Notes

n.1We prefer to follow the mainstream Buddhist Sanskrit usage of manuscripts and inscriptions by spelling bodhisatva with a single rather than a double t, the latter being a convention of modern editors. See Gouriswar Bhattacharya, “How to Justify the Spelling of the Buddhist Hybrid Sanskrit Term Bodhisatva?” in From Turfan to Ajanta: Festschrift for Dieter Schlingloff on the Occasion of his Eightieth Birthday, ed. Eli Franco and Monika Zin (Rupandehi: Lumbini International Research Institute, 2010), 2:35–50. Note that this is also the spelling used in Gāndhārī, as well as in Khotanese, Tibetan lexicography, and old Thai documents.

n.2Liland et al., forthcoming.

n.3In Braarvig and Pagel 2006.

n.4Braarvig and Pagel 2006.

n.5Liland et al., forthcoming.

n.6This homage to Mañjuśrī is only included in MS.

n.7This sentence is missing in Tib.

n.8According to Tib and Chi, “You do not strike your ankles against each other when you walk.”

n.9This refers to the idea of trirāśi, the three types of sentient beings. See glossary.

n.10This list is missing in MS.

n.11Dh reads 聖法 (āryadharma) for āryaskandha, interpreting it to refer to the “noble teaching.”

n.12This verse apparently plays on the word pravrajita (rab tu byung ba), which can mean both to go astray (in the first line) and to leave home and become a monk (in the fourth line). This wordplay seems to have been lost to some degree in both the Tibetan and Chinese translations. In Tibetan, it seems that pravrajita was originally translated as mid but then changed to ming by later scribes. The word in the Tibetan manuscript in the first line is clearly ming rather than mid, as there is a tsheg between the word and the shad.

n.13This probably refers to the strength to swim, indicating that the water is a contributing factor for a fish’s ability to move, without being a conscious factor involved in this process.

n.14This sentence, which is missing in MS and Xu but found in Tib and Dh, follows the interpretation of Dh, which seems to make most sense in the context. D reads khyim bdag rnams rba klong dang bral ba’i phyir chos thams cad ni brjod du med pa’o.

n.15The last four sentences are missing in MS and Xu but are found in Tib and Dh.

n.16Skt: cittākulavigama; Tib: sems ’khrug pa med par; Xu: 眼如浮雲聚亂散相; Dh: 眼如浮雲剎那離相. We have chosen to follow the Chinese interpretation of the Sanskrit and understand cittākula as a variation of cittakṣaṇa, employed here to avoid repetition in the next example. Later in the text the same list gives vyaktākula (“ever changing”) translated as gsal ba la brug pa in Tibetan. This would perhaps make more sense.

n.17Vāyuyantra; rlung gi ’khrul ’khor. This is a hapax legomenon as far as we can tell, but this interpretation seems to fit well in the context.

n.18MS does not mention the four elements.

n.19Here and in the previous sentence, “that upon which ignorance rests” follow the Skt pratītyājñānaṃ. D reads mig la brten te mi shes pa.

n.20The deer will cause its own death through the injuries inflicted when trying to pull itself from the trap.

n.21D: mi dang mi ma yin pa (“human and nonhuman beings”) is not mentioned in MS.

n.22MS has the verb in the singular, but we have chosen to read it as plural to conform to the context.

n.23MS and the two Chinese translations here agree that this line should rather read “there will be mendicants of the buddhas.” We have chosen to follow the Tibetan rendering as it seems to make more sense.

n.24The last few sentences, from “I must create,” are present in Tib and both Chinese translations but are missing in MS.

n.25D has bshad, but MS (cchedikāro), both Chinese translations, and Q (kun gcod) agree on “eliminate.”

n.26Skt and Tib here start the sentence with “When he saw this emanation…,” but both Chinese translations do not, and we have chosen to follow the Chinese as it makes most sense in the context.

n.27MS anupaliptā́ś ca bhavanti (“and are not soiled”) does not occur in D, Q, Xu, or Dh at this point, but it does occur when the same phrase is repeated further down. Therefore, as it seems to be a stock phrase in the translation we here follow MS.

n.28The shaking of “ten million fields” refers to the idea that when a buddha attains awakening and touches the ground to be his witness, the entire buddha field shakes and reverberates. Here, ten million (i.e., infinite) buddhas, and thus ten million buddha fields, are referred to.

n.29MS: asaṅga­jñāna­saṃpreṣitaṃ; Xu: 令無礙智; Dh: 觀無礙智故; D and Q: thams cad mkhyen pa’i ye shes la btang ba’i phyir. The Tibetan translation seems to have misread sarva- for asaṅga-.

n.30MS’s guṇavantaḥ and peśalaḥ not in D and Q.

n.31“The disciples of the Buddha” (buddhaśrāvakāḥ) is absent in D.

n.32MS has satvārthat­rayānuttaraṃ, which seems to be an error when compared to Chi and Tib, which the translation here follows.

n.33Here “virtue” translates Skt dharma ; Tib chos.

n.34In a presentation more common in the Tibetan tradition the second statement has “all defiled things” (āsrava, zag pa) instead of “all conditioned things” (saṃskāra).

n.35Skt repeats anupama, while Tib has dpe med and zla med.

n.36This list constitutes the twelve branches of excellent speech (dvādaśaka­dharma­pravacana).

n.37MS: jñeya does not occur in Tib.

n.38MS: manojña (“agreeable”) here occurs for a second time.

n.39D and Q instead have rgya mtsho chen por (“into the great ocean”) here, but we follow MS and Chi, which seem to make more sense.

n.40“The Mahācakravāḍa Mountains” is absent in Skt.

n.41Following MS. D reads these two classes of gods together: gzhan ’phrul dbang byed kyi bdud kyi ris rnams.

n.42This power is later labeled the power of direct knowledge of recollection and realization by means of divine sight (divyacakṣuranu­smṛtisākṣātkriyā­jñānabala).

n.43D and Q read the end of this sentence a bit differently, adding Brahmā and Śakra to the list of those who do not turn the wheel of teaching like the Buddha. This might be a misreading of the Sanskrit, which states that the Buddha turns the holy wheel (brāhmaṃ cakraṃ). This is to some extent supported by the fact that the list is omitted in the Sanskrit.

n.44Sentence missing in MS.

n.45We follow Dh’s interpretation of this phrase and not MS, D, Q, and Xu, which all agree that “anger arises when one attains one’s goal.”

n.46The previous two sentences are missing in MS.

n.47D and Q disagree on this line, Q being closest to MS. Here we follow Q and MS; D: thar ba’i gnas las phyir yang mi ldog ’gyur, Q: mi ldog gyur cing gti mug rgyags pa med.

n.48This sentence is not found in MS.

n.49MS omits a few sentences here that are found in D and Q and which we have translated here. The passage in D reads ’dus ma byas ji lta bar ’dus byas kyang de ltar khong du chud par bya’o/ de ci’i phyir zhe na/ gang chos thams cad kyi rang bzhin de ni dngos po med pa’o/ gang dngos po med pa de la gnyis med de/ lus med pa/ mngon par ’dus ma byas pa de.

n.50This verse is not found in MS but occurs in Tib and Chi.

n.51MS: samāhita; Tib mnyam pa.

n.52The phrase nyams pa med do / de bzhin gshegs pa’i shes rab gaṅ zhe na (“…is never given up. What is the wisdom of the Tathāgata?”) is lacking in MS.

n.53D has mi zad pa (“imperishable” or ”inexhaustible”), but MS has advaya and Q has mi gnyis pa, which seems to be the correct reading.

n.54D: mi myur ba/ rab tu myur ba ma yin not in MS.

n.55This sentence is not found in MS.

n.56This image does not seem to be entirely consistent, but the point seems to be that the number of sentient beings in the universe exceeds the number of water molecules.

n.57This sentence occurs further down in Tib.

n.58MS: vṛddhatara; D and Q: rgan pa dang gzhon pa (vṛddhataruṇa?)

n.59D has mnyam pa, but Q’s reading ma tshim par seems to be correct for MS aśamā and makes more sense.

n.60D: bdag gi bde ba gtong ba in place of astabdhā seems to be a dittography, as it also occurs later.

n.61MS here adds -anusmṛti­sahadharma­bodhyaṅga-, which is not attested in the Tibetan translations and seems redundant.

n.62This sentence in MS does not appear in Tib.

n.63This sentence in MS does not appear in Tib.

n.64Tibetan has mi ’gyur for vikāra (“adjust”) here, perhaps reading avikāra.

n.65“Unconcerned” here translates upekṣa (btang snyoms), which we have translated elsewhere as “equanimity.” The word is used here in a slightly different sense, and we found no single English word that adequately conveys both uses.

n.66These two sentences do not occur in Chi or Tib.

n.67This sentence is missing in Tib.

n.68Part of this sentence is missing in Tib.

n.69Reading kṣetraṃ (“piece of land”). The alternative reading is kaṣṭha, in which case “relying on a dry piece of wood,” i.e., a plow.

n.70MS has śuddhavāci, but Tib and Chi agree on śuddhacāri.

n.71Xu gives trust in actions and in the ripening of their fruits as two separate points, making it a list of ten.

n.72“Practice” here translates dharma (chos).

n.73This is probably a reference to the idea that the chances of obtaining a human birth are the same as the chances of a turtle that comes to the surface of the ocean once every hundred years putting its head through a single small yoke floating somewhere in the great oceans of the world. See for example Saṃyutta Nikāya 56.47 and 56.48.

n.74The Sanskrit has a negation in this expression, implying that the bodhisatva could not, in that existence, honor and serve those who are worthy of veneration. We follow the Tibetan here, which has no negation, implying that the bodhisatva had to honor and serve the beings who ruled over him.

n.75This is probably a reference to the epic battle of the Rāmāyaṇa in which Rāma conquered Rāvaṇa (mānamandara; “the proud one of Mandara”).

n.76MS reads śakya­buddha­guṇa­śikṣarakṣanā, which we interpret as śākyabuddha, though Tib (D) has sangs rgyas yon tan bslab pa bslab nus shes.

n.77This sentence is missing in MS.

n.78This sentence is missing in Tib.

n.79This follows Tib. MS says, “For one who does not understand by means of wisdom, everything becomes difficult to grasp.”

n.80Skt: śrutamūla and Xu: 聽聞根本 agree on this interpretation, but Tib (D and Q) have thos pa dang tshul khrims, apparently reading śrutaśīla.

n.81Skt Vāyuna is missing in Tib.

n.82Skt and Tib do not correspond in the previous six verse lines (Skt is missing a line): arthalābhena mitro smi bhāṣaṃte anyamany ime ’ jambūdvipe [narthaṃ(?) s.a] rvvaṃ buddha­kṣetrā varo balo; D: don rnyed na ni mdza’ ’o zhes/ phan tshun du ni rab tu smra/ don ni byed par ma gyur na/ gcig la gcig ni gsod par byed/ ’dzam bu yi ni gling ’di dang / sangs rgyas kyi ni zhing kun tu; Xu: 善友交談論 義利由斯獲 而眾生不求 反更相加害 我以贍部洲 及諸佛國土. This reading follows the Tibetan.

n.83Skt: aniruddha ; Tib: ’gags pa; Xu: 滅. We follow the Tibetan and Chinese interpretation.

n.84This sentence does not occur in Tib but is present in both Skt and Chi.

n.85D: mi rtag pa does not occur in Skt.

n.86The phrase so tenāntareṇa sapta­buddha­sahasrāṇi ārāgitavān does not appear in the Tibetan.

n.87There is a play on words here that cannot be captured in translation, as the Sanskrit word bhāryā can mean both “wife” and “burden.”

n.88Skt: bhāra.

n.89Skt: māyāgrāma; there is a play on words here in the Sanskrit in that the word for “village of seduction” (māyāgrāma) is similar to the word for “woman” (mātrgrāma). Tib has bu med, indicating that they read mātrgrāma here. The word for seduction can also be translated as “illusion,” with the sense that the woman is a trickster.

n.90This sentence does not occur in Tib.

n.91This question is not found in Tib.

n.92The following section, until the end of the verses, can only be found in MS. It is absent in both Tib and Chi.

n.93The phrase cāribhir gocarāsī, which is here translated as “master of conduct,” is a play on the name of the seer, Cāritragocara.

n.94A satisfactory reading has not been established for this verse of MS, and the translation is therefore tentative.

n.95The word pacyī (“roast”) here could also mean “ripen.”

n.96A satisfactory reading has not been established for the second line of this verse of MS, and the translation is therefore tentative.

n.97The reading of this verse is unclear, and the word “Māra” is a guess.

n.98There is a play on words here that is lost in translation, as the Sanskrit word aṅga means both “factor” and “limb.”

n.99This item is not found in Tib.

n.100Item not found in MS.

n.101Part of this sentence (D: bdud dang bdud kyi ris kyi lha’i bu rnams dang / de ma yin ba gzhan phas kyi rgol ba thams cad kyis, “to attacks by Māra and his retinue, the gods, and all other opponents”) is not found in MS.

n.102This sentence is not found in MS.

n.103This verse in the Sanskrit is exactly the same as the previous one, although the Tibetan translation is different.

n.104Bodhisatvas are not restricted to the present but are connected to the past and the future.

n.105Their goal is awakening.

n.106Śrāvakas.

n.107There seems to be some confusion between the texts at this point, as to who is talking. We have chosen to follow the Tibetan.

n.108There is a play on words here, as the word for “ease” is kṣema , the same as the name of the boy’s father.

n.109Skt: maṇḍala.

n.110The word translated as “true” here is yukta, which we have otherwise translated as “rational” in the surrounding text. The text plays on the different meanings of the word yukta.

n.111Both Tibs have mi dmigs pa, and Dh has 攀緣有所, but Skt āraṃbhataḥ and Xu 緣境 seem to make more sense.

n.112These are the same words that were translated as “vows” and “keeping to vows” above.

n.113Skt: vyādhi; Tib: ’chi.

n.114This follows the Tibetan reading; the Sanskrit does not have the negatives.

n.115The word translated as “suchness” here is tathatā. This is formed from the same word tathā (“such,” “thus”) that we find in tathāgata , “The Thus Come” (āgata) or “The Thus Gone” (gata), i.e., gone or come in the same way as earlier buddhas. But, to the more philosophically minded, the title Tathāgata was also understood as “the one who has attained, or understood, suchness,” according to the scholastic commentary traditions of Buddhism. In the following the text develops the idea of the Tathāgata by playing on the double meaning of the word vigraha, namely “multiplicity” and “body.” The bodhisatva displays this suchness through the multiplicity of his bodily existence for the sake of all beings.

n.116This sentence is missing in MS.

n.117Tib has eighty thousand years old; Xu has eighty; Dh has eighty thousand; since his Dharma lasted for five hundred years we have chosen eighty, since this seems comparatively more reasonable.

n.118Skt has sixty, while both Tibetan renditions have sixty thousand.

n.119Tib has “Pratyekabuddhayāna” instead of “wisdom.” We follow MS.

n.120D: zad pa’i rnam pa (“destructible form”); we follow the Sanskrit reading.

n.121This aspect of learning is not in the Sanskrit but is repeated later when the same list is returned to.

n.122This clause is missing in the Tibetan.

n.123The meaning of line one seems clear, but the Tibetan reading is divergent‍—D: mis ni chos rnyed stong par mi byed do; Q: chos rnyed stong pa nyid du mi byed min. In line 2, Skt has samādhi, whereas Tib has rlom sems. We follow the Sanskrit here.

n.124These two sentences are missing in MS.

n.125This sentence is missing in MS.

n.126“Acts of desire in the past” is not found in Tib.

n.127MS has adrutā vispaṣṭā, but Tib (both D and Q) only has brtags pa ma yin, perhaps a mistake for brtabs pa ma yin.

n.128The Tibetan and Chinese translations differ on the reading of this sentence, and although opposite in meaning, it seems that both ways of understanding it are possible. We have, however, chosen to follow the Sanskrit reading. MS: yā paṃcasūpādāna­skandheṣv aprati­ṣṭhitatā jñānaskandhaparijñedam; D: gang len pa’i phung po lnga rnams la gnas pa/ rnam par shes pa’i phung po yongs su shes pa; Xu: 是名為識不應依趣; Dh: 謂若了知所取五蘊.

n.129This paragraph does not occur in MS.

n.130In MS, ˚ādvay[ā]˚ is missing.

n.131MS has dharmajñānaṃ.

n.132Not in MS; D: nges par gyur pa; Q: skyon med pa.

n.133MS: aprāptā; Tib: thob pa.

n.134There is a play on words here that is lost in translation, as the word we have translated as “phenomena” is dharma .

n.135Skt alpabhāṣyatā not in Tib.

n.136This is a bit obscure to us, and we have not found a reference to support this understanding.

n.137MS: kamanīyatā; Tib: yid du ’ong ba agrees, but Akṣ (p. 550) has las su rung ba (karmaṇyatā; “adeptness”) when providing the same list, which makes more sense.

n.138Skt: āraṃbaṇa; Tib: sgrib pa. We have chosen to follow the Sanskrit; the parallel passage in Akṣ has sgrib pa.

n.139Both Skt and Tib agree on dhyānavimokṣair, but Chi has 彼對治/所對治, and the same sentence in Akṣ has bsam gtan dang mi mthun pa’i phyogs, which would suggest dhyānavipakṣair.

n.140Skt, D, and Q all differ here, and we follow Skt. The anomaly in Q might be a misspelling.

n.141The text seems to imply the opposite, but we have chosen this interpretation as Akṣ reads in this way, and it seems to make more sense.

n.142There seems to be considerable difference between the different texts here, and this reading follows the Sanskrit.

n.143There is some discrepancy between the Sanskrit and Tibetan here, and this reading follows the Sanskrit, which is supported by the same phrase in Akṣ.

n.144MS and Q have this reading, but D has the opposite. In Akṣ the same divergence between Tibetan versions is found.

n.145MS has anāśrayā, but Tib has zag pa med pa (anāsravā); in light of the interpretation of Akṣ (mi gnas pa), we have chosen to follow the Sanskrit reading.

n.146In Xu this is the end of the eleventh chapter, and the following twelfth chapter is entitled 大自在天授記 (“The Prediction of Awakening of Maheśvara”).

n.147Skt, Tib, and Dh all have bodhi instead of bhoga, but Xu reads in this way, suggesting later corruption of the passage. Akṣ also presents the same point as Xu, and as this seems to make more sense we have chosen to follow this reading.

n.148The Sanskrit actually spells this name Dīpavati in this paragraph, but since it is later spelled Dīpapati, we have chosen this latter spelling throughout for clarity.

n.149This is a reference to the single yoke and the blind turtle and is elaborated upon in Xu: 又似盲龜難遇浮孔百千大劫時或一遇我今奉見甚為希有.

n.150The exact nature of this structure is not certain, and the two Chinese translations both give different interpretations. The translation here follows the Tibetan. The Sanskrit for this has not yet been established, and when this becomes available it will become clearer.

n.151This follows Skt. Tib has a smaller number. The same applies for the next reference.

n.152As this says four hundred, and not four thousand, the numbers do not add up in the following section. All available versions, however, agree on this number (except Dh, who has five hundred).

n.153Q has dbul po shin tu chad pa, but this is not supported by any of the other versions.

n.154This information, as well as the following colophon, is found in Tib, but not in MS.