultural parallel in the Western concept of ghosts, but readers should keep in mind that this is the translator’s interpretation and that there are also important differences between Western and Buddhist notions of supernatural beings.

n.115Tibetan ngan du spyad seems to be a nonstandard rendition of Sanskrit duścarita. Mvy (Sakaki 1682) lists Tibetan nyes pa spyad pa as the translation equivalent.

n.116English translation adopted from Edgerton; see BHSD, s.v. “mithyājīva.”

n.117The title of this unidentified work is according to the Sanskrit versions. The Tibetan reads brgya bsdus pa’i mdo (*Śatavargasūtra?). Lévi (1932, p. 97, n. 9) identified a partial correspondence of this citation in a passage of the Pāli Mahākammavibhaṅgasutta (MN III, 214–15). See also Kudo 2004, p. 248, n. 20.

n.118Tentative translation. Presumably, the reader is supposed to supply “…leads to rebirth in the realm of ghosts.” The quotation nevertheless remains cryptic, and its connection to the theme of the paragraph is unclear.

n.119MS[B] (Kudo 2004, p. 77) reads apamāna (“contempt, disregard”; German Verachtung, Gerinschätzung; see pw, s.v.) and furthermore adds adhimāna (“conceit”; the Tibetan translation equivalent would be lhag pa’i nga rgyal but is omitted in the Tibetan), which is defined as lying about one’s spiritual accomplishments and constitutes a very grave offence according to the Vinaya.

n.120Tibetan nga’o zhes pa’i nga rgyal literally means “the conceit ‘I [am].’ ” Mvy (Sakaki 1949) lists the variant nga’o snyam pa’i nga rgyal as standard translation equivalent for the Sanskrit asmimāna. This is defined as the pride of identifying with the five skandhas (psycho-physical aggregates that constitute the empirical person) and regarding them as “self” and “mine” based on false views.

n.121According to the Tibetan chung ba’i nga rgyal. The Sanskrit translation equivalent would be ūnamāna (see Mvy [Sakaki 1951], where the prescribed translation equivalent for Sanskrit ūnamāna is cung zad snyam pa’i nga rgyal. Ūnamāna, according to Vasubandhu (with Sthiramati’s commentary; see Jacobi 1932, pp. 33–4, commentary on Triṃśikā, v. 12a), conveys the following meaning: when encountering a person who far outshines one’s own moral and other qualities, one downplays the difference by saying that one is only slightly inferior to that person. The Sanskrit in all editions instead has mithyāmāna, “hypocrisy” (Lévi 1932, p. 47; Kudo 2004, pp. 76 and 77). Furthermore, we have here followed the Tibetan text’s listing of four different kinds of pride instead of the Sanskrit version’s five. MS[A]: mānaḥ, abhimānaḥ, adhimānaḥ, asmimānaḥ, mithyāmānaḥ; MS[B]: mānaḥ, apamānaḥ (or alpamānaḥ? see Lévi 1932, p. 47, n. 1), adhimānaḥ, asmiṃmānaḥ [sic], mithyāmānaḥ (Kudo 2004, pp. 76 and 77, respectively). For definitions of the different kinds of pride enumerated in Abhidharma literature, see Abhidh-k-bh ad V.10a = Abhidh-k-bh(P), 284,23–285,18): sapta mānāḥ māno ’timāno mānātimāno ’smimāno ’bhimāna ūnamāno mithyāmānaśca, etc. See also Sthiramati’s commentary on Vasubandhu’s Triṃśikā, v. 12a (Jacobi 1932), which mentions mānaḥ, atimānaḥ, asmimānaḥ, abhimānaḥ, ūnamānaḥ, mithyāmānaḥ.

n.122According to the Sanskrit. The Tibetan leaves out asuralokopapatti[-pariṇāmitaṃ] (Lévi 1932, p. 47; Kudo 2004, pp. 76 and 77) and just reads der “[transfer to] there.”

n.123The Tibetan literally means “taking what was not offered or given.”

n.124Tibetan here translates very literally: gzhan gyi nor la chags sems su byed pa (“coveting others’ possessions”). The standardized translation according to Mvy (Sakaki 1696) is brnab sems.

n.125According to the Tibetan. The Sanskrit text instead positively states the cultivation of the ten virtuous actions as the cause for rebirth as a human and, unlike the Tibetan, qualifies the extent to which the ten virtuous actions have been cultivated, namely, only weakly or halfheartedly (mandabhāvita) (see Lévi 1932, p. 47; Kudo 2004, pp. 76 and 77; MS[B] in fact reads abhāvita, which is explained by Kudo as “scribal error for subhāvita”). The Sanskrit does not list the ten virtuous courses of action but merely states, “the threefold bodily actions, the fourfold vocal actions, and the threefold mental actions.”

n.126The Sanskrit translation equivalent for the Tibetan ’dod pa’i khams is kāmadhātu (see Mvy [Sakaki 3072]). The Sanskrit manuscripts, however, have kāmāvacara instead (see Kudo 2004, pp. 78 and 79). The Tibetan ’dod pa’i khams na spyod pa’i lha would translate to Sanskrit as kāmadhātvāvacaradeva, which literally means “a deva whose sphere/range of activity/action is the realm of desire (kāmadhātu).”

n.127An alternative translation of Tibetan nam mkha’ mtha’ yas so snyam nas may be “experiencing that space is infinite,” since conceptualization, and thus thought, is said to be suspended in these meditative states.

n.128The Sanskrit editions only list the names of the four formless absorptions of the formless realm without their standard descriptions, which the Tibetan translation provides (see Lévi 1932, p. 47; Kudo 2004, pp. 78 and 79). Mvy (Sakaki 1492–95) gives the following as Sanskrit equivalents of the module describing the formless attainments: sa sarvaśo rūpasaṃjñānāṃ samatikramāt pratighasaṃjñānām astaṃgamān nānātvasaṃjñānām amanasikārād anantam ākāśam ity ākāśānantyāyatanam upasaṃpadya viharati. For an alternative rendering/interpretation, especially of the first of the four formless absorptions, see Karen Liljenberg, trans., The Question of Maitreya (1), Toh 85 (84000: Translating the Words of the Buddha, 2016), 1.­20.

n.129This sentence corresponds to the Sanskrit sa sarvaśa ākāśānantyāyatanaṃ samatikramyānantaṃ vijñānam iti vijñānānantyāyatanam upasaṃpadya viharati (Mvy [Sakaki 1493]).

n.130This sentence corresponds to the Sanskrit sa sarvaśo vijñānānantyāyatanaṃ samatikramya nāsti kiṃcid ity ākiṃcanyāyatanam upasaṃpadya viharati (Mvy [Sakaki 1494]).

n.131This sentence corresponds to the Sanskrit sa sarvaśo ākiṃcanyāyatanaṃ samatikramya naivasaṃjñānāsaṃjñāyatanam upasaṃpadya viharati (Mvy [Sakaki 1495]).

n.132An alternative translation for “accumulated” (Tibetan gsags pa, Sanskrit upacita) would be “increased,” “augmented,” or “intensified” (see Lévi [1932, p. 121], who translates as “est aggravé”). The abhidharmic differentiation between “action that is carried out” (karma kṛtaṃ) and “action that is accumulated” (karmopacita) is explained in detail in Vasubandhu’s Abhidharmakośabhāṣya, verse IV.120 (see Abhidh-k-bh at AKK IV.120 = Abhidh-k-bh(P) 271,20–272,3; English translation La Vallée Poussin/Pruden 1988–90, pp. 701–2): “Action ‘done’ (kṛta) is distinguished from ‘accumulated’ (upacita) action. What are the characteristics and conditions of accumulated action? 120. Action is termed ‘accumulated’ by reason of its intentional character, by its completion, by the absence of regret and opposition, by its accompaniments, and by its retribution.”

n.133According to the Sanskrit. “Action” (karma; las) is missing from the Tibetan text.

n.134According to the Tibetan ’gyod pa, the Sanskrit equivalent of which is, according to Mvy (Sakaki 1980), kaukṛtya. The orthography and derivation of the Sanskrit verb form āstīryati (emendation by Lévi 1932, p. 47, n. 8; MS[A] ārttīyati [Kudo 2004, p. 80]) found in the extant Sanskrit is problematic: see also the Pāli addiyati/aṭṭiyati (“to be worried, to feel loathing”; see CPD, s.vv. “addiyati,” “aṭṭiyati,” respectively); BHSD, s.v. “āstīryati,” gives only this occurrence in the Karmavibhaṅga as reference.

n.135A slightly variant string of verbs is found in the Sanskrit: āstīryati jihreti vigarhati vijugupsati deśayati ācaṣṭe vyaktīkaroti, “one is distressed, ashamed, offended, and disgusted by acts that one has done, and one confesses, tells, and lays it open” (Lévi 1932, pp. 47–48; Kudo 2004, pp. 80 and 81).

n.136This translation follows the Sanskrit more closely than the Tibetan: yat karma kāyena paripūrayitavyam, tatra praduṣṭacitto vācam bhāṣate evaṃ te kariṣyāmīti, idaṃ karmopacitaṃ na kṛtaṃ (Lévi 1932, p. 47; for variant readings see Kudo 2004, pp. 80 and 81), with the exception of the part “but then does not actually follow through,” which is not expressed in the Sanskrit. The Tibetan differs slightly and is partly unclear with regard to the verb tenses and moods: “An action that is complete(d) with the body and with the mind, and when one even says, ‘I will carry out this action!’ [but] then does not carry it out‍—this kind of action is accumulated but not carried out” (de la bsags la ma byas pa’i las yod de de gang zhe na/ las gang lus kyis yongs su rdzogs par byas pa (= paripūritaṃ?) de/ sems kyis yongs su rdzogs par byas (= cittena paripūritaṃ?) shing tshig tu yang las ’di bya’o zhes smras la de ma byas pa ste/ las de lta bu ni bsags la ma byas pa’o).

n.137According to the Tibetan (which differs from the Sanskrit), except for the first sentence, the English rendering of which follows the Sanskrit syntax: “What kind of action is both done and accumulated?” (tatra katamat karma kṛtaṃ copacitañca). The answer in the Sanskrit is simply “intentional action” (yat karma sāṃcetanikaṃ; Kudo 2004, p. 80, paragraph 25). The whole paragraph in the Sanskrit runs thus: tatra katamat karma kṛtaṃ copacitaṃ ca. ucyate. yat karma sāṃcetanikam (Lévi 1932, p. 48). Then follows a citation of two verses from the Udānavarga (31.23, 24)/Dhammapada (I.1, 2). This quotation is missing in the Tibetan, but see 1.­93 below for a variant of this quotation. The Tibetan reads, “What kind of accumulated action is [considered] both carried out and accumulated when carried out?”

