² v. Lankāvatārasūtra, ed. Nanjō, , 2, pp. 63–6Google Scholar, and the other sources quoted in Ruegg, , Théorie, 74 f.Google Scholar

³ Saṃdhinirmocanasūtra 7.15, 24Google Scholar; cf. Théorie, 73–4.Google Scholar

⁴ Laṅkāvatārasūtra 2, p. 63.Google Scholar

⁵ v. Laṅkāvatārasūtra 2, pp. 63–5Google Scholar; MSA Bh. 3.2.

⁶ Laṅkāvatārasūtra 2, pp. 65–6; MSA Bh. 3.11: aparinirvāṇadharmaka. There are two categories of persons not attaining nirvāṇa, those who do not attain it for a certain length of time (tatkālāparinirvāṇadhannan) and those who never do so (atyantāparinivāṇadharman). The theory that some persons are destined never to attain nirvāṇa and buddhahood is considered characteristic of the Yogācārin school, which does not admit the doctrine of universal buddhahood implied by the usual interpretation of the ekayāna theory (see Saṃdhinirmocanasūtra 7.24) and the theory of the tathāgatagarbha present in all sentient beings. (MSA 9.37 does not, it seems, refer to the fully developed tathāgatagarbha theory which is based on three factors—the irradiation of the dharmakāya, the non-differentiation of the tathatā, and the presence of the gotra [see RGV 1.27 f.]—and concerns only the non-differentiation of the tathatā, and the tathāgatatva, which all beings possess as their embryonic essence. Cf. below, n. 50.)

The agotra doctrine to the extent that it assumes a class of spiritual ‘outcastes’ being evidently incompatible with the tathāgatagarbha theory, the question arises as to the significance of the allusion to persons without a gotra in RGV 1.41. The reference there seems to be to a hypothetical case (opposed to the author's own view expressed before in RGV 1.40–41c), which is not, however, admitted by the author; and the revised reading of pāda 1.41d agotrāṇāṃ na tad yataḥ (cf. Schmithausen, L., WZKS, XV, 1971, 145)Google Scholar ‘since this is not so for those without gotra’ makes this interpretation easier (see p. 346). Indeed, according to RGVV 1.41, any allusion to an. icchantika who does not attain nirvāṇa is to be interpreted as referring to a certain interval of time (kālāntarābhiprāya) only, and not to a permanent incapacity. On the icchantika cf. Ruegg, D. S., Le traité du tathāgatagarbha de Bu ston Rin chen grub, Paris, 1973, p. 12, n. 1Google Scholar. The aparinirvāṇagotra is also mentioned in RGVV 1.32–3, 1.38, and 1.41, and the aparinirvāṇadharman in 1.41.

⁷ cf. MSA, ch. 3; Madhyāntavibhāgabhāṣya and °ṭīkā, 2.1, 4.15–16.

⁸ cf. Théorie, 123f.Google Scholar

⁹ v. Théorie, 177f.Google Scholar; MSA 11.53–9; Madhyāntavibhāgaṭīkā 3.1a, 22. On the equivalence of nirvāṇa and buddhahood, see RGV 1.87.

¹⁰ The meanings ‘family’ and ‘seed’ have been radically separated by Edgerton, F., Buddhist Hybrid Sanskrit dictionary, New Haven, 1953Google Scholar, under his headings (1) and (3) respectively (see above, p. 354).

¹¹ Bodhisattvabhūmi 2.4 (pp. 326–7)Google Scholar. Cf. triratnavaṃśa in RGVV 1.26.

¹² Daśabhūmikasūtra 1U (= p. 16)Google Scholar. Cf. MSABh. 5.4–5.

¹³ Buddhāvataṃsaka, ch. 43, fol. 76 a 7 (Peking ed.). (Cf. Takasaki, J., IBK, VII, 1, 1958, 48–53.)Google Scholar

¹⁴ Gaṇḍavyūha, ch. 31, p. 221Google Scholar: ayaṃ sa tathāgatagarbha āgacchati, yaḥ sarvasattvānām avidyāṇḍakośaṃ nirbhetsyati/ ayaṃ sa dharmarājakulodita āgacchati, yo ‘saṅgavaravimaladharmarājapaṭṭam ābandhiