191. gnas gyur pa; āśrayaparivṛtti. I chose to translate this technical term with “shift in one’s basis [of existence]” or “shift of the basis [of existence]” instead of “transformation of the basis” as is usually the case. The āśrayaparivṛtti in Saṃdh. is an attainment that is obtained after the ālayavijñāna has ceased (see Schmithausen 1987, p. 198 and Schmithausen 2014, p. 37). In this sūtra, the ālaya therefore does not seem to be equated with the āśraya. In chapter 10, the basis is evoked in relation to the truth body (dharmakāya). According to 10.2 and 8.35, it appears that the basis one possesses once all corruption has been eliminated is none other than the truth body after it has been purified of adventitious defilements (see Xing 2005, p. 97), at least in the case of the bodhisattvas. The dhāraṇī in 10.8 makes it clear that conceptions of being defiled or purified have in fact no raison d’être. From the perspective of true reality, they are completely adventitious. As stated by the Buddha at the conclusion of 10.8, bodhisattvas exchange the body afflicted by corruption for the body of truth or actual body, the dharmakāya. In line with this interpretation, āśraya is read as a quasi-synonym for kāya as in the expression āśrayapādātṛ (on this term, see Schmithausen 2014, p. 331, §272.1). This reading seems to be confirmed, for example, in YBht P ’i 30b4f (see ibid., pp. 521–22, §483), which explains the āśrayaparivṛtti as the completely purified dharmadhātu, which is permanent and inconceivable. Schmithausen adds, “At the same time, it [this passage of the Ybht] stresses the permanence of āśrayaparivṛtti (in the ontological perspective), precluding thereby a causal process in the strict sense.” Elsewhere, Schmithausen refers to this term as “the [accomplished] āśrayaparivṛtti or purification of the tathatā” (Schmithausen 2014, p. 527, 536ff.). While there is certainly a multiplicity of interpretations with regard to this complex matter, it seems to me that this reading is precisely what is meant in Saṃdh. In that sense, āśrayaparivṛtti, as a result (phala), corresponds to an unveiling (see Schmithausen 2014, p. 537) or purification of the basis in the form of a return, a restoration, a restitution, or a re-entry into the dharmakāya, tathatā, or dharmadhātu. The synonyms given in the list above (true reality, awakening, and nirvāṇa) show that āśrayaparivṛtti does not refer to afflicted dharmas, which would be the case if the ālayavijñāna was meant here. To conclude this discussion, it seems on the basis of Sakuma’s work (Sakuma 1990) that we are in presence of (at least) two models of āśrayaparivṛtti: (1) an originally ontological model, as found in the Śrāvakabhūmi, in which the psychophysical base (lit. the basis of existence) of the person practicing śamatha and vipaśyanā is transformed, as dauṣṭhulya is replaced by praśrabdhi; and (2) a cognitive or epistemic model using this originally ontological terminology to express the purification of the tathatā. In this model, the purification as an elimination of the dauṣṭhulya alone is the manifestation of the dharmakāya that is not the creation of a causal process transforming an entity conceived in ontological terms, as repeatedly stated throughout the later chapters of Saṃdh. In this latter model, the cognitive purification of the tathatā as a causal process can only make sense from the perspective of conventional truth (see chapter 3). From the ultimate standpoint of realization, nothing was ever purified by anyone (cf. dhāraṇī in 10.8). As a consequence of this (and following William Waldron’s suggestion), I would like to make clear that the “shift in one’s basis of existence” referred to in Saṃdh. is a cognitive restoration of the basis in which the attainment of gnosis plays a central role, from the perspective of conventional truth. In the present context, one should therefore refrain from interpreting the term āśrayaparivṛtti as implying any ontological commitment to the process thereby described. To conclude on this point, I understand āśrayaparivṛtti in Saṃdh. as implying a “doctrine of (re-)embodiment” as explained in Radich 2007, p. 1109ff. At the end of the path, one has as a basis of existence the truth body in lieu of the body afflicted by corruption as mentioned above, hence the notion of a shift.
192. chos; dharma.
193. D: byams pa mdo’i sde nas shin tu rgyas pa’i sde dang / rmad du byung ba’i chos kyi sde dang / gtan la bab par bstan pa’i sde’i bar dag so sor gcig tu bzlums te … [F.29.a] mdo’i sde la sogs pa de dag nyid ji snyed bzung ba dang / bsams pa so sor gcig tu bzlums te yid la byed pa ni ’dres pa chen por gyur pa’i chos la dmigs pa yin par rig par bya’o. F, folio 39.a: so sor gcig tu bsdu ba te … bsam pa de dag nyid gcig tu bsdu ba byas te. I followed F and did not translate so sor in the sentence pertaining to the highly universal teaching (’dres pa chen por gyur pa’i chos).
195. rtog pa; vitarka. For the translation of vitarka and vicāra, see Cousins 1992, p. 147.
196. See F, folio 40.a: gsal zhing che bar myong ba’i rjes su dpyod pa. This construction is also found in the following sentences in F.
197. Lit. “experienced as subtle,” according to F, or “the experience of the subtle” according to D.
199. See 8.9: “What is the unimpeded mind? Maitreya, it is the referential object of mental stillness, the mind that takes the image as an object.”
200. See 8.9: “Blessed One, what is a mental appearance? Maitreya, this is the referential object of insight, the conceptual image that is the object of concentration.”
201. nye ba’i nyon mongs; upakleśa.
202. These two expressions refer to dharmapratisaṃvid and arthapratisaṃvid. Lamotte and Powers opted for “object” and Keenan for “meaning” for artha, while dharma is usually translated with “Dharma.” For an interpretation of the term in the sense of the translation suggested above, refer to Nance 2012, pp. 58–59, 72, 74–75, 135, 227–28, and 233–35. The same interpretation is found in Lamotte 1970, p. 1617ff.: dharma is translated with “designation” and artha with “chose.”
203. ji lta ba bzhin du yod pa nyid; yathāvadbhāvikatā. On yāvadbhāvikatā and yathāvadbhāvikatā, see Takasaki 1966, p. 173.
204. mthar thug pa; paryanta, in the sense here of “entirety.”
205. For the Sanskrit of these seven, see Nagao 1964, p. 43.
206. See F, folio 41.a: rnam par rig pa’i de bzhin nyid ni gang / ’du byed rnam par rig pa tsam mo.
207. log par sgrub pa; mithyāpratipatti. In F, sgrub pa is translated with nan tan (see for example F, folio 41.b).
208. D: byams pa de la yang dag par sgrub pa’i de bzhin nyid des ni thos pa thams cad ’dres pa’i chos la dmigs pa’i zhi gnas dang lhag mthong gi bsdus pa’i shes rab mtshungs shing mnyam mo. F, folio 41.b: byams pa gang yang dag pa’i nan tan de bzhin nyid des ni thos pa dang bsams pa dang / bsgoms pa’i ’dres pa la dmigs pa’i zhi gnas dang lhag mthong gis yongs su zin pa’i shes rab mtshungs shing mnyam mo. Xuanzang’s translation confirms the reading found in D: 聽聞正法緣總境界勝奢摩他毘缽舍那所受慧平等平等 (Cbeta, Taishō 676). However, F makes sense from the perspective of the meaning of this chapter.
209. See Z, folio 38.a: rnam par grol bar byed instead of D: rnam par grol bar shes.
210. See Z, folio 37.b, F, folio 43.a: chos de dag nyid la instead of D: chos de dag nyid las.
211. We find slightly variant readings in F and D. D: byams pa de la rab tu rig par byed pa ni mngon sum du byas pa’i chos de dag nyid las rnam par grol bar shes pa dang / gzhan dag la yang rgya cher ston pa dang / yang dag par ston pa gang yin pa ste. F, folio 43.a: byams pa de la shes par byed pa ni gang mngon sum du byas pa’i chos de dag nyid la rnam par grol bar shes pa dang / rgya cher yang gzhan la ston cing ’chad pa dang / yang dag par ston pa’o.
212. In accordance with F, folio 44.b, shes pa should be read here also as ye shes.
213. The phrase is repeated throughout the text. Here shes pa should be read as ye shes, a reading confirmed by F.
215. mtshan ma corresponds here to a reflection (gzugs brnyan; pratibimba).
216. so sor brtags pa; pratisaṃkhyā. D: zhes bka’ stsal pa gang lags pa de la sems kyis so sor brtag pa ni gang lags.
217. rnam par rig pa’i de bzhin nyid; see 8.20, in which the true reality of representations corresponds to the fact that “all conditioned phenomena are just a representation.” In the preceding sentence, I added the term ‘truth’ (dharma) to render a frequent collocation that clarifies the meaning of this statement.
218. Lamotte reads chos dang don respectively as “Dharma” and “chose” (“thing”); see Lamotte 1935, p. 225. Powers reads them as “doctrine” and “meaning”; see Powers 1995, p. 189. However, as explained above (see 8.19–21), I understand these two technical terms to refer to “designation” and “objects of designations.” Lamotte’s rendering of the syntax of the entire passage appears to be inaccurate: “Pour celui qui…, il y a…”
221. skyes pa dang / bud med kyi bsnyen bkur, probably for upasthāna-kāri/-kārikā, “(a woman) serving, doing service to (a man, sexually; said of a courtesan)”; see Edgerton 1953, p. 143.
222. In the case of *viśuddhyālambana, one can read the Sanskrit compound as a genitive or dative tatpuruṣa; see Schmithausen 2014, p. 363, n. 1648 and p. 362, n. 1644. I read rnam par dag pa’i de bzhin nyid as *viśuddhitathatā in accordance with Schmithausen 2014, pp. 362–63, §306.5, and n. 1647.
223. gzugs brnyan; pratibimba; see 8.1–10 for the meaning of the image in the context of contemplative practice.
224. See Schmithausen 2014, p. 366, n. 1664 quoting YBht P’i 83a5f: de de la mi dmigs pa gang yin pa and F, folio 46.b: de’ang mi dmigs pa instead of D, folio 34.b: de la de dmigs pa gang yin pa.
225. lus dang longs spyod la lta ba; 顧戀身財.
226. This reference remains obscure. Could this point be directed at outcast bodhisattvas (byang chub sems dpa’ gdol ba; bodhisattvacāṇḍāla), namely, bodhisattvas taking pride in detachment who practice in the way of hearers? See Conze 1975, p. 438ff.?
227. D has bcom ldan ’das ci tsam gyis na zhi gnas kyi lam yongs su dag pa lags/ byams pa gang gi tshe rmugs pa dang gnyid legs par rab tu choms par gyur pa’o/ /bcom ldan ’das ji tsam gyis na lhag mthong gi lam yongs su dag pa lags/ byams pa gang gi tshe rgod pa dad ’gyod pa legs par rab tu choms par gyur pa’o, but one should read bcom ldan ’das ci tsam gyis na zhi gnas kyi lam yongs su dag pa lags/ byams pa gang gi tshe rgod pa dad ’gyod pa legs par rab tu choms par gyur pa’o/ /bcom ldan ’das ji tsam gyis na lhag mthong gi lam yongs su dag pa lags/ byams pa gang gi tshe rmugs pa dang gnyid dang the tshom legs par rab tu choms par gyur pa’o. See F, folio 47.a, which seems to indicate that the list of terms in D is incomplete and in the wrong order.
228. theg pa chen po dang ldan pa; mahāyānapratisaṃyukta.
229. nyon mongs pa dang shes bya’i sgrib pa; kleśajñeyāvaraṇa.
230. The reading of this passage found in D is problematic on account of the double la particle in the second part of the sentence: de legs par bcom pas thams cad la chags pa med pa dang / thogs pa med pa’i shes pa dang / mthong ba thob cing dgos pa yongs su grub pa’i dmigs pa la chos kyi sku shin tu rnam par dag pa la gnas pa yin no. Compare with VD, folio 77.a: de legs par bcom pas chos kyi sku shin tu rnam par dag pa la gnas pa na/ dgos pa yongs su grub pa’i dmigs pa la/ thams cad du chags pa med pa dang/ thogs pa med pa’i shes pa dang mthong ba thob po. Kǫ, folio 39.a: de legs par bcom pas thams cad la chags pa med pa dang / thogs pa med pa’i shes pa dang / mthong ba thob cing dgos pa yongs su grub pa’i dmigs pa la cha shas kyi sku shin tu rnam par dag pa la gnas pa yin no. S, folio 51.b: de legs par bcom pa’i phyir/ chos kyi sku shin tu rnam par dag pa la gnas pas/ dgos pa yongs su grub pa’i dmigs pa la thams cad du mi thogs mi chags pa’i ye shes mthong ba rab tu thob bo. F, folio 48.a: de bcom pa’i phyir chos kyi sku shin tu rnam par dag pa la gnas pas/ dgos pa yongs su grub pa’i dmigs pa la thams cad du mi thogs mi chags pa’i ye shes mthong ba rab tu thob po. Bd, folio 54: de legs par bcom pas chos kyi sku shin tu rnam par dag pa la gnas pa na/ dgongs pa yongs su grub pa’i dmigs pa thams cad la chags pa med pa dang / thogs pa med pa’i shes pa dang / mthong ba thob po. L, folio 48.a: de bcom pa’i phyir/ chos kyi sku shin tu rnam par dag pa la gnas pas/ dgos pa yongs su grub pa’i dmigs pa la thams cad du mi thogs mi chags pa’i ye shes mthong ba rab tu thob bo. (similar to F, folio 48.a). He, folio 102b: de bcom ldan pa’i phyir chos kyi sku shin tu rnam par dag pa la gnas pas dgongs pa yongs su grub pa’i dmigs pa dang/ thams cad du mi thogs mi chags pa’i ye shes mthong ba rab tu ’thob po/ First, a few general remarks: Kǫ follows the syntax of D with an important variation cha shas kyi sku instead of chos kyi sku in D. In some other editions, such as F, L, and He as well as in Bd (which seems to be a compromise between D and K ǫ, on one side, and F and L on the other side), the syntax of the sentence is quite distinct from D and Kǫ; see for example VD in which entire blocks of text are found in a different order. In addition, we find in other minor variant readings, such as dgongs pa in apposition to yongs su grub pa’i dmigs pa (see Bd) instead of the more usual dgos pa.I therefore suggest the following emendations: One should read ye shes mthong ba instead of shes pa dang / mthong ba since we find in various forms the well-known collocation ma chags ma thogs pa’i ye shes mthong ba in Mvyut: ’das pa’i dus la ma chags ma thogs pa’i ye shes gzigs par ’jug go; ’das pa’i dus la ma chags ma thogs pa’i ye shes mthong ba ’jug pa, atīte ’dhvany asaṅgam apratihataṃ jñānadarśanaṃ pravartate (Mvyut 151) or da ltar gyi dus la ma chags ma thogs pa’i ye shes gzigs par ’jug go; da ltar gyi dus la ma chags ma thogs pa’i ye shes mthong ba ’jug pa, pratyutpanne ’dhvany asaṅgam apratihatam jñānadarśanaṃ pravartate (Mvyut 153).Based on the fact that jñānadarśana results from having discarded the obstructions, a doctrine already present in the Pāli tradition, F and L probably give a better account of the logical sequence of this passage: (1) first, obstructions are eliminated; (2) thereupon, one remains in the dharmakāya, which has been completely purified from these obstructions; (3) as a consequence of this, the insight into the accomplishment of the goal/intention, which is the real object, arises together with gnosis free from attachment and hindrance. In this context, jñānadarśana is interpreted as a dvandva (“gnosis and vision”).My suggestion for this passage would thus be quite close to Bd, folio 54: de legs par bcom pas chos kyi sku shin tu rnam par dag pa la gnas pa na (or gnas pas)/ dgos pa yongs su grub pa’i dmigs pa la thams cad du chags pa med pa dang / thogs pa med pa’i ye shes mthong ba thob po. (underlined words are additions to Bd. dang between shes and mthong ba was omitted).
231. Some editions (e.g., F) include ji ltar in the second clause: “once bodhisattvas have obtained mental stillness and insight, how do they attain the unsurpassable, complete, and perfect awakening?” The difference is not significant because the attainment of the fourth object of mental stillness and insight, the accomplishment of the goal, corresponds to the attainment of awakening (see 8.2).
233. F, folio 48.b reads legs par rtogs pa instead of D, which has simply bzung ba.
234. lhag par btang snyoms; adhyupekṣya.
236. F, folio 49.a ye shes instead of shes pa in D.
237. D reads, de la nang gi so so’i bdag nyid la so sor rang rig pa de bzhin nyid rnam pa bdun so sor rtog pa’i shes pa rnam pa bdun skye bar ’gyur te. Compare with F, folio 49.a: de bzhin nyid rnam pa bdun bden bden pa’i rnam pa nang gi so so rang gis shes par bya ba rab tu rtogs pa’i ye shes skye bar ’gyur te. I think F is more in the spirit of this paragraph than D. The notion of so sor rtog pa’i shes pa is at odds with nang gi so so’i bdag nyid la so sor rang rig pa, to which it stands in apposition in D. Instead, I’d rather read here rab tu rtogs pa’i ye shes as found in F.
238. yang dag pa nyid skyon med pa (D) or yang dag pa mi ’gyur ba (F); samyaktvaniyama.
239. These are the first three of the four objects of mental stillness and insight as explained in 8.2. The fourth is the accomplishment of the goal.
242. Lit. “They are skillful with regard to the arising of the mind as it is if they know the sixteen points of the arising of the mind. The sixteen points of the arising of the mind are the arising of…”
245. D reads yang dag pa’i tshor ba, probably in the sense of yang dag pa’i don gyi (or la) tshor ba, but F, folio 50.a has instead ’gag par ’gyur ba’i tshor ba (but the sentence in F is not built according to a pronominal relative-correlative structure as it is in D).
246. D reads rig pa in the sense of rnam par rig pa, while F, folio 50.a has tshor ba instead, which makes more sense in the present context. This reading is supported by Xuanzang’s reading: 一者所依粗重受二者彼果境界受 (Cbeta, Taishō 676), in which 受 refers to vedanā. I therefore emended the entire paragraph accordingly.
247. In the sense of sensations being the cause for future sensations as explained above (see 8.38.1.iv).
248. Sanskrit reads evam eva; see Matsuda 2013, p. 940 ad Lamotte XIII.39.
249. D: rigs kyi bu rnams dang / rigs kyi bu mo dag gis ’di la shin tu brtson par bya ba’i rigs so. Sanskrit: ayam atra kulaputraiḥ kuladuhitṛbhir vā tīvravyāyāmair bhavituṃ (see Matsuda 2013, p. 940 ad Lamotte VIII.39).
250. de’i tshe, tasyāṃ velāyam (cf. Sanskrit text in Matsuda 2013, p. 940 ad Lamotte VIII.39).
251. D: chos rnams gdags pa rnam gzhag gang yin pa/ /de ni rnal ’byor bag yod don chen yin; in Sanskrit: dharmāna prajñaptivyavasthito yo hi yoge pramattā na mahā [rtha] so hī. (see Matsuda 2013, p. 940 ad Lamotte VIII.40). The sentence is built according to a correlative-relative structure yaḥ … sa…, literally “that which is … is…” The Sanskrit here is of a hybrid nature as pointed out by Matsuda. Instead of pramattā na, apramāda would be expected for the Tibetan bag yod.
252. D: gang dag chos der brten nas rnal ’byor ’dir/ /yang dag brtson pa de dag byang chub ’thob; Sanskrit: taṃ dharmaṃ niśrāya ye atra yoge samyakprayukta te labha(ṃ)ti bodhi (see Matsuda 2013, p. 940 ad Lamotte VIII.40).
253. D: gang dag glags lta de skad rgol ba las/ thar bar lta ba chos kun chub byed pa/ /byams pa de dag rnal ’byor ’di las ni/ /thag ring gnas sa ring ba ji bzhin no; Sanskrit: upāraṃbhaprekṣā iti vādamokṣaprekṣā ye dharmaṃ sarva pu .. ../ (see Matsuda 2013, p. 940 ad Lamotte VIII.40).
254. blo ldan; dhīmān (cf. Sanskrit text in Matsuda 2013, p. 940 ad Lamotte VIII.40).
255. The Tibetan reads sems can don zhes while the Sanskrit has satvārthasāra (cf. Sanskrit text in Matsuda 2013, p. 940 ad Lamotte VIII.40). Sāra means both “real meaning”/“quintessence” and “wealth”/“riches.”
256. D: blo ldan sems can don zhes de dag las/ /lan byed rig nas sems can don brtson min; Sanskrit: satvārthasāro na tu kāra tebhyaṃ viditva satvārthaprayukta dhīmān.
257. Eva in Sanskrit (see Matsuda 2013: p. 940 ad Lamotte VIII.40).
258. … ca … ca in Sanskrit (see Matsuda 2013, p. 940 ad Lamotte VIII.40).
259. zang zing med pa; nirāmiṣa (in the sense of “disinterested, not expecting a reward”; see Matsuda 2013, p. 940 ad Lamotte VIII.40 for the Sanskrit term).
260. Sanskrit: punar (see Matsuda 2013, p. 940 ad Lamotte VIII.40).
261. Anagharatna (cf. Sanskrit text in Matsuda 2013, p. 940 ad Lamotte VIII.40).
262. spongs zhing rgyu; caraṃti bhikṣāṃ (cf. Sanskrit text in Matsuda 2013, p. 940 ad Lamotte VIII.40). The term bhikṣā literally corresponds to the French concept of “mendicité.”
263. gzung bar bgyi; dhārayāmi (cf. Sanskrit text in Matsuda 2013, p. 940 ad Lamotte VIII.41). I suggest reading dhārayāmi, which is a causative present, as an optative here.
265. Sanskrit reads adhimukti; see Matsuda 1995, p. 67.
266. Sanskrit reads suparibhāvitādhimuktikṣanti; see Matsuda 1995, p. 67.
267. See 7.31: “they develop devotion for [this teaching] and commission its transcription into writing. Once it has been put into writing, they keep it in mind, read it, venerate it, propagate it, expound it, chant it aloud, contemplate it, and apply it in their practice.”
268. D reads shin tu bsgoms pa’i phyir bzod pas sa de las yang dag par ’das nas byang chub sems dpa’i yang dag pa nyid skyon med pa la ’jug go. The term bzod pas appears to be out of place here and should have been found in the preceding clause, as is the case in the Sanskrit text, which in addition contains no parallel to phyir.
269. byang chub kyi phyogs dang mthun pa’i chos; bodhyaṅga. D, folio 39.b reads byang chub kyi phyogs dang ’thun ba’i chos ji ltar thob pa dag gis, but F, folio 52.a has byang chub kyi phyogs kyi chos ji ltar thob pa de dag la.
270. D: ’bad pas yan lag des yongs su rdzogs pa yin yang byang chub kyi phyogs dang ’thun ba’i chos ji ltar thob pa dag gis de la mang du gnas par bya ba dang / snyoms par ’jug pa la sred pa dang / chos la sred pa las sems lhag par btang snyoms su ’jug mi nus pas. The expression de la mang du gnas par bya ba corresponds to tadbahulavihārin. Sanskrit: na tu śaknoti yathāpratilabdhair bodhipakṣair dharmais tadbahulavihārī samāpattidharmatṝṣṇāyāś ca cittam adhyupekṣituṃ (see Matsuda 1995, p. 68). The subject is singular masculine. I used a plural for bodhisattva in this chapter to avoid the gender issue since bodhisattvas include both males and females.
272. Sanskrit reads sarvasūkṣāpattidauṣṭhulyamalavigatām (see Matsuda 1995, p. 69).
273. Sanskrit has apramāṇajñānāvabhāsena. It seems that jñāna has been translated by shes pa in D while the more frequent corresponding Tibetan term ye shes is usually found in F. Sanskrit: apramāṇajñānāvabhāsena sanniśrayatām upādāya tasya samādhes tasyāś ca śrutadhāraṇyās tṛtīyā bhūmiḥ prabhākarīty ucyate (see Matsuda 1995, p. 69).
274. The Sanskrit fragment (see Matsuda 1995, p. 69) does not contain the Sanskrit equivalent for me’i ’od, which is most probably agnyarci, as suggested by Matsuda. In this case, understanding ’od/arci as “flame” makes more sense. I translated arcibhūtatva as “set ablaze.”
276. chos kyi tshogs; dharmasaṃbhāra. The Sanskrit (see Matsuda 1995, p. 69) differs from the Tibetan (Go, folio 35.a reads chos kyi tshogs, just like the more recent witnesses): nabhopamsya dauṣṭhulyakāyasya mahāmeghopamena dharmakāyena spharanāc chādanatām upadāya daśamī bhūmir dharmameghaty ucyate. The process described here corresponds to the āśrayaparāvṛtti (i.e., the shift in one’s basis of existence), which is completed on the eleventh stage.