n.138According to the Sanskrit. This paragraph is problematic, as the Sanskrit (see Lévi 1932, p. 48) differs from the Tibetan, and the two Sanskrit recensions edited by Kudo also differ between each other (see Kudo 2004, pp. 82 and 83): yat karma saṃcetanīyaṃ svapnāntar{a} kṛtaṃ kāritaṃ vā (MS[A]); yat karmaṃ sāṃcetanikaṃ na kṛtaṃ na kāritaṃ vā (MS[B]). The Tibetan is not very clear but seems to mean “Now, what kind of nonaccumulated action exists that is done but not accumulated when done? Intentional action that is not accumulated (read D: bsags pa), such as, for instance, action done or caused to be done in a dream.” Other editions (Y, J, K, N, and C) read ma bsams par (“unintentional”) for D ma bsags par.

n.139Lévi (1932, p. 122) translates as “during sleep” (pendant le sommeil). Although it is possible to translate the Sanskrit svapna (Tibetan rmi lam) as sleep, the intentionality mentioned in all editions does not seem to call for this translation. According to Vasubandhu’s Abhidharma­kośabhāṣya (Abhidh-k-bh ad IV,120), the presence of intention in an action defines the action as accumulated. Different Buddhist schools of thought, however, come to different conclusions regarding the question of whether karma can be created during dreams.

n.140The Tibetan literally reads, “What kind of action leads, for the person who possesses it, to passing away from the hells after having been reborn as a hell being but only after having completely exhausted the lifespan of the hell realm?”

n.141The Degé reading of Devadatta’s name is lhas byin. Y, J, K, N, and C read the variant lhas sbyin. For Kokālika, see DPPN, s.v. “kokālika.” For Kokālika and his relationship with Devadatta, see the Kokālikasutta, SN III.10, especially vv. 657–78 (Norman 2001, pp. 85–87); AN V, 171–74 and II, 3; and SN I, 149ff. The name Kokālika is here translated into Tibetan as dus min (*kukālika?), which is not attested in any dictionary. Kokālika, in the transliterated form ko kā li ka, also appears in The Limits of Life (Āyuḥparyanta­sūtra, Toh 307), 1.­70.

n.142According to the Tibetan. The Sanskrit of this paragraph differs: “The person, having committed actions that will lead to rebirth in the hell realms, does not regret, etc., the actions but also does not overly rejoice in what they have done, resulting in the lifespan of the hell realm being cut in half (see Lévi 1932, p. 49). It is also mentioned in the Sanskrit that the actions are accumulated; this is omitted in the Tibetan, which is consistent with the definition of this category of karmic action in the Abhidharma­kośabhāṣya.

n.143On the Avīci hell, see Guenther 1986, pp. 58–59: “The torture of the denizens of the mNar.med (Avīci) hell consists in being thrust into huge iron kettles filled with molten […] bronze or copper from vast cauldrons and then being boiled over an unbearably hot fire. As has been written: Some are cast into iron kettles, Head down like the ingredients of rice soup. Because of the uninterrupted pain this hell is called mNar.med (Avīci).” See also Khenpo Könchok Gyaltsen 1989, p. 98, and The Limits of Life (Āyuḥparyanta­sūtra, Toh 307), 1.­80–1.­84.

n.144This is the name in the Sanskrit. In fact, the elephant’s name was originally Nālāgiri, according to the sources (the story is found in the Pāli Vinaya). Only after the Buddha had tamed the elephant, and those who had witnessed the event had heaped all their ornaments on him in reverence, did his name become Dhanapāla (“Protector of Wealth”).

n.145The Sanskrit editions here add “acting upon Devadatta’s instruction/advice” (MS[A] and MS[B] read Devadattasyādeśena; Kudo 2004, pp. 86 and 87, respectively), which is how we generally know the story from Pāli sources. Also, earlier the Sanskrit text mentions Devadatta, who is not mentioned at all in the Tibetan translation: tena Devadatasahayena ānanta(r)yakarma kṛtam (MS[A], Kudo 2004, p. 86).

n.146Tibetan mi dga’ ba. This translation is according to the Sanskrit, which reads saṃvigna (“alarmed, terrified”), which in Buddhist texts describes a state of having an acute awareness of the shortcomings of saṃsāra and consequently developing a strong desire for spiritual liberation. See also pw, s.v. “saṃvega”: German Verlangen nach Befreiung (“a longing for liberation”). Appleton (2014, p. 20) translates as “experienced profound shock.” The common sense of the Tibetan mi dga’ ba, literally “unhappy,” does not quite seem to capture this sense.

n.147According to Kudo (2004, p. 250, n. 22), this version of the Ajātaśatru episode of the Śrāmaṇyaphala­sūtra does not match any of the known versions in Pāli, Sanskrit, or Chinese.

n.148According to the Sanskrit (pratisaṃdadhāti kuśalamulāni; see Lévi 1932, p. 49). The Tibetan reads “accumulated the roots of virtue” (dge ba’i rtsa ba bsags pa).

n.149This is a free translation of the seemingly idiomatic expression (Sanskrit) asthibhir api (“even with my bones”). Lévi (1932, p. 122) translates this literally, according to the Sanskrit: “Même avec mes ossements.” The Tibetan translation reads rus pa yan cad kyang, literally “down to the bone.” The Sanskrit manuscripts read asthibhir iti (MS[A])/aṣṭhibhir api (MS[B]) Bhagavantaṃ śaraṇaṃ gacchāmi (Kudo 2004, pp. 86 and 87, respectively). Lévi’s copy of manuscript A apparently reads aṣṭabhir (see Lévi 1932, p. 49, n. 11): “Eight times [I go for refuge to the Bhagavān].” The reading asthibhir, however, is confirmed by the Tibetan translation.

n.150The last two sentences (from “I have carried out intolerable actions” to “he went silent”) are not in the Sanskrit. The Sanskrit reads (after the refuge formula, buddhaṃ bhagavantaṃ śaraṇaṃ gacchāmi): The very moment he was reborn [in hell] he passed away (sa upapannamātra eva cyavati; Lévi 1932, p. 50).

n.151According to the Tibetan. The Sanskrit reads “there” instead of “such a one” (Sanskrit amutra … tatra).

n.152This is the spelling in the Sanskrit manuscripts. For the Śyāmākajātaka (= dkar sham kyi skyes pa’i rabs), see Kudo 2004, pp. 250–51, n. 23.

n.153Sanskrit differs slightly: yathā Bhagavāno jātake Śyāmākajātakaprabhṛtiṣu praṇidhānavaśād upapattir varṇyate, “For example, in a story of a former birth of the Bhagavān, e.g., in the Śyāmākajātaka , etc., rebirth due to a strong aspiration is related” (Lévi 1932, p. 50).

n.154According to the Tibetan. The Sanskrit reads, “A person does not dedicate his action to a rebirth in a certain place, saying, ‘May I be reborn there!’ ”

n.155Tibetan rnam par smin pa. Sanskrit (all editions) reads vipakṣa for vipāka. Vipakṣa is a hyper-Sanskritism of the Middle Indic vipakka (see BHSD, s.v. “vipakṣa”). Lévi (1932, p. 50) emended the Buddhist Hybrid Sanskrit reading to vipāka on the basis of the Tibetan.

n.156According to the Tibetan, which reads (literally) “In this very life or in another life” (de ni tshe ’di nyid dam tshe gzhan la). The Sanskrit, according to the manuscripts, reads yat karma tasminn eva janmāntare deśāntaragatasya vipacyate śubham aśubhaṃ vā (Kudo 2004, pp. 88 and 89): “Karma that ripens as good or bad karmic result in this or another life for someone who has gone abroad (literally ‘to another country’).” Lévi (1932, p. 50) emended the text following the Tibetan translation and added vā after tasminn eva janmāntare. However, since vā is not contained in either MS[A] or MS[B] according to Kudo’s edition, and unless both manuscripts are faulty, tasminn eva janmāntare could also mean “within this very life.” This interpretation seems consistent with 1.­95 below, which explicitly says that some actions can ripen in “in this very lifetime.” The Tibetan translation, however, suggests that -antare has the sense of Tibetan gzhan (“another”) here, and that the Sanskrit text as it is needs to be emended.

n.157Literally “a merchant, captain, caravan leader” (Tibetan ded dpon, Sanskrit sārthavāha).

n.158According to Kudo, this is the form of the name used throughout in MS[B] (Kudo 2004, p. 92, n. 5).

n.159According to the Tibetan (bza’ shing gi ra ba). The Sanskrit reads “park” (udyānaṃ).

n.160Sanskrit adds tasya pitaram pūrvaṃgamaṃ kṛtvā, “having made your father their leader” (Lévi 1932, p. 51).

n.161Sanskrit MS[A] reads samudrabhūmi; MS[B] suvarṇṇabhūmi (Kudo 2004, pp. 88 and 89, respectively). Lévi (1932, p. 123) translates literally as “Land of Gold” (Terre de l’Ore). See also the glossary entry on this toponym.

n.162According to the Tibetan. The Sanskrit reads, “to acquire merchandise” (dravyopārjanaṃ; Lévi 1932, p. 51).

n.163According to the Tibetan. The Sanskrit reads, “I/we will beg her/make her understand” (tam arthaṃ vijñāpayāmi (MS[A])/vijñāpayāmaḥ (MS[B]); Kudo 2004, pp. 90 and 91, respectively).

n.164According to the Tibetan (rkang pa nas bzung). Sanskrit (Lévi 1932, p. 51 passim) reads pādapatanaṃ kṛtvā (“she threw herself at his feet”).

n.165According to the Tibetan, which literally says “It must not be that we do not go for certain!” The Sanskrit (Kudo 2004, p. 90, MS[A]) reads “We must go now!” (atha gantavyaṃ iti).

n.166According to the Tibetan. The Sanskrit similarly reads “on the thirteenth day” (trayodaśyāṃ, from trayodaśī, f. “the thirteenth day of a half moon”; see MW). However, neither the Tibetan nor the Sanskrit specifies whether the thirteenth day of the first (śuklapakṣa) or the second half (kṛṣṇapakṣa) of the month is intended. Lévi (1932, p. 125) translated “the thirteen of us” (Nous, les treize, nous partirons).

n.167According to the Tibetan. The Sanskrit (Kudo 2004, p. 90, MS[A]) reads mātuḥ pādaṃ mastake datvā prakrāntaḥ (“hitting her on the head with his foot, he set out”). The Sanskrit datvā should probably be emended to hatvā. See Klaus 1983, p. 50, v. 51b).

n.168Only MS[A] preserves this passage (see Kudo 2004, p. 92): Maitroyajñas carmapraṭamukhyān avabaddhaṃ tāmaraghaṭañ ca gṛhya samudrakṛta utīrṇṇaḥ. Lévi’s emended edition (Lévi 1932, p. 52) reads Maitrāyajñaś ca ma. pra. ṭamukhyān avabaddhaṃ tāmraghataṃ ca gṛhya samudrakūla uttīrṇaḥ. Lévi translates with the help of his edition of the Tibetan (according to N: byams pa mchod sbyin ni zangs kyi ril [D ral] chen kha bskya [D skya] ’dar gyis bcad pa zhig la [D |] ’ju ’ju nas rgya mtsho mthar phyin te): “Maitrāyajña, lui, s’accrochant à un grand vase de cuivre qui avait le col fermé par une étoffe, put atteindre le rivage” (see also Lévi 1932, p. 52, n. 3). The N reading seems to be closer to the extant Sanskrit and makes most sense.

n.169According to the Tibetan. Mvy (Sakaki 6058) has avasaktapaṭadāmakalāpaḥ = dar gyi lda ldi mang po btags pa; Kudo 2004, p. 90 (MS[A]): avasaktapaṭṭadāmakālāpaṃ; D: dar gyi lda ldi mang po bres pa. The Sanskrit seems to mean “suspended bundles/tassels or ornaments (kalāpa) made from silk ribbons.”

n.170Tibetan: bsgo ba (“to say, when used of superiors, hence mostly to bid, to order.” See Jäschke 1972, s.v. “sgo ba”). The Sanskrit simply says uktāḥ (“they said [to him]”).

n.171The Sanskrit āryaputra can, besides the more literal translation “son of a reputable family,” also mean “husband” when a woman is addressing her man (see pw, s.v. “āryaputra”).

n.172Tibetan is only partially successful in emulating the Tibetan diction (’phags pa’i bu ’di ni sngan cad khyod la med pa’i yul yin). The Sanskrit (Kudo, 2004, p. 92, MS[A]) reads āryaputra, tavāyaṃ pṛth(i)vīpradeśa apūrvam asmākam aviditaṃ na nirgantavyaṃ | yadi nirgacchasi sarvathā uttarābhimukho na{r}gantavyam iti (“Husband, this spot of the earth is new to you. You should not leave here without our knowledge. However, if you [have to] go away, do not go to the north”). The Sanskrit reads throughout (Kudo 2004, pp. 94 and 96; Lévi 1932, p. 52–53) uttarābhimukhena na gantavyaṃ (“should/must not go north”).

n.173The Sanskrit here repeats “for many years, many hundreds of years, many thousands of years, and many hundreds of thousands of years.”

n.174This is the common translation equivalent for the Sanskrit vaiḍūrya, Tibetan bai dū rya. The Sanskrit vaiḍūrya, however, was suggested by Alfred Master (1944) to be a Sanskritization of a Middle Indic form related to Ardhamāgadhī [ve]ruḷiya, Pāli [ve]ḷuriya, that is to be identified with beryl rather than lapis lazuli (at least not until later in Indian history). The words (English) beryl and (Pāli) veḷuriya are etymologically related.

n.175Tibetan: gnas ’di nas phyi rol tu byung ta re, literally “you should not go outside from this place.”

n.176According to the Tibetan. As above, the Sanskrit reads “you must not go north!”

n.177Mvy (Sakaki 4944): pratyekanarakaḥ (atyekanarakaḥ) (= nyi tse’i sems can dmyal ba) literally means “a denizen of a lesser hell,” i.e., a hell with less severe punishment. The Sanskrit equivalent may rather be prādeśika, and thus “a local/limited/lesser/ephemeral [hell].” However, see also Edgerton (BHSD, s.v. “pratyeka”), who cites the Karmavibhaṅga: “°narakaḥ Karmav 53.13–14; (tasminn eva janmani) pratyekasvargaṃ pra°narakaṃ (n. sg.) cānubhūtam 57.2 (refers to story of which 53.13–14 is part); here private, personal heaven and hell seem to fit, since the one who experiences both seems to be a single individual (at a given time; a former inhabitant of the pra°naraka is released as the new one arrives).”

n.178According to the Tibetan. The Sanskrit adds “he is our ancestor” (asmākaṃ pūrvapuruśo ’sti; Lévi 1932, p. 53).

n.179According to the Sanskrit pūrvaṇgamaṃ kṛtvā (Lévi 1932, p. 53).

n.180Tibetan bu yang khyod gcig pur zad. This expression is recorded in Jäschke’s dictionary: “you are my only son” (Jäschke 1972, s.v. “ ’dzad pa”).

n.181This is a rather free translation of the Tibetan gser gnas su dong dong ba las.

n.182Lévi translates the Sanskrit apsarasaḥ/apsaras (Tibetan lha mo) as “Nymphes célestes” (celestial nymphs).

n.183According to the Tibetan (bar du snga ma bzhin te), which simply leaves out the repeated arrival at the different cities that was related earlier by Maitrāyajña. The Sanskrit manuscripts, however, repeat the whole sequence. See Kudo 2004, p. 100.

n.184Reading C and H de na instead of D de ni.

n.185Or “armored with”? See Lévi’s translation (1932, p. 126).

n.186This spelling of the toponym is confirmed by several premodern Indian lexicographic works (see Patkar 1953, p. 297, s.v. “Tamluk”). Tāmalipta (present-day Tamluk) was an ancient Indian port city on the Hugli (anglicized also as Hooghli) River, a distributary of the Ganges in West Bengal. The famous Chinese pilgrim Faxian is reported to have left India from Tāmalipta on his way home (see Ch’en 1964, p. 91). Other spellings are found in lexicographic and other sources (see pw): tamoliptī, tāmralipta, dāmalipta, tamālikā, tamālinī.

n.187For lta having the sense “indeed, thus then, evidently,” see Jäschke 1972, s.v. “lta.”

n.188The Tibetan here omits Maitrāyajña’s first question about the lifespan in this individual hell: “ ‘What is the lifespan [here]?’ He replied, ‘six thousand years’ ” (kitrāyuḥ; Kudo suggests emending to ki(ṃ) tv āyuḥ or kiadāyuḥ). Lévi (1932, p. 54) also omits it, perhaps based on the Tibetan translation (N) that he consulted (see ibid., p. 195; see also Kudo 2004, p. 104, n. 1).

n.189According to the Tibetan [sha] rnyil ba (“broken down” = archaic Tibetan for ’gyel ba?). Lévi records sha ril ba (“the meat that is falling down [from my own head]”?) as a variant reading in his edition of N (Lévi 1932, p. 195). The flesh, however, is not mentioned at all in the Sanskrit, which reads ata eva mastakāc chidyamānād yat pūyaśoṇitaṃ sravati (“this here, the pus and blood that flows from [my] cut head”; Kudo 2004, p. 104, MS[A]). See also this description in the Maitrakanyakāvadāna, at 111+ (Klaus 1983, p. 81): svaśiraḥ-pravigalita-śoṇita-vasā-rasāhāra-mātra-vidhṛta-prāṇa-śeṣaṃ (“[Maitrakanyaka saw that hell being] for whom a [meagre] rest of the life force was preserved only by the food of blood and liquor oozing from his own head”).

n.190According to the Tibetan. The Sanskrit differs: “With his whole being he prostrated to his parents and said…” (sarveṇa bhāvena mātāpitroḥ praṇipātaṃ kṛtvāha; Kudo 2004, p. 104).

n.191According to the Tibetan. The Sanskrit differs in the details: pada c also mentions “humans” (nara) beside “snakes” (uraga): ūrddhaṃ bhavāgravitatām adharasya (’)vīciṃ | tiryagpradhān agaṇitān api lokadhātūn | ā[tma] «nsva»rāsuraṇaroragabhūtakāye | satvāni yāny upagatāni svakhī bhavantu (Kudo 2004, p. 104, MS[A]).

n.192According to the Sanskrit yuktās. The Degé edition’s orthography seems to reflect the Sanskrit: rigs pa dang ldan (“those who are able/devoted to/engaged in [practice and moral discipline]?”). C reads rig pa dang ldan (“learned”).

n.193According to the Tibetan. This last sentence occurs in the Sanskrit only after the insertion of two more textual references (32.b and 32.c) that are omitted in the Tibetan (see Lévi 1932, p. 55; Kudo 2004, p. 108, MS[A]: 32.c), which furthermore varies significantly from the Sanskrit: sa tatra prītyahāraḥ sthitvā paripūrṇeṣu ṣaṣṭivarṣasahasreṣu kālagataḥ (“He [Maitrayajña] remained there with joy as his nourishment and died after six thousand years” (Kudo 2004, p. 108, n. 17 gives {a}paripūrṇeṣu as possible variant reading of MS[A]; MS[B] missing).

n.194Inserted according to the Sanskrit atīva (“extreme; exceedingly, very, excessively”).

n.195There is a paragraph here in the Sanskrit (32.e; see Lévi 1932, p. 56) that is missing in the Tibetan.

n.196Tibetan rim gro byed pa corresponds to the Sanskrit upasthānam (Mvy [Sakaki 1762]); the Sanskrit here, however, reads gaurava.

n.197According to the Tibetan. The Sanskrit reads slightly differently (Kudo 2004, pp. 112 and 113): yathā mayi tathā mātāpit[ṛ]ṣ[u] ācāryopādh‹y›āyeṣu vacanakāriṇāṃ samo vipāka ihaloke paraloke ca (“as for me, so for your parents, preceptor, and teacher‍—the karmic result for one who obeys orders/those meriting respect is the same, here in this world and in the next”).

n.198The translation of this stanza is based on the Sanskrit. See also AN V.36, Kāladānasutta (III.42). Here in the Tibetan only the first half of the verse is given.

n.199The Tibetan [yid ni] legs par mos [’gyur na] is meant to render the Sanskrit [manasā ca] prasannena, which in most translations of this verse is translated as “pure mind/intent.” This Tibetan rendering is not the standardized or expected translation equivalent of the Sanskrit pra + √sad (see Mvy [Sakaki 7295]: dang ba’am gsal ba; Mvy gives adhimukti [“reverence”] as translation equivalent for mos pa). However, the Tibetan translation has clearly favored the sense of “faith” (or “reverence, devotion”) for prasanna over its second main sense (“pure, clear”), probably in order that the verse better fit the theme of its present context of faith and devotion.

n.200The translation of this stanza tries to follow the Tibetan, although the Tibetan text is partly unclear. Specifically, the Tibetan word order of the first pada is strange: chos rnams sngon du yid ’gro ste (see also Uv 31, 24: chos kyi sngon du yid ’gro ste). For the well-known parallels of these famous verses, see Dhp I.1, 2 (Pāli) and Uv 31, 24. Note that the Tibetan translation here is slightly different from the Sanskrit and from the Tibetan translation of the Uv in the Kangyur.

n.201The Tibetan translators appear to have read Nagaraśikhin (D grong khyer gtsug phud). The Sanskrit editions, however, confirm Tagaraśikhin as the correct form of the name. On Tagarasikhi in the Pāli Canon, see DPPN, s.v.

n.202Tibetan bsod snyoms, Sanskrit piṇḍapata (see Mvy [Sakaki 8671]); Lévi (1932, p. 57) emended to sūpaḥ. MS[A] and MS[B], however, read yūpaḥ and yūṣaḥ, respectively. The latter means “(bone) broth” (see Kudo 2004, pp. 112 and 113; pw, s.v.). The text does not mention to whom the broth was offered.

n.203According to the Sanskrit. The Tibetan omits the Sanskrit tasmin nagare (“in this city”; Lévi 1932, p. 57).

n.204See, however, Lévi 1932, p. 129: “Et de plus ce fut pour lui le germee de son salut.”

n.205According to the Tibetan. The Sanskrit manuscripts give Śikhaṇḍī as the name of the prince.

n.206Tibetan ’o dod can zhes bya ba grong khyer. MS[A] has the (perhaps orthographically more correct) Rorukā, while MS[B] and Lévi read Raurukaṃ (Kudo 2004, pp. 114 and 115; Lévi 1932, p. 129).

n.207Other versions of this story may be found in the Divyāvadāna (no. 38; cf. Klaus 1983) and the Mūlasarvāstivādin Vinaya. We are unable at this point to provide exact references. The Tibetan translation of the Vinaya of the Mūlasarvāstivāda school is currently being translated by 84000.

n.208According to the Tibetan. The Sanskrit reads tena kāraṇena kiṃ nāsti nānākaraṇaṃ ucyate (“Why is there no difference between those causes [i.e., between the Buddha and one’s parents]?” Kudo 2004, p. 117, reading MS[B]). The locution of the Tibetan translation is different but preserves the sense of the Sanskrit.

n.209According to the Tibetan. The Sanskrit reads “the Awakened One shows the path toward awakening; he is the revealer of the path for those for whom no path has yet arisen. Revering him yields immeasurable karmic fruit and, ultimately, awakening” (according to Lévi 1932, p. 58).

n.210According to the Sanskrit. The Tibetan is unclear: “At the time of his parinirvāṇa he set his parents on the path to liberation.”

n.211The Tibetan and Sanskrit literally say “to the son(s)” (bu la; putrān).

n.212The translation of this unclear expression follows Lévi’s French translation: “Allons! Conduis–moi dans un lieu inhabité,” together with the Sanskrit. The Tibetan expression kha ’tsho ba is unclear. Lévi (1932, p. 58, n. 12) reads kha ’tshe ba (Sanskrit avasanam/avaśanam = avasānam?) but does not provide a translation for his edition of the Tibetan.

n.213According to the Tibetan. The Sanskrit adds “There is no realization of the goal” (nāsti phalaprāptiḥ).

n.214Sanskrit differs: “Except for this, one’s parents, one’s teacher, and one’s preceptor are equal” (evaṃvidhaṃ varjayitvā anyathā samasamā mātāpitaraḥ ācāryopadhyāyāḥ; Kudo 2004, p. 117, following MS[B], MS[A] defective; Lévi 1932, p. 59).

n.215This is a free rendering of the idiomatic Tibetan: pha ma bu la byams pa ni rkang dang rus pa’i khong nas byams so. The Sanskrit literally reads “Parents’ love for their son enters the marrow and stays there” (mātāpitroḥ putrasnehaḥ yāvad asthimajjām āśritya tiṣṭhati; Kudo 2004, p. 117, MS[B]; Lévi 1932, p. 59).

n.216Tibetan reads pho brang ’khor skyong. Mvy (Sakaki 1361) gives the later, standardized version of this name as yul ’khor skyong.

n.217According to the Tibetan. The Sanskrit here adds the name Śoṇa (MS[B]) or Sronita (MS[A]) (= Śroṇa[koṭikarṇa]? see Kudo 2004, p. 260, n. 33). For Rāṣṭrapāla, see the Pāli Raṭṭhapālasutta, MN 82 (Bhikkhu Ñāṇamoli and Bhikkhu Bodhi 2009, pp. 677–91); see also DPPN, s.v. “Śona–Kuṭikaṇṇa/–koṭikaṇṇa.”

n.218This translates the D reading: rab tu ma phyung ba. N reads rab tu ma byung ba. The Buddha gives ordination (rab tu ’byin) rather than receives ordination (rab tu ’byung).

n.219According to the Sanskrit: adyāpi (Lévi 1932, p. 59).

n.220Y, K, C, and N read nga chung yang pha mas ma gnang na/bar rab tu mi dbyung ste; D: da dung yang …

n.221According to the reading of D: mig rub par gyur pa lta bu’i pha mas. Y, K, and N read mig rub par gyur pa lta bu’o/ pha ma. The expression mig rub par gyur (= Sanskrit cakṣuṣī antarhite, “the eyes disappeared/shut”) seems uncommon. The only parallel to this story, according to Kudo, is found in the Mahāvastu; there, however, it is the Buddha’s mother who goes blind (see Kudo 2004, 260, n. 34).

n.222According to the Tibetan (rim gror byed par ’gyur ro). The Sanskrit differs slightly: “When he is grown up, he will protect us when we are old” (saṃvardhito no vṛddhībhūtān pālayiṣyati).

n.223J and C omit this sentence.

n.224This almost certainly constitutes a quotation from a Buddhist scripture, but we were unable to identify the source. See Lévi 1932, p. 59; Kudo 2004, pp. 118 and 119 and, for a synopsis of different versions of this list, n. 35). See also the list of essentially the same items in the Pāli Canon at AN III.39 (III, 43–44).

n.225This sentence is problematic in the Tibetan. Should we adopt the variant reading in Y and K: ’khor ba’i mthar phyin par bya instead of D, J, C, and N: da dung pha mas ma gnang na rab tu mi ’byung gi bar du phyin par bya? The first corresponds better with the extant Sanskrit (Lévi 1932, p. 59; Kudo 2004, pp. 118 and 119). For the sake of comparison, we here give the whole paragraph, highlighting in bold the corresponding passages: na evam ācāryopādhyāyāḥ ‹|› kevalam eva kāruṇyaṃ puraskṛtya katham asyānādikālavṛttasya saṃṣārasya paryantaṃ kuryād iti (Kudo 2004, p. 119, MS[B]; Lévi p. 59).

n.226This last sentence of the quotation differs in the Skt.: evam anyonyaniśritāḥ sukhino bhaviṣyanti, “In this way relying/leaning on each other will be conducive to their happiness” (Kudo 2004, p. 119, MS[B]; Lévi p. 59).

n.227As to this text and its variant titles, see Kudo 2004, pp. 262–63, n. 37.

n.228The precious elephant and the precious horse are part of the “seven treasures/jewels of a wheel-turning monarch” (the wheel, precious jewel, queen, minister, elephant, general, and horse).

n.229According to the Tibetan. The Sanskrit differs slightly: MS[A] and MS[B] add dīrgharātraṃ (“for a long time”), and the (future) wheel-turning monarch “carries his parents himself or has them carried” (ācāryopādhyāyān svayaṃ vahati vā vāhayati vā; Kudo 2004, pp. 120 and 121, respectively).

n.230Tibetan mchod pa’i gnas, Sanskrit gṛhasthānāṃ mātāpitṛ pravrajitāḥ pūjyāḥ (Kudo 2004, p. 121, MS[B]).

n.231The Sanskrit contains another reference here that is missing in the Tibetan (see Lévi 1932, p. 60; Kudo 2004, pp. 121 and 122).

n.232This paragraph was translated according to the Sanskrit yaś ca punar […] evaṃ mātrāpitṝṇāṃ putraiḥ pratyupakāraḥ kṛto bhavati (Lévi 1932, 61,2–5). The Tibetan is unclear and does not seem to fit the context: “It is said that if someone makes someone else develop faith in the Buddha, the Dharma, and the Saṅgha, establishes one in the five precepts, and encourages one to seek refuge (skyabs su gtong ba) in the path that is praised by the noble ones, then through merely saluting and greeting them respectfully with one’s palms joined in reverence (thal mo sbyar ba) and providing them with a mat, robes, alms, bedding, and medicine against illness, it is impossible to repay their kindness (phan pa’i lan lon par mi nus so). In this way, the preceptor and the teacher are more distinguished than the parents.”

n.233This sentence occurs later in the Sanskrit version (Lévi 1932, p. 64), after a portion of text in the Sanskrit (Lévi 1932, pp. 61–64) that is not contained in the Tibetan. The missing portions gives examples from Buddhist scriptures in which individuals are said to have appeased different lands, areas, or beings such as Mahendra (or Mahinda in Pāli), who, according to tradition, was sent to convert the island of Sri Lanka to Buddhism.

n.234According to the Sanskrit abhiprasādita (Kudo 2004, p. 127). The Tibetan has mngon par mos pa, which according to Negi renders the Sanskrit abhilāṣa.

n.235This sentence is not preserved in MS[A], but only in MS[B]. See Kudo 2004, p. 126, especially nn. 7, 127, and 226. Lévi (1932, p. 64) gives the reading of MS[B] in the main text of his edition.

n.236For the most part, this passage, starting from “Monks, suppose someone took their parents on their back and roamed the Jambu continent,” follows the structure of the text as it is preserved in the Degé edition, unless indicated otherwise in the notes. Sometimes it was necessary to resort to the corresponding Sanskrit portions when extant, for the Tibetan is oftentimes obscure. Overall, the extant Sanskrit manuscripts preserve a different version and different readings, and the exact original Sanskrit form of the Tibetan text is difficult to reconstruct.

n.237This paragraph is essentially a praise of the Blessed One’s words or Dharma or his teaching. See also the famous quotation in the Vakkalisutta, SN III, 120: yo kho vakkali, dhammaṃ passati so maṃ passati, yo maṃ passati so dhammaṃ passati, dhammaṃ hi vakkali, passanto maṃ passati. Maṃ passanto dhammaṃ passati (“He who sees the Dhamma, Vakkali, sees me; he who sees me sees the Dhamma. Truly seeing the Dhamma, one sees me; seeing me one sees the Dhamma”).

n.238This phrase deviates from earlier paragraphs and from the Sanskrit, which continues as before with the reply (ucyate).

n.239This is his name given in the Sanskrit. The Tibetan reads khye’u grags pa, which, according to Lévi (1932, p. 65, n. 5), may be the result of an erroneous reading in the Sanskrit manuscript written in a Gupta-type script: the Tibetan translators may have read yathāyaśo- for yathāyagopakaḥ (MS[A]; Kudo 2004, pp. 273–74, n. 49).

n.240MS[A]: paścime pi; MS[B]: paścime bhave (“in a later reincarnation”). These add anyatamasmin gṛhe (MS[A]) and anyatarasmin gṛhe (MS[B]) “in a certain home” (see Kudo 2004, pp. 130 and 131, respectively). The Tibetan phyis (“later”) generally corresponds with MS[A] paścime pi but leaves out “in a certain home.”

n.241Sanskrit mūlanakṣatra, Tibetan skar ma snrubs. The nineteenth of the twenty-eight “lunar mansions” (Sanskrit nakṣatra).

n.242According to the Tibetan rigs phung bar byed par ’gyur te (“to kill, destroy”). MS[B] reads ayaṃ mā nirmūlaṃ eva kulaṃ kariṣyati (“lest he should eradicate this family”); MS[A] omits kulaṃ (Kudo 2004, pp. 130 and 131). The Tibetan does not emulate the pun involving the Sanskrit word mūla.

n.243According to the Tibetan. The Sanskrit adds here, “All this the Blessed One told the monks from beginning to end” (etac ca sarvaṃ anupūrveṇa Bhagavānā bhikṣūṇāṃ kathitaṃ; see Lévi 1932, p. 66).

n.244Here the Sanskrit text adds another story, which is missing in the Tibetan (see Kudo 2004, pp. 132 and 133; Lévi 1932, p. 66).

n.245According to the Sanskrit. The Tibetan kha sra ba las is obscure here but seems to correspond to the Sanskrit kṛcchreṇa, “with difficulty” (here rendered as “reluctantly”).

n.246According to the Tibetan. At this point the Sanskrit additionally contains the summary of a story from a former life of Aniruddha, together with the instruction to tell that story in full (Kudo 2004, p. 274, n. 50).

n.247According to the Tibetan. The Sanskrit contains additional references to stories of great donors of the Buddha, beginning with Miṇḍhaka (= Meṇḍhaka?). See Mūlasarvāstivāda Vinaya, Divyāvadāna nos. 9 and 10; for exact references, see Kudo 2004, p. 278, n. 52.

n.248Here the context demands the translation “suffering” for the Tibetan sdug bsngal (Sanskrit duḥkhita) instead of “unhappy,” as in the foregoing paragraphs.

n.249Here the Tibetan translation has tshegs chen po where the Sanskrit reads kṛcchrāt, instead of kha sra ba las as in 1.­106 above.

n.250According to the Tibetan. The Sanskrit additionally contains a story from an otherwise unknown Daridrakārasyāvadāna (see Kudo 2004, pp. 140 and 141 and n. 54).

n.251According to the Tibetan log pa’i ngang tshul can (= duḥśīleṣu?); the Sanskrit adds abrahmacāriṣu (“[and] who conduct themselves unethically/immorally”).

n.252The phrase “in former births” is added here to provide context.

n.253It is not clear what the Tibetan expression gser gyi yag mag [= yag ma (“rug”; “saddlecloth”)?] bting ba exactly means. The Sanskrit reads suvarṇāstīrṇaṃ niryātayiṣyati (“he will spread out gold covering [the entire ground of Prince Jeta’s grove]”). It is clear, though, that the expression refers to the famous story of the gift of Prince Jeta’s grove to the Buddha and his saṅgha by Anāthapiṇḍada as a dwelling for the monks.

n.254According to the Tibetan. The Sanskrit adds Kāśmīrāyāṃ; see Lévi 1932, p. 72; Kudo 2004, pp. 149 and 150.

n.255This refers to the story told at 1.­4 above.

n.256Literally “he departed” (Tibetan slar log pa dang, Sanskrit saṃprasthitaḥ).

n.257The Tibetan mkhas pa (Sanskrit vidvāṃsaḥ) usually means “learned, skilled, experienced; knowing, wise; a scholar,” etc. Here, however, it more likely just means that he knew who the sthavira Maudgalyāyana was.

n.258According to the Tibetan. The Sanskrit has a plural here: taśmiṃś ca gṛhadvāre deśāntarābhyāgatāḥ puruṣā vidvāṃsaḥ pūrvasthitāḥ te taṃ dṛṣṭvā vismayaṃ prāptāḥ, “At that door had been standing some learned people who had come from a foreign country” (Lévi 1932, p. 72).

n.259The reading nyer dga’ bo (“Upananda”) is according to the Degé edition. Y and K: bsnyen dga’ bo; N: nye dga’ bo. Mvy (Sakaki 3278) records the reading nandopanandau nāgarājānau = klu’i rgyal po dga’ dang nye dga’ gnyis.

n.260The Sanskrit additionally has “at night” (rātrau).

n.261According to the Tibetan. The verse is slightly different in the Sanskrit (Lévi 1932, p. 73; Kudo 2004, p. 159), where pada c reads “The world is full of delusion.” A similar verse is cited by Patrul Orgyan Jigme Chökyi Wangpo (1808–1887) in his celebrated kun bzang bla ma’i zhal lung, F.35.a–b (“The Words of My Perfect Teacher,” p. 51), but is said there to have been spoken by Kātyāyana.

n.262According to the Sanskrit yamaloke (Lévi 1932, p. 74; Kudo 2004, pp. 158 and 159). The Tibetan has yi dags (= Sanskrit preta) instead.

n.263According to the Tibetan. The Sanskrit leaves out the asuras.

n.264Here the Sanskrit contains an additional story that illustrates this karma category; missing in the Tibetan.

n.265MS[A] has an additional paragraph here, 43a, that is missing in MS[B] and in the Tibetan translation.

n.266According to the Tibetan. The Sanskrit has pratyekabuddha (Tibetan rang sangs rgyas) instead of Tibetan dgra bcom pa (= arhat ) (see Lévi 1932, p. 75; Kudo 2004, pp. 162 and 163, respectively).

n.267Kudo (2004, p. 293, n. 61) translates the Tibetan bde ba (Sanskrit sukhin) as “blessed.” The Tibetan bde ba usually means “happiness, joy” or “happy,” and the Sanskrit sukha means “happiness,” “well-being.” The kind of person referred to here seems to be someone who is physically healthy and well but still suffers because they are not liberated.

n.268Tibetan rgyal po man ta.

n.269According to the Tibetan. The Sanskrit differs: yathā mahādhana­brāhmaṇagṛhapatayo rājā ca māndhātā (Lévi 1932, p. 75; Kudo 2004, pp. 162 and 163). Lévi (1932, p. 139) translates following the Tibetan. The identity of the individuals referred to here is unclear.

n.270Degé spells the name ka ra ma sha; N reads kar ma sha, according to Lévi 1932, p. 75, n. 3. This person is not mentioned in the Sanskrit. The Sanskrit readings are unclear: MS[B] reads ya(thā)rhann alpapuṇyaḥ, where Alpapuṇya might be a proper name (Kudo 2004, p. 165; Lévi 1932, p. 75, n. 3); MS[A] reads yathātyapuṇyaḥ (“as [someone] with an extreme lack of merit”?).

n.271The Sanskrit manuscripts have the following variants of this person’s name: MS[A] Śroṇottaraḥ; MS[B] Śoṇottaraḥ (Kudo 2004, pp. 164 and 165). The exact identity of this person is unknown. The Sanskrit of Lévi’s edition (Lévi 1932, 75; corresponding with MS[A]) reads yathārhann apuṇyaḥ. MS[B] (Kudo 2004, p. 165) has the variant alpapuṇya cittena sukhī na kāyena (“As, for example, an arhat with little[/no] merit is someone who is joyful in mind but not [well] in body.”

n.272According to the Sanskrit kapikacchumiśraḥ. D reads ka pi ta su ka ra bsregs pa te. This passage is problematic. The Tibetan could also mean “[a ball made of] cow dung and burnt sugar (ka ra bsregs) in Juniper resin (ka pi ta).” The N, Y, and K reading of bsres te instead of bsregs, however, may be confirmed by the Sanskrit miśra (“mixed with”), and ka pi ta su ka ra might be an attempt at rendering Sanskrit kapikacchu phonetically in Tibetan. For kapikacchu (Mucuna pruriens or “cowhage”), see s.v. under pw and Medicinal Plant Names Services , Kew Royal Botanic Gardens, last accessed July 21, 2020.

n.273Tibetan literally reads, “Acts are more powerful when their objects have the nature of diamonds.”

n.274This verse and how exactly it is supposed to illustrate the story of Śoṇottara remains obscure. Our translation is a free rendering of the Sanskrit reading of MS[A], which, at this time, makes most sense to us: Offering someone a soap ball containing cowhage, a plant whose hairy seed pods cause stinging and itching, seems like a harmless prank at first. The action becomes more serious when considering the victim/object of such an action‍—a pratyekabuddha. The origin or prevalence in other Buddhist literature of this verse is unknown. The Tibetan seems to be in closer correspondence with MS[B] (see also Lévi 1932, p. 76, n. 1). MS[A] reads karmāṇi nūnaṃ balikatarāṇi dharmeṣu vajrakalpatareṣu | yatra vaśībhūtā api anubhavāmo duḥkhāni karmāṇi || (Kudo 2004, p. 164).

n.275The identity of this figure is not known (see Kudo 2004, p. 296, for further references).

n.276Reading J, N, and C ba ku la instead of D la ku ba. According to the Pāli sources, Bakula was the son of a householder of Kosambī; see DPPN, s.v.; Kudo 2004, p. 297, n. 63.

n.277MS[A] omits this entire paragraph.

n.278Free rendering of the Tibetan. The Tibetan syntax here is slightly awkward. Tibetan may literally mean: “What kind of action leads to a person not being well in both body and mind” or “When endowed with what [kind of] action is a person not well in both body and mind” (las gang dang ldan na/ gang zag lus kyang mi bde la sems kyang mi bde ba yod de).

n.279According to the Tibetan. The Sanskrit uses the plural form.

n.280According to the Tibetan. Our translation is tentative. It is not clear whether cough or phlegm is meant: the Tibetan reads lud (“phlegm”; see lud pa lu ba “to cough up phlegm”), but the Sanskrit reads kāsa (“cough”).

n.281Translation tentative (see Jäschke 1972, s.v. “dysentery”). The Tibetan reads rims (“infectious disease, plague”), the Sanskrit jvara (“fever”).

n.282According to the Tibetan shu ba; the Sanskrit reads pāṇḍuroga (“jaundice”).

n.283According to the Tibetan. The Sanskrit additionally mentions pāmā (“scabies”).

n.284Tibetan (transliterated Sanskrit) sha ri ka, Sanskrit sārikā. A bird of the genus Turdus in the thrush family. The exact species referred to here is unknown (the blackbird?). According to Böhtlingk (pw, s.v. “sārikā”), a kind of crow (“Predigerkrähe”).

n.285Tibetan (transliterated Sanskrit) kA ran Da va, Sanskrit kāraṇḍava. A kind of duck or goose. The exact genus and species are not determinable. Different religious texts in the South Asian context refer to different kinds of aquatic birds with the name kāraṇḍava.

n.286Tibetan ngur pa, Sanskrit cakravāka. This bird has been identified as Anas casarca (pw) or Tadorna ferruginea (ruddy shelduck). Populations of the eastern hemisphere are migratory birds who winter in the Indian subcontinent. According to Indian mythology, these ducks are monogamous and utter their characteristic cry when separated; see also pw, s.v. “cakravāka.”

n.287Tibetan dred mo, missing in Sanskrit. See, however, Jäschke 1972, s.v. “dred mo”: “hyena.”

n.288Tibetan sbrul nag po, Sanskrit kṛṣṇasarpa. Coluber naja or Naja naja.

n.289Translation with the help of the (emended) Sanskrit, which literally reads “with crooked and underdeveloped sense faculties” (durgandho bhavati jihmendriyo ’vyaktendriyaḥ; Lévi 1932, p. 77). The Tibetan translation does not seem to make good sense. The Tibetan (lce’i dbang po) apparently reads the Sanskrit jihvendriya (“with a taste organ tasting (only) bad smells”; lce dbang po dri nga bar). The idea of this passage seems to be that the animals mentioned are found in bad-smelling places. Lévi (ibid., p. 77, n. 8), however, felt certain that it must be corrected to jihmendriya, which he translates as “les sens obtus” (ibid., p. 141). MS[B] does contain the reading jihmendriya, which was apparently overlooked by Lévi (see Kudo 2004, p. 171; the respective folios are missing in MS[A]).

n.290Tibetan sbrul, Sanskrit ajagara. The Sanskrit refers to large snakes or constrictors such as a boa.

n.291According to the Tibetan sbrang ma, which corresponds to the Sanskrit bhramara. The Sanskrit editions, however, have makṣika (“the fly”).

n.292Tibetan of A (p. 841) mang kun should be corrected to mang ku na, Sanskrit maṅkuna “bug”. Lévi, in his edition of N (1932, p. 204), records the correct reading maṅkuna (= mang ku na). (The original of the N edition was not consulted to verify Lévi’s edition.) The term is missing altogether in all the extant editions of the Sanskrit text.

n.293It is not clear, or not known to us, what exactly “the ten kinds of external things” refers to. The Abhidharmakośabhāṣya mentions eleven tangible things in connection with the skandha of matter (“things that have the nature of being tangible”): spṛśyamekādaśātmakam […] spraṣṭavyamekādaśadravyasvabhāvam | catvāri mahābhūtāni ślakṣṇatvaṃ karkaśatvaṃ gurutvaṃ laghutvaṃ śītaṃ jighatsā pipāsā ceti (Abhid–k–bh 7,9–10 ad AKK I.10b). It seems more plausible, however, that the Karmavibhaṅga wishes to establish a (karmic) connection between the ten nonvirtuous actions or the ten virtuous actions and the state or quality of one’s rebirth environment. Therefore, this may be referring to the five elements and their qualities: earth, water, fire, air, and space; and smell, taste, touch, color, and sound.

n.294According to the Tibetan rgud par ’gyur (“deteriorate, decline”). The Sanskrit reads abhivṛddhiḥ, “growth, increase; success, prosperity” (Lévi 1932, p. 78). See also Kudo 2004, p. 171, for MS[B]: abhivṛddhir (MS[A] lacuna); p. 171, n. 8; and p. 32 (uddeśa), paragraph 51A (n. 13). MS[B] seems to have originally contained two variants of this sentence: “The karmic ripening of the ten virtuous courses of action consists in a proliferation of external things” and “the karmic ripening of the ten nonvirtuous courses of action consists in a decline/destruction of external things.” MS[A] does not contain the second sentence, and the Tibetan translation appears to have confused, or perhaps deliberately interchanged, the two.

n.295For essentially the same, but more systematic, presentation of the karmic results of the ten nonvirtuous courses of action, see the Abhidharma­kośabhāṣya ad AKK IV.85 (La Vallée Poussin/Pruden 1988–90, pp. 669–71).

n.296This is in Sanskrit the adhipatiphala or the “predominating karmic result” of an action according to Vasubandhu’s Abhidharma­kośabhāṣya.

n.297Tibetan las de nyid kyi rnam par smin pas, Sanskrit tasyaiva ca karmaṇo vipākena (according to MS[B], Kudo 2004, p. 173; MS[A] lacuna) implies rebirth under the same conditions under which the action was carried out. Both the Tibetan and Sanskrit of this second form or aspect of karmic result imply that, due to karmic forces other than the action concerned, one is reborn in the human realm, as is explicitly mentioned in the Abhidharma­kośabhāṣya: “If a transgressor is reborn in a human existence” (La Vallée Poussin/Pruden 1988–90, p. 669). This constitutes the second of the different karmic results: the “outflowing result” in the Abhidharma­kośabhāṣya (niṣyandaphala; La Vallée Poussin/Pruden 1988–90, p. 669, IV.85a,b).

n.298According to the Tibetan, except for “locusts,” which follows the Sanskrit reading, where the Tibetan list has ba lang (“ox, bull; cow”). The Sanskrit list contains hail (aśani), birds ( śuka , lit. “parrots”?), locusts (śalabha), mice (mūṣika), and vermin (kīṭa).

n.299According to the Tibetan. The Sanskrit text has lacunae here (see Lévi 1932, p. 78, n. 4; Kudo 2004, pp. 172 and 173) but probably did not read “bad smelling” but “grasses, brush, and forests obstructing one’s passage” (tṛṇavankuśadurga­(va)sarvadurgādīni; Kudo 2004, p. 173, n. 4).

n.300Reading N nag po ’char ldan gyi tshe rabs (see Lévi 1932, p. 78, n. 5) instead of D nag po ’char ldan gyis tshe rabs.

n.301According to the Sanskrit. The Tibetan is not clear. Kudo (2004, p. 304, n. 66) writes “In this section […], three avadānas are referred to: the avadānas of Śvaprapada, Susudhī, and Kālodāyin,” which is true for the Sanskrit text. However, the references in the Sanskrit are unclear, and the Tibetan is not very clear either. There are no equivalents for the Sanskrit names Śvabhrapada and Susudhī in the Tibetan translation‍—the former is completely missing from the Tibetan. Lévi interpreted the Tibetan as referring to a single avadāna, that of Kālodāyin, and translates literally: “here one should relate the jātaka of Kālodāyin, the story of a previous rebirth, placed in a house where the harvest of the year had been good” (Lévi 1932, p. 78, n. 5). The Sanskrit (ibid.; Kudo 2004, pp. 172 and 173) reads tasyaiva karmaṇo vipākena sampannagṛhāvāsaṃ praviśanti. atrāvadānaṃ Śvabhrapādasya Susudhī dārikā Kāsirājñaḥ patnī Devāvataraṇe Kālodāyinaḥ pūrvajanmany avadānaṃ vaktavyaṃ. The Tibetan approximately says, “The karmic result equivalent to the action is illustrated by prosperous laymen and laywomen (lo legs pa’i khyim na gnas pa’i rtog pa brjod pa). Here, one should relate the story of the king of Benares’s wife (ka shi rgyal po dga’ bas) and that of the former birth (rtogs brjod pa) of Kālodāyin in [the town of] Devāvataraṇa.”

n.302According to the Tibetan. The Sanskrit reads “diseases of the teeth” (dantaroga).

n.303According to the Tibetan. The Sanskrit reads “one will obtain/get false explanations/expositions [of the Dharma?]” (abhūtavyākhyānaṃ pratilabhate; Lévi 1932, p. 79).

n.304According to the Tibetan. The Sanskrit additionally reads “there will be division within one’s family and friends” (jātivyasanā mitravyasanā bhavanti).

n.305According to the Tibetan. The Sanskrit reads “gorges and crevices will manifest” (kandaraśvabhrādīni prādurbhavanti; Lévi 1932, 79).

n.306See Jäschke 1972, s.v. “btsun pa,” under sense 3: tshig mi btsun pa (“was explained to me: one whom nobody believes”).

n.307According to the Tibetan. The Sanskrit reads, “one’s possessions will be desired by others” (paraprārthanīyabhogā bhavanti; Lévi 1932, p. 79).

n.308According to Kudo (2004, pp. 174 and 175), MS[A] and MS[B] both read apratikūla, not pratikūla as in Lévi’s edition (Lévi 1932, p. 79). The Tibetan reads yid du mi ’ong ba mthong bar ’gyur ro, which may confirm the reading pratikūla, even though the Tibetan reading does not have the expected standard equivalent (Mvy [Sakaki 2647]: pratikūlaḥ = mi mthun pa).

n.309Degé has the reading rku ba, a verb meaning “to steal, rob.” We translate Y bku ba (“stench”; Jäschke 1972, s.v.), as given in the apparatus of the Comparative Edition of the Kangyur (K: dgu ba).

n.310According to the Tibetan. The Sanskrit differs (Lévi 1932, pp. 79–80): tiktakaṭukabhāvāny api picumandakoṣātakīviṣatiktālābuprabhṛtīni phalāni prādurbhavanti, “bitter and pungent fruits such as the neem tree (picumanda/picumarda), the ribbed gourd (koṣātakī, Luffa acutangula), creepers, the poisonous, bitter bottle gourd, and so forth will manifest.”

n.311According to the Tibetan (’jig rten rgyang pan pa’i gtsug lag la sogs pa la dad par ’gyur ro); the Sanskrit equivalently reads, “[the karmic result will be] the wrong view of annihilation and belief in the treatises of the materialists” (ucchedadṛṣṭiḥ lokāyatādiṣu ca śāstreṣu prasādo bhavati). gtsug lag can render the Sanskrit śāstra (see McKeown 2010, pp. 126–29). It is not clear in this passage, however, whether a particular treatise is intended, or which. No original writings of the Indian Lokāyata or Cārvāka school have survived.

n.312This term is difficult to translate into English with one word. In the English language the terms annihilationist and annihilationism, as well as destructionism or extinctionism, usually denote a discrete Christian sectarian belief, and we have therefore avoided them here.

n.313Tibetan ’jig rten rgyang pan pa, Sanskrit lokāyata (“materialist,” the doctrine or the philosophical school of Cārvāka; see pw, s.v. “lokāyata”). Mvy (Sakaki 3520) gives the alternative (or correct?) spelling ’jig rten rgyang phen pa. Other spelling variants are ’jig rten rgyang ’phen pa and ’jig rten rgyang phan pa. Cārvāka is the name of a disciple of Bṛhaspati, the mythical founder of the philosophical school called lokāyata. Lokāyata‍—literally “turned toward [this] world”‍—is the doctrine or view that there is no other world beyond this empirical world. The meaning or derivation of the Tibetan term is unclear. However, Jamgön Kongtrül (’jam mgon kong sprul), in Light of the Sun, explains the name in a literal sense: “In their thoughts and behavior they act like they cast (’phen) far away (rgyang) any regard or concern for their future lives” (folios 3.a–3.b). However, it seems to us that the Tibetan is a slightly odd but fairly literal rendering of the Sanskrit name of the school: “those who consider (’phen; see Jäschke 1972, p. 357, s.v. “ ’phen pa,” sense 3) [only] the range/extent (rgyang, āyata) of this world (’jig rten, loka).”

n.314According to the Tibetan. The Sanskrit here adds as an example a reference to the chieftain Padāśva (Pāli Pāyāsi; see the Pāyāsisutta, DN 23): yathā padāśvasya rājaputrasya yaḥ kumārakāśyapena śvetikāyāṃ vinīto lokāyatikaḥ (see Lévi 1932, p. 80, n. 3).