277. I follow Wonch’uk’s ṭīkā (vol. thi, folio 269.a) regarding the meaning of the analogy with the sky/space ( nam mkha’ lta bu): gnas ngan len gyi lus nam mkha’ lta bu la chos kyi sku yongs su rdzogs pa sprin chen po lta bus khyab cing khebs pas de’i phyir bcu pa ni chos kyi sprin gyis zhes bya’o zhes bya ba ni bcu pa chos kyi sprin rnam par bshad pa’o/ /’di ltar bdag dang chos su ’dzin pas yongs su bsgos pa’i sa bon gang yin pa de ni las su mi rung ba’i phyir gnas ngan len zhes bya’o/ /stong pa nyid rnam pa gnyis dang / bdag med pa’i don la sgrib pa’i phyir mdo las nam mkha’ lta bu’o zhes gsungs so/ /sa bcu po de la chos kyi dbyings rnam pa bcu’i chos kyi sku yongs su rdzogs pa ni sprin chen po dang mtshungs te/ chos kyi dbyings mngon du gyur pa ni chos kyi sku yongs su rdzogs pa dang / bsgoms pa las byung ba’i chos kyi sku gnas ngan len nam mkha’ lta bu la khebs pa’i phyir chos kyi sprin zhes bya’o.
278. shes bya’i rnam pa thams cad [mkhyen pa]; jñeyasarvākāra [jñatā-] (see Matsuda 1995, p. 69).
279. On the opposition between affliction and purification in this sentence, see Lamotte 1935, pp. 240–41, n. 13.
281. gang la; yatredānīm (see Matsuda 1995, p. 70).
282. The variant reading found in F differs significantly from the Sanskrit edited by Matsuda in comparison to the Tibetan translation in D. I translated this passage from the Sanskrit (see Matsuda 1995, p. 70): (āścaryā bhaga)van yāvad mahānuśaṃsā mahāphalā anuttarā samyaksaṃbodhī yatredānīm evaṃ ma(hā)saṃmohajālaṃ saṃpracālya mahac ca dauṣṭhulyagahanaṃ samatikramya bodhisatvā anuttarāṃ samyaksaṃbodhim abhisaṃbudhyaṃte. D: bcom ldan ’das gang la de ltar/ byang chub sems dpa’ rnams de ltar kun tu rmongs pa’i dra ba chen po rab tu dral zhing de’i gnas dan len thibs po chen po las yang dag par ’das te/ bla na med pa yang dag par rdzogs pa’i byang chub mngon par rdzogs par ’tshang rgya ba’i bla na med pa yang dag par rdzogs pa’i byang chub ji tsam du phan yon che zhing ’bras bu che ba ni ngo mtshar lags so. F, folio 54.b: bcom ldan ’das ji tsam du bla na med pa yang dag par rdzogs pa’i byang chub de ’bras bu che zhing legs pa che ba dang / de ltar gti mug chen po’i dra ba gsal nas ngan len che zhing sdug pa las yang dag par ’das te/ byang chub sems dpa’ rnams bla na med pa yang dag par rdzogs pa’i byang chub mngon par rdzogs par ’tshang rgya ba ni ngo mtshar che’o.
283. D: chos rnam par dag pa dang / ches shin tu rnam par dag pa yin par rig par bya’o; compare with Kǫ, folio 55.b: ches rnam par dag pa dang / ches shin tu rnam par dag pa yin par rig par bya’o.
284. D: dgos pa med for nirupalepa (see Mvyut 6672)? F, folio 55.b: dgongs pa med pa. D and Xuanzang’s translation agree: 無緣無待發大願心 (Cbeta, Taishō 676).
285. D: khams, but F, folio 56.b reads mkhas pa.
286. D; Kǫ, folio 47.b: khams gya nom pa (syn. khams bzang po), praṇītadhātu (kam) (see Mvyut 7670), 妙界 (see Yokoyama, Kōitsu, and Takayuki Hirosawa, eds., Sanskrit-Tibetan Index for the Yogâcārabhūmi-śāstra, accessed August 29, 2016, http://www.buddhism-dict.net/cgi-bin/xpr-ddb.pl?q=%E5%A6%99%E7%95%8C). As a translation for this term, Lamotte (1935, p. 245) suggests “complexion” and Powers (1995, p. 243) gives “constituents.” However, F, folio 57.a has dam pa’i dbyings su instead of khams gya nom pa las. Obermiller (1933, p. 207, n. 3) gives us the context in which gya nom pa (praṇīta) is usually used: “Emancipation, in the sense that (the extinction of Phenomenal Existence) represents the state of bliss and purity (which is completely free from the defiling forces of the Phenomenal World).” Xuanzang’s translation reads: 亦能獲得上界勝解 (Cbeta, Taishō 676), in which 上界 refers to the formless and form realms.
287. chos la nges par rtog pa’i bzod pa; dharmanidhyānakśānti. The term kśānti is used here in the sense of “acceptance.” In the Kīṭāgiri sutta (M. I, 480), the phrase dhammanijjhānakkhanti is glossed in the following way: “Monks, I do not say that the attainment of gnosis is all at once. Rather, the attainment of gnosis is after gradual training, gradual action, gradual practice. And how is there the attainment of gnosis after gradual training, gradual action, gradual practice? There is the case where, when confidence has arisen, one visits the teacher. Having visited, one grows close. Having grown close, one lends ear. Having lent ear, one hears the Dhamma. Having heard the Dhamma, one remembers it. Remembering, one penetrates the meaning of the teachings. Penetrating the meaning, one comes to an agreement through pondering the teachings.” Kathañca bhikkhave anupubbasikkhā anupubbakiriyā anupubbapaṭipadā aññārādhanā hoti: idha bhikkhave saddhājāto upasaṅkamati, upasaṅkamanto payirupāsati, payirupāsanto sotaṃ odahati, ohitasoto dhammaṃ suṇāti, sutvā dhammaṃ dhāreti, dhatānaṃ dhammānaṃ atthaṃ upaparikkhati, atthaṃ upaparikkhato dhammā nijjhānaṃ khamanti, dhammanijjhānakkhantiyā sati chando jāyati, chandajāto ussahati, ussahitvā tuleti, tulayitvā padahati, pahitatto samāno kāyena ceva paramaṃ saccaṃ sacchikaroti, paññāya ca naṃ paṭivijjha passati.” (Translation and citation are from Punnaji 2017).
288. byas shing bsags pa; 造作増長; kṛta-upacita.
289. rnam grangs is used here as a synonym for thabs.
290. Lit., “Likewise, beings who are suffering from being subject to the suffering resulting from being conditioned cannot be made happy by any method consisting in merely assisting them with material objects.”
291. According to Lamotte and Keenan, this refers to 9.13: “The perfections are without attachment, disinterested, free from faults, nonconceptual, and dedication of merit.”
292. Translating rnam par dag pa with “purity” does not work here as can be seen in the case of the following purifications (2–7), which are formulated according to the exact same lexicographical and syntactical structure as (1).
293. shag kyis ’chags; codanā; 諫誨 (see Yokoyama, Kōitsu, and Takayuki Hirosawa, eds., Sanskrit-Tibetan Index for the Yogâcārabhūmi-śāstra, accessed August 30, 2016, http://www.buddhism-dict.net/cgi-bin/xpr-ddb.pl?q=%E8%AB%AB%E8%AA%A8). For codanā, see Edgerton 1953, p. 234.
294. rigs pa’i gnas lnga po; pañcavidyā. The five sciences are grammar, logic, philosophy, medicine, and crafts.
295. ma ’tshal ba (D), mi zad pa (F); akṣaya.
296. D: sar grub pa.
297. In D consistently expressed through the term ngo bo nyid, while in F rang bzhin is used instead.
298. D: ngo bo nyid dang bcas pa nyid kyang ci’i slad du mi ’dzin lags. F, folio 62.a: rang bzhin ma mchis pa’i rnams kyang ci’i slad du mi ’dzin.
299. D: spyan ras gzigs dbang phyug nga ni ngo bo nyid kyis ngo bo nyid med pa nyid ’dzin par mi smra mod kyi. Compare with F, folio 62.a: spyan ras gzigs dbang phyug nga ni rang bzhin med pa nyid la/ rang bzhin med par mi bzhad de; VD, folio 89.a: spyan ras gzigs dbang phyug nga ni ngo bo nyid med pa nyid kyis ngo bo nyid ’dzin par mi smra mod kyi; and Xuanzang’s translation: 我終不說以無自性性取無自性性 (Cbeta, Taishō 676). The meanings of these translations can be interpreted as being similar, although their phrasing is quite different.
300. nye ba’i pha rol tu phyin pa. The prefix nye ba (for the upasarga upa-) refers here to proximity, intimacy, and immediacy (as defilements manifest in the present case). Reading this term as “subsidiary” or “secondary” does not make sense since the three kinds of transcendence are presented in increasing order from the lesser to the greater.
301. See 9.4: “The eighth stage is called Immovable because what lacks phenomenal appearance is spontaneously accomplished and [the bodhisattvas] are unshaken by the manifestation of defilements resulting from phenomenal appearance.” See also 9.5.
302. F has only two aspects; see F, folio 63.a. Xuanzang’s translation has three: 善男子略有三種 (Cbeta, Taishō 676).
303. For Schmithausen’s translation of 9.29, see Schmithausen 2014, p. 563.
304. Regarding the fact some editions have gnyis (“two”) and others gsum (“three”); see Schmithausen 2014, p. 563, nn. 2311–12.
306. Could this clause, de las gzhan pa’i yon tan rnams, refer to those who are not bodhisattvas?
307. See the verses below in D: de dag ’gal ba snyam du sems pa na/ /rnam par rmongs nas blo gros sna tshogs ’gyur.
308. See translation of VinSg 16 in Sakuma 1990, p. 202: “Der Dharmakāya der Tathāgatas ist dadurch charakterisiert, daß die [ihn konstituierende] ‘Umgestaltung der Grundlage’ daraus hervorgegangen ist, daß man die [Bodhisattva-]Stufen und Vollkommenheiten durch intensive Übung gemeistert hat.”
309. mngon par ’du bya ba med pa; anabhisaṃskāraṇa.
310. On vimuktikāya, see Radich 2007, p. 1254ff.
311. As with the compound in the opening question above, I read sprul pa’i sku’i mtshan nyid (nirmāṇakāyalakṣaṇa) as a bahuvrīhi.
312. byin gyis brlabs; adhiṣṭhita.
313. F, folio 65.a: bstan pa la.
314. dmigs pa la nye bar gtod pa.
315. D: sems can gnas pa, but F, folio 65.b reads sems gnas pa, which corresponds to Xuanzang’s translation: 心安住事 (Cbeta, Taishō 676).
316. mngon du bya ba; sākṣātkāra.
317. gsal ba; paṭu (?). See Mvyut 6695: spyod pa mi gsal ba; apaṭupracāraḥ. Another possibility for gsal ba would be saṃprakhyāna. Edgerton gives as synonym asaṃmoṣa (“absence of confusion”). As an equivalent for saṃprakhyāna, a Tibetan synonym of gsal ba is dran pa. See Edgerton 1953, pp. 83 and 576.
318. D: bsgom pa las yongs su skyob pa’i sbyor ba gsal ba. Compare with F, folio 66.a: bsgom pas yongs su skyob pa’i sbyor ba gsal ba.
319. pham pa’i gnas lta bu [’i chos]; pārājayikasthānīya [dharmāḥ]. See Edgerton 1953, p. 342.
320. bar du gcod pa’i chos; antarāyikadharmāḥ (see Mvyut 9324).
322. D, folio 50.b: gnas pa. F, folio 67.a: rnam par bzhag pa. They are synonyms for vyavasthāna.
323. See 8.21, in which the exact same enumeration is found. See D, folio 31.b: mgo gcig tu lan gdab pa dang / rnam par dbye ba dang / dris te lan gdab pa dang / gzhag pa dang / gsang ba dang.
324. ’thob pa. Usually “obtainment” or “attainment.”
325. D: so so’i shes pa; compare with F, folio 67.a: dam bcas for the Sanskrit pratijñā. Yoshimizu opts for “objects that are known” (see Yoshimizu 2010, p. 142), although it is clear that so so’i shes pa is the Literal translation into Tibetan of pratijñā.
326. Yoshimizu 2010, p. 142 reads the correlative/relative sentence (yat … tat …; … gang dag yin pa de dag …) as meaning “Whatever is…, that is…” Alternatively, this grammatical construction could be literally translated with the following syntactic structure: “That which is … is…” While Yoshimizu’s translation is technically correct, reading gang dag yin pa in the sense of “whatever” in the present case is unnecessary since this grammatical structure is usually used to give a definition of a technical term. As a consequence, we do not need to mirror the Sanskrit correlative/relative structure in English. The result is a more simple and fluid rendering of the text.
327. In this paragraph, I read the compounds ending with lakṣaṇa as bahuvrīhis. Yoshimizu translated these compounds as tatpuruṣas. It seems to me that reading them as bahuvrīhis makes the entire passage much easier to understand.
328. de mngon sum du dmigs pa’i mtshan nyid; tatpratyakṣopalabdhilakṣaṇa (see Mvyut 4405).
329. de la gnas pa mngon sum du dmigs pa’i mtshan nyid; tadāśritya pratyakṣopalabdhilakṣaṇa/tadāśritya āśritapratyakṣopalabdhilakṣaṇa (see Mvyut 4406). See Yoshimizu 2010, p. 144: “the characteristic of the direct cognition [of something] depending on the [imperceptible object to be inferred].” The definition of this term reads, according to Yoshimizu, “The characteristic of the direct cognition [of something] depending on the [imperceptible object to be inferred consists in] such kinds of direct cognition, through which something directly not [cognizable] is inferred.”
330. rang gi rigs kyi dpe nye bar sbyar ba’i mtshan nyid; svajātīyadṛṣṭāntopasaṃhāralakṣaṇa (see Mvyut 4407). The term upasaṃhāra means “establishing,” in the way of the sādhana with respect to the sādhya. It is therefore also translated into Tibetan as nye bar sgrub, a synonym for nye ba sbyar ba, which is used for the Sanskrit sādhana too.
331. lung shin tu rnam par dag pa gtan la phab bar bstan pa’i mtshan nyid. See Mvyut 4409: lung shin tu rnam par dag pas gtan la bab par bstan pa’i mtshan nyid/ lung shin tu rnam par dag pas gtan la dbab par bstan pa’i mtshan nyid; suviśuddhāgamopadeśalakṣaṇa.
332. For the sake of readability, I inverted the order of the clauses in the sentences explaining the five points mentioned here. If we translate the Tibetan literally, the pattern would be: [example 1, example 2, etc.] are [the logical proof to be defined].
333. D: gang gis mngon sum du ma gyur pa la rjes su dpag par bya ba dang / de lta bu dang ’thun pa gang yin pa de ni de la gnas pa mngon sum du dmigs pa’i mtshan nyid yin no. Yoshimizu’s translation of this clause seems incorrect: “The characteristic of the direct cognition [of something] depending on the [imperceptible object to be inferred consists in] such kinds of direct cognition, through which something directly not [cognizable] is inferred, as…” followed by sentences (1), (2), and (3).
334. In this explanation of tadāśritya pratyakṣopalabdhilakṣaṇam, I do not understand why Yoshimizu takes elements of sentence (1) into sentence (2), in violation of the Tibetan syntax which is quite clear in the present case. See Yoshimizu 2010, p. 144.
335. Yoshimizu adds the concept of vipraṇa to this sentence, which is not found in the Tibetan. See Yoshimizu 2010, p. 145.
336. The analogies given as examples seem to be instances of [para]prasiddhānumāṇa ([tha snyad du gzhan la] grags pa’i rjes dpag) in that the perception of the analogy must be renowned (grags pa) in the world (or established from the perspective of the person to be persuaded), thereby offering a certain level of consensus, which is essential for the validity of this kind of logical proof.
337. This probably refers to the various kinds of suffering, which includes the suffering inherent to the conditioned phenomena as well as to the twelve factors of conditioned existence.
338. rang dbang med pa; asvatantra.
339. Yoshimizu segments this passage in a different way (see Yoshimizu 2010, p. 145), as it appears that she did not understand its syntactic structure (or chose not to follow it). The point of these instances in the form of established perceptions for which there is a consensus is to show how one thing (that which must be established) is established from the other (the commonly established perception). The relation here is again of the type sādhya/sādhana, this time through an instance belonging to the same class of phenomena. In these sentences, we have the following construction: X la Y dmigs pa nye bar sbyar ba. In the present case, X (i.e., external and internal conditioned phenomena) is the sādhya, and Y is the sādhana (i.e., clauses 1–2), which makes the svajātīyadṛṣṭāntopasaṃhāralakṣaṇam look like a type of prasiddhānumāṇa.
340. These three proofs are (A) a perception that is a direct cognition of the [thing to establish]; (B) a perception that is a direct cognition [of something existing] in dependence on the [thing to establish]; and (C) a demonstration through an instance belonging to the same class.
341. See beginning of 10.7 above. Yoshimizu translates rigs pa brtag pa yongs su dag pa with “the reasoning to be investigated.” See Yoshimizu 2010, p. 145.
342. dge sbyong bzhi for dge sbyong chos bzhi; see Mvyut 8708: dge sbyong du byed pa’i chos bzhi ming la; catvārah śramaṇakārakadharmāḥ.
343. This passage is interesting because the five defining characteristics of the principle of reason are reduced to three core ideas. Since pariniṣpannalakṣaṇa merely refers to the definition of correct reasoning with regard to the other four defining characteristics, it is understandable that it is not included in this list. However, it is fascinating to see that svajātīyadṛṣṭāntopasaṃhāralakṣaṇa is also excluded here, which might confirm that this proof was considered to be a form of prasiddhānumāṇa in spite of its seemingly inductive character resulting from the use of instances or examples upon which there is a consensus. However, the “engine” of the proof in the case of this valid cognition is not an induction but the deduction ensuing from facts that are accepted as conventions by virtue of consensus. As a consequence, one could argue that the svajātīyadṛṣṭāntopasaṃhāralakṣaṇa has a monotonic aspect explaining why it is not inductive in spite of its empirical aspect. The analogies used in this kind of reasoning are, in a way, carved in marble, in the sense of well-established principles that cannot be refuted by new information drawn from further experience or perception, which is precisely the reason why these reasonings have the capacity to establish the sādhya. They are by nature a deduction from a universal law or principle. Hence their possible inclusion in the category of anumāṇa as prasiddhānumāṇa since they surely do not correspond only to a direct cognition.
344. de las gzhan dang ’thun par dmigs pa’i mtshan nyid. See Mvyut 4410: de las gzhan dang mthun par dmigs pa’i mtshan nyid/ de las gzhan dang mthun par mngon sum du dmigs pa’i mtshan nyid; tadanyasārūpyopalabdhilakṣaṇa.
345. de la gzhan dang mi ’thun par dmigs pa’i mtshan nyid. See Mvyut 4411: de las gzhan dang mi mthun pa mngon sum du dmigs pa’i mtshan nyid/ de las gzhan dang mi mthun par dmigs pa’i mtshan nyid/ de las gzhan dang mi mthun par mngon sum du dmigs pa’i mtshan nyid; tadanyavairūpyopalabdhilakṣaṇa.
346. thams cad ’thun par dmigs pa’i mtshan nyid. See Mvyut 4412: thams cad mthun par dmigs pa’i mtshan nyid; sarvasārūpyopalabdhilakṣaṇa.
347. thams cad mi ’thun par dmigs pa’i mtshan nyid. See Mvyut 4413: thams cad mi mthun par dmigs pa’i mtshan nyid; sarvavairūpyopalabdhilakṣaṇa.
348. gzhan gyi rigs kyi dpe nye bar sbyar ba’i mtshan nyid. See Mvyut 4414: gzhan gyi rigs kyi dpe nye bar sbyar ba’i mtshan nyid; anyajātīyadṛṣṭāntopasaṃhāralakṣaṇa.
349. This refers to point (IV) above.
350. For an examination of similarities between Saṃdh. and Hetuvidyā, see Yoshimizu 2010.
351. This refers to point (IV) above.
352. This refers to point (I) above.
353. This point is not easy to unravel. I understand it in the following way. If the perception [of a logical proof] that does not conform with any[thing] could be used to establish the thesis, it also could be used to establish as well something that is not the thesis. As a consequence, it would be included in the reasons proving that which is not the thesis and would be therefore inconclusive. In other words, the perception of the proof would be present in both the sapakṣa and the vipakṣa. The proof of something that does not conform to anything would be necessarily also found in the perception of that which does not conform with that which must be established. As one proceeds to examine the proof, its absence of conformity with the premise is enough to disqualify it, whether it is conforming with something other than the premise or with nothing else.
354. This refers to point (VI) above.
355. This refers to point (III) above.
356. This refers to point (II) above.
357. This point is also not easy to understand. I take it to mean the following: if a perception that conforms with all [things] (i.e., with anything) demonstrates the thesis, it follows that the dharmin/pakṣa (all phenomena) constitutes the entire sapakṣa and there is not even the possibility of having a vipakṣa. The demonstration based on such perceptions is therefore inconclusive because it represents a tautology based on a circular argument that nothing could invalidate in the absence of a sapakṣa and a vipakṣa.
358. “Whether tathāgatas…”; see also 4.10 and 7.9. This quote is found in various other canonical scriptures with minor variations.
359. F, folio 69.b: ye shes.
360. ldem por dgongs pa.
361. I read kun nas nyon mongs pa’i chos and rnam par byang ba’i chos as bahuvrīhis.
362. The verses might have been corrupted. The prose section (D, folio 53.a) reads, ’jam dpal kun nas nyon mongs pa’i chos gang dag yin pa dang / rnam par byang ba’i chos gang dag yin pa de dag thams cad ni g.yo ba med pa gang zag med pa yin te/ de’i phyir ngas chos rnams rnam pa thams cad du byed pa med par bstan to, which does not match the variant reading found in the verses (D, folio 53.b): kun nas nyon mongs chos dang rnam par byang ba’i chos/ /thams cad byed pa med cing gang zag med pa yin/ /de phyir de dag byed pa med par ngas bshad do.
363. 10.8 before the gāthās, in which the dhāraṇī is given, could be seen as a mchan ’grel of these verses (i.e., a ‘fill-in commentary’).
365. sprul pa; nirmāṇa. From the Sanskrit point of view, the juxtaposition of anabhisaṃskāra and nirmāṇa must have created a cognitive dissonance as it represents a paradox that can only be solved through the notion of nonduality, the topic of the next paragraph. The term anabhisaṃskāra expresses the notion of something that is uncreated, not brought about, and not the result of any conditioning process— something uncontrived, effortless, spontaneous. In contrast, nirmāṇa implies creation, construction, emanation, formation, composition, and transformation. A solution to this quandary that would not invoke nonduality is to understand the term anabhisaṃskāra as stressing primarily the idea of effortlessness as in the example of the dream given by the Buddha below. Another interpretation could be the apparition in a mirror. A reflection may seem real but is actually neither going nor coming anywhere. It is unproduced and nonexistent, not even “a thing.”
366. The question is to determine how the arising of anything is possible on the level of the relative truth in the absence of a causal process.
367. D: ’jam dpal sems yod pa yang ma yin/ sems med pa yang ma yin te/ sems rang dbang med pa nyid dang / sems kyi dbang nyid yin pa’i phyir ro.
368. yul; viṣaya.
370. Advayalakṣaṇa can be read in various ways: (1) as a genitive tatpuruṣa: “the defining characteristic of the nondual/of nonduality”; (2) as a karmadhāraya: “the nondual defining characteristic”; or (3) as a bahuvrīhi (which occurs frequently with lakṣaṇa as the second member of the compound): “who has the defining characteristic of nonduality/the nondual defining characteristic” or “who is characterized by nonduality.” On the basis of the question and the first part of the answer (D: bcom ldan ’das de bzhin gshegs pa’i mngon par rdzogs par byang chub pa gang lags pa dang / chos kyi ’khor lo bskor ba gang lags pa dang / yongs su mya ngan las ’das pa chen po gang lags pa de dag gi mtshan nyid ni/ ji lta bur rig par bgyi lags/ ’jam dpal gnyis su med pa’i mtshan nyid yin te), I would tend to read the compound as a genitive tatpuruṣa: “Mañjuśrī, [you should understand it] as the defining characteristic of nonduality.” However, the following sentences in the answer are built with the verb yin, implying a series of expressions referring to the tathāgatas through the use of nominalized verbal adjectives such as byang chub pa (bodha) or bskor ba; see D, folio 54.a–b: mngon par rdzogs par byang chub pa yang ma yin/ mngon par rdzogs par byang ma chub pa yang ma yin/ chos kyi ’khor lo bskor ba yang ma yin/ chos kyi ’khor lo mi bskor ba yang ma yin/ yongs su mya ngan las ’das pa chen po yang ma yin/ yongs su mya ngan las ’das pa chen po med pa yang ma yin te/ chos kyi sku shin tu rnam par dag pa nyid kyi phyir dang / sprul pa’i sku kun tu ston pa nyid kyi phyir ro. Lamotte translated the passage into French according to the structure defined by “il n’y a pas,” which would correspond to the verb yod in Tibetan, not yin. As a consequence, I chose to translate advayalakṣaṇa as a bahuv rīhi qualifying the tathāgatas.