n.315Sanskrit reads tathaiṣāṃ daśānām bāhyānām bhāvānām atīva prādurbhāvo bhavati, “[To the degree to which beings cultivate the ten nonvirtuous courses of action,], to that degree the ten outer (material) things will appear in excess” (Lévi 1932, p. 80).

n.316The Sanskrit treats this list of substances differently (Lévi 1932, p. 80): anenaiva ca kāraṇena mahāsaṃvartakalpe bhaviṣyati samayo ’nāgate ’dhvani yat tilā bhaviṣyanti tilapiṣṭaṃ bhaviṣyati tailaṃ na bhaviṣyati ikṣur bhaviṣyati ikṣuraso na bhaviṣyati guḍo na bhaviṣyati. na khaṇḍaṃ bhaviṣyati na śarkarā bhaviṣyanti. gāvo bhaviṣyanti kṣīraṃ bhaviṣyati dadhi bhaviṣyati navanītaṃ na bhaviṣyati na ghṛtaṃ na gḥrtamaṇḍo bhaviṣyati. evam anupūrveṇa sarveṇa sarve rasā antardhāsyanti. MS[A] and MS[B] contain slightly different readings, too; see Kudo 2004, pp. 176 and 177, respectively.

n.317According to the Tibetan. The Sanskrit reads, “In this way, one after the other/step by step, all the different tastes disappear” (evam anupūrveṇa sarveṇa sarve rasā antardhāsyanti; Lévi 1932, p. 80). We have here the same deviation from the Sanskrit as in paragraph 1.­132 above: according to the Tibetan translation, engaging in the ten nonvirtuous actions leads to deterioration of the “ten external things.”

n.318According to the Tibetan. This paragraph and the nine that follow it, corresponding to 51b–61b in Lévi’s edition (1932, pp. 80–82, n. 8), are missing in all extant Sanskrit versions. Also, the karmic results of some of the actions are transposed: while at 1.­137 above, slander leads to painful sensations through stepping on pebbles, etc., the opposite action, at 1.­149, giving up speaking divisively, leads to not living in an environment with deep ravines, etc., the opposite of which is the result of idle talk at 1.­139.

n.319The Tibetan literally means “external things” (phyi rol gyi dngos po rnams).

n.320Free translation of C sa’i bcud dag, which corresponds to the Sanskrit pṛthivīrasaḥ, “juices, saps, essences or the nutrients, potency of the earth or the soil” (Mvy [Sakaki 5286]). D reads mthu dang [gzi byin med par mi ’gyur ro]. See also the expression bhūmirasaḥ in the Abhidharma­kośabhāṣya, AKK III.98 (Abhidh-k-bh(P) 186,27; La Vallée Poussin/Pruden 1988–90, p. 488), which may be a related idea.

n.321This is a tentative translation of the Tibetan ngan skyugs kyi ljan ljin; J and C read ngan skyugs kyi ljon. ngan skyugs usually means “vomiting.”

n.322The following list of five categories that constitute the five precepts or rules of conduct (Sanskrit pañcaśīla, Tibetan bslab pa nga) and their (ten, respectively,) evil consequences are not in the Sanskrit version. The pañcaśīla constitute the very foundation of (lay) Buddhist ethics.

n.323Translating H bzhin instead of D zhing. The parallel sentence structure suggests that this is the preferred reading: mi bde bzhin nyal zhing mi bde bzhin sad pa dang.

n.324Tibetan literally means “at the breakup of the body, following one’s death,” which is reminiscent of a stock phrase in Pāli and Sanskrit Buddhist literature meaning simply death: kāyasya bhedāt. The expression was therefore simplified here.

n.325Literally “wife” (Tibetan chung ma).

n.326Literally “sons” (Tibetan bu).

n.327Tibetan blon po, Sanskrit amātya; literally “relative” (German Angehöriger; Hausgenosse; see pw).

n.328Family in the paternal line: Tibetan nye du, Sanskrit jñāti (German Verwandte väterlicherseits; see pw).

n.329Family in the maternal line: Tibetan snag gi gnyen mtshams, Sanskrit sālohita. The translation of these kinship terms is largely based on the Sanskrit equivalents given in Mvy (Sakaki 3681, 3910, and 3912).

n.330Tentative translation of the Tibetan de’i lus las lha rnams ’phang ba. What exactly this sentence refers to is not clear. Behind it may be the indigenous Tibetan and pre-Buddhist idea of the bla, the life essence or vital principle, which resides in the body and which is connected to the concept of the ’go ba’i lha lnga, the five individual patron deities that are said to be born simultaneously with a child and remain in certain parts of the body to protect the individual (see Samuel 1993, p. 187).

n.331Tibetan gzung ba’i tshig tu mi ’gyur ba might also mean that one’s own words will not be believed or will be incomprehensible. See also 1.­164 below.

n.332Tibetan ’bru’i chang dang sbyar ba’i chang bag med pa’i gnas (= surāmaireyapramādasthāna; Mvy [Sakaki 8505]). Several kinds of alcoholic beverages are known from Buddhist scriptures, summarized in the canonical formula suramaireyamadyapramādasthāna (“the failures of mindfulness due to becoming inebriated by alcoholic drinks made from fermented grains and other ingredients”). Alcoholic beverages were mostly produced from fermented grains‍—mostly rice in India and barely in Tibet‍—but also from various fruits and sugar cane juice. Pramādasthāna (Tibetan bag med pa’i gnas) is sometimes translated as “negligence” (see La Vallée Poussin/Pruden 1988–90, p. 607).

n.333Reading D (dge sbyong). K has dge slong, which corresponds to the Sanskrit bhikṣu (“monk”).

n.334This is a free translation of a stock phrase in Pāli/Sanskrit Buddhist literature: “someone with uncontrolled sense doors.” Pāli indriyesu guttadvāra (“having the doors of the senses guarded, practicing self-control”; PED, s.v. “gutta”).

n.335Literal translation of the Tibetan bud med rnams la shin tu bag med par ’gyur ba.

n.336Cp. this list to the one in The Sūtra of Nandika (Nandikasūtra, Toh 334), 1.­9, and to the one in Vinīta 2010, pp. 127–31.

n.337The Sanskrit text continues again at this point (Lévi 1932, p. 82). See n.­324.

n.338MS[A] reads dvādaśānuśaṃsā (“twelve blessings”; Kudo 2004, p. 178).

n.339According to the Tibetan. The Sanskrit differs: “What are the ten blessings/benefits of paying homage at a tathāgata’s stūpa with the palms of one’s hands joined in reverence, for example at the four major caityas such as the one in Lumbinī or the Mahābodhi temple?” (katame daśānuśaṃsā madhyadeśe caturmahācaityalumbinīmahā-bodhiprabhṛtiṣu tathāgatacaityāñjalikarmapraṇipāte; Lévi 1932, p. 83).

n.340According to the Tibetan rgya chen po’i khyim. The Sanskrit differs: “One will be reborn in Madhyadeśa; one will obtain excellent clothes and an excellent family” (Lévi 1932, pp. 83–84).

n.341According to the Tibetan. The Sanskrit differs; see Lévi 1932, pp. 82–83.

n.342The Tibetan phyag ’tshal ba literally means “making obeisance” (Sanskrit vandanā).

n.343According to the Tibetan. The Sanskrit reads maheśākhya, “distinguished, exalted, eminent; powerful” (Lévi 1932, p. 84). The Tibetan gzi byin chen po (“splendor, glory; majesty, charisma”) seems to render the Sanskrit maheśākhya here. The standard translation equivalent given in Mvy (Sakaki 6411) is dbang che ba/dbang che bar grags pa.

n.344This paragraph has no equivalent in the Sanskrit.

n.345Literally “heated” (Tibetan gdung ba med pa, Sanskrit anavatapta).

n.346According to the Tibetan ’jig rten gyi rten du ’gyur ba; there is no corresponding Sanskrit.

n.347Tibetan gzung ba’i tshig tu ’gyur ba, Sanskrit ādeyavākyo bhavati. Lévi 1932, p. 144: “having persuasive words” (on a la parole persuasive). See also 1.­158 above.

n.348Translating N thos pa instead of D ’thob pa, confirmed by Sanskrit śṛṇoti (Lévi 1932, p. 87; Kudo 2004, pp. 184 and 185).

n.349This paragraph has no equivalent in the Sanskrit.

n.350Tibetan ba dan, Sanskrit patākā; for the meaning of the Buddhist symbol of the flag in Tibetan Buddhist iconography, see Beer 2003, p. 174.

n.351According to the Tibetan. The Sanskrit reads daśa, “ten” (Lévi 1932, p. 89; Kudo 2004, pp. 184 and 185).

n.352Tibetan na bza’, Sanskrit vastra. Mvy, however, gives as the standard translation equivalent for Sanskrit vastra as ras sam gos (Sakaki 5846).

n.353Tibetan khrel yod, Sanskrit apatrāpya.

n.354Sanskrit (Lévi 1932, p. 100) reads “for/of the world” (lokasya).

n.355Sanskrit reads “to be purified” (viśudhyati).

n.356According to the Tibetan (’jig rten pa rnams ’khor zhing ’du bar ’gyur ba dang). This translation is tentative. Sanskrit reads “one will be approachable” (abhigamanīyaś ca bhavati).

n.357According to the Tibetan. The Sanskrit syntax differs: mālābhūto bhavati lokasya, “one becomes the garland of the world” (Lévi 1932, pp. 98 and 147).

n.358According to the Tibetan. The Sanskrit differs slightly: sarvajanapriyo bhavati, “one will be liked by/popular with everyone” (Lévi 1932, p. 98).

n.359According to the Tibetan. The Sanskrit reads māṃsacakṣur naśyati, “the fleshly eye is/will be destroyed” (Lévi 1932, p. 101; Kudo 2004, p. 205, MS[B]‍—MS[A] omits naśyati divyacakṣuḥ).

n.360Literally, “one will obtain the divine eye” (Tibetan lha’i mig, Sanskrit divyacakṣus).

n.361I.e., the wisdom will arise that lets one see what one should do and what one should not do.

n.362According to the Tibetan, which does not explicitly indicate whether mun pa here is intended in a literal sense (as darkness) or metaphorical sense (as the lack of understanding and clarity). The Sanskrit is more explicit: avidyāndhakāro (“the darkness of ignorance”).

n.363Tibetan byug spos. The Sanskrit has gandha (“perfume”; Lévi 1932, p. 151), for which one would rather expect the Tibetan dri (see Mvy [Sakaki 1861]).

n.364See 1.­170 above. Tibetan reads thos pa here instead of ’thob pa, which is most probably an error. The Sanskrit is completely identical for both sentences (Lévi 1932, p. 103; Kudo 2004, pp. 208 and 209).

n.365There is no corresponding paragraph in the extant Sanskrit editions.

n.366The Tibetan text of the following seven paragraphs is not contained in either MS[A], MS[B], or Lévi’s edition. However, they correspond (more or less accurately) to a different, fragmentary manuscript, MS[C], which was also edited by Kudo (2004, pp. 218–24).

n.367Tibetan de bzhin gshegs pa’i mchod rten byed pa’i phan yon bco brgyad yod de. The Sanskrit reads pratiṣṭhāpana (“consecrating”).

n.368The Tibetan here has a slightly variant expression: mig tu sdug par ’gyur ba, instead of blta na sdug par ’gyur ba in earlier instances of the same phrase.

n.369This sense of the Tibetan rim gro byed pa dang ldan pa seems to be confirmed by the roughly corresponding Sanskrit upasthāyakair avaikalyam bhavati, “there will be no lack/shortage of servants” (Lévi 1932, p. 90; see also Kudo 2004, p. 186 and 187). MS[A] reads upasthāyikair vaikalyaṃ bhaviṣyati.

n.370The Tibetan mdog bzang is likely a very literal translation of the Sanskrit adjective varṇavant (“possessing a [nice] color”). The meaning “beautiful” is documented in the PED for the Pāli vaṇṇavant (see PED, s.v.). It is debatable whether the Sanskrit varṇa (“color”) referred to skin color. Since this term is problematic, we have opted for a more neutral translation.

n.371According to the Sanskrit of MS[C], pariṣadam āvarjayati (Kudo 2004, p. 220: viśāradaḥ pariṣadam upasaṃkramya pariṣadam āvarjayati, “Having fearlessly approached an assembly, he/she wins over that assembly”). The Tibetan ’khor ’dun pa is unclear.

n.372While the Tibetan continues to translate phan yon (= Sanskrit a-/ānuśaṃsa), the Sanskrit here reads guṇa (= Tibetan yon tan), “merits, rewards, good results,” instead of ānuśaṃsa in MS[A] and MS[B] (Kudo 2004, pp. 192 and 193; Lévi 1932, p. 94). Only MS[C] reads ānuśaṃsā (Kudo 2004, p. 221).

n.373Lévi 1932, p. 94; MS[B] supratiṣṭhita-caraṇo; MS[A], MS[C] susaṃsthita-caraṇo (see Kudo 2004, pp. 192, 193, and 221, respectively). The Tibetan (shin tu gnas pa) corresponds to MS[B] supratiṣṭhita; see Mvy (Sakaki 265).

n.374The Sanskrit text of MS[A] and MS[B] does not match the Tibetan translation, which seems rather to correspond with MS[C]. See Kudo 2004, pp. 221 and 222; Lévi 1932, pp. 96–97.

n.375For the exact referent of the word shelter as “lodging for travelers,” etc., see Lévi 1932, p. 97, n. 9.

n.376According to the Tibetan grong gi gtso bo, grong khyer gyi gtso bo, grong rdal gyi gtso bo, ri brags kyi rgyal po, rgyal phran, rgyal po btsan pa. The Sanskrit and Tibetan lists do not match. The Sanskrit in Lévi’s edition (ibid., 1932, p. 96) reads: rājā bhavati prādeśikaḥ. rājā bhavati māṇḍalikaḥ. rājā bhavati jambudvīpādhipatiḥ. rājā bhavati dvīpadvayādhipatiḥ. rājā bhavati dvīpatrayādhipatiḥ. rājā bhavati caturdvīpādhipatiś cakravartī. Here, rājā bhavati prādeśikaḥ, “ruler/chief of a district” or “landowner,” seems to have no correspondence in the Tibetan; rājā bhavati māṇḍalikaḥ, “ruler of a province” (Kudo 2004, pp. 194 and 195), may correspond with rgyal phran; Tibetan ri brags kyi rgyal po seems to correspond to Sanskrit rājā syāṃ kārṣarvvaṭayaḥ of MS[C], Kudo 2004, p. 221 (for the Sanskrit kārvaṭika/karvaṭaka [“mountain hamlet”], see BHSD, s.v. “kārvaṭika”).

n.377According to the Tibetan. The Sanskrit differs slightly: rājā bhavati Jambudvīpādhipatiḥ (“ruler of the Jambu continent”), rājā bhavati dvīpadvayādhipatiḥ (“ruler of two continents”), rājā bhavati dvīpatrayādhipatiḥ (“ruler of three continents”), rājā bhavati caturdvīpādhipatiś cakravartī (“ruler of all four continents, a wheel-turning monarch”) (Lévi 1932, p. 96; Kudo 2004, pp. 194 and 195, respectively).

n.378The following list of the different heavens and the stages of liberation and awakening are only contained in the Tibetan translation and in Sanskrit MS[C] (Kudo 2004, pp. 221 and 222).

n.379Reading Y and K tshangs ris kyi lha rnams (instead of D tshangs rigs kyi lha rnams), which is confirmed by the Sanskrit brahmakāyikānāṃ devānaṃ (Kudo 2004, p. 222; see also Mvy [Sakaki 3058]).

n.380Sanskrit ākiñcanyāyatana; the Tibetan term used here (chung zad med pa’i skye mched) is not the standard translation. One would instead expect the standardized form recorded in Mvy (Sakaki 3112): ci yang med pa’i skye mched (kyi lha rnams).

n.381According to the Sanskrit saced ākāṃkṣet pratyekāṃ bodhiṃ sākṣāt kuryām iti. The Tibetan is unclear: byang chub mngon du bya’o (“May [I] realize awakening”?).

n.382Tibetan dpral ba dag, Sanskrit (vi)śuddhalalāṭaḥ. Perhaps a mark of beauty? According to a widespread and popular South Asian belief, Brahmā (or some deity) appears on the sixth day after a child is born to inscribe the child’s fate on its forehead. See the similar passage in Toh 339 (Bruno Galasek-Hul and Lama Kunga Thartse Rinpoche 2021, 1.­86), dpral ba’i dbyes legs pa (“a good-sized [i.e., broad] forehead”). A broad forehead is one of the eighty secondary physical characteristics of a great being or a buddha (aśīty anuvyañjanāni). See also Jäschke (s.v. “dbyes”), who seems to have taken it as a general characteristic of beauty.

n.383According to the Tibetan. The Sanskrit reads prahasitavadanaḥ (MS[C], Kudo 2004, p. 223), “having a laughing face.”

n.384According to the Tibetan phan yon. The Sanskrit MS[C] reads ānuśaṃsā; MS[A] and MS[B] guṇāḥ.

n.385According to the Sanskrit of MS[C], māragocara (Kudo 2004, p. 253), and the Tibetan bdud kyi spyod yul. MS[A] and MS[B] read bālagocara (Lévi 1932, p. 104, n. 3; Kudo 2004, pp. 210 and 211). Sanskrit MS[C] breaks off after the word māragocara (Kudo 2004, p. 223).

n.386According to the Tibetan. The Sanskrit reads “The gods will envy one” (devā asya spṛhayanti).

n.387Tibetan dgon pa la gnas pa, Sanskrit araṇyavāsa. “Forest life” is one of the so-called thirteen dhūtaṅgas/dhūtaguṇas (“ascetic practices”) that are optional (i.e., not prescribed by the Vinaya) for monks and nuns to develop certain qualities such as contentment and detachment.

n.388According to the Sanskrit saṃgaṇikā vivarjayati. The Tibetan translates the Sanskrit saṃgaṇikā as ’du ’dzi (“noise, crowd, bustle, tumult”). The word s aṃgaṇikā (“society, crowd, association”) is well known from Pāli and Sanskrit Buddhist texts (see BHSD, s.v.). That the Tibetan translators rendered it with saṃsargaḥ, the standard translation equivalent suggested by Mvy (Sakaki 6535), is strange. Lévi (1932, p. 104, n. 5) comments that the Tibetan translators apparently were not familiar with its Buddhist usage. Its opposite is “solitude” (viveka), which is addressed in the next sentence.

n.389Translating the D reading (rab tu dben pa la) sten pa. Y ra bsan; K par gnas; H pa la bsten.

n.390The Sanskrit reads dhyānālambanaṃ cittam bhavati (“the mind becomes a support for contemplation”).

n.391These are the two stages of Buddhist meditation: Tibetan zhi gnas, Sanskrit śamatha; and Tibetan lhag mthong, Sanskrit vipaśyanā. This available Sanskrit editions do not use these terms. The Sanskrit edition by Lévi (ibid., 1932, p. 105) reads: “The celibate will attain meditative concentration easily” (brahmacaryasya alpāyāsena samādhim adhigacchati).

n.392“Living on alms” is another of the optional thirteen dhūtaṅgas/dhūtaguṇas (“ascetic practices”).

n.393This translates the reading of D, ston pa, which is confirmed by the Sanskrit dīpayati (Y and K read sten).

n.394Tibetan and Sanskrit literally mean “One will become somebody who will have illuminated [the teachings] for future generations” (Tibetan phyi ma’i skye bo la snang bar sbyas ’gyur ba, Sanskrit paścimāyā janatāyā ālokaḥ kṛto bahavati; Lévi 1932, p. 105).

n.395According to the Sanskrit upaghāta (“injury, damage, hurt; assault”); see CPD, Apte, s.v. “upaghāta.” The Tibetan nyam nga ba does not appear to be a standardized translation equivalent for upaghāta.

n.396According to the Tibetan. The Sanskrit differs: piṇḍapātaparacittasya bhikṣoḥ sarvā diśo ’pratikūlā bhavanti gamanāya (Lévi 1932, p. 105).

n.397One enters to beg for alms food, i.e., one is confident that one will not break one’s vows.

n.398According to the Tibetan gan du ’gro ba. The Sanskrit reads, “with confidence one instructs (anuśāsati) one’s disciples” (see Lévi 1932, p. 105).

n.399According to the Tibetan. The Sanskrit reads: “one’s words will be respected” (Lévi 1932, p. 105: grāhyaṃ cāsya vaco bhavati).

n.400According to the Tibetan. The Sanskrit does not contain this sentence but instead reads, “these are the ten kinds of confidence.”

n.401A, D to’u de ya; J, C to’u da ya. Pāli Todeyya. A rich brahmin from Tudigāma. See DPPN, s.v. “Todeyya.” His story is related in the narrative frame of the Sanskrit version of this sūtra (see Kudo 2004, pp. 2–26 and Lévi 1932, pp. 21–29) and in the commentaries on the Pāli Cūḷakammavibhaṅgasutta (MN 135) and the Subhasutta (DN 10). It is, however, missing from the Tibetan versions. It is the story of the conversion of Śuka. After his death, Taudeya was reborn as a dog in his son’s house, where he is identified by the Buddha on one of his visits during his alms round. The Bhagavān calls the barking dog by the name Taudeya, whereupon it runs into the house, jumps onto the bed, and cannot be removed from it. At first, Śuka does not believe the Buddha’s assertion that the dog is his deceased father and becomes angry. Later, however, he is convinced when the Buddha makes the dog unbury a treasure in the house, of which no one knew but Taudeya. After being the addressee of the Buddha’s teachings on karmic cause and effect on several occasions, Śuka eventually becomes a lay follower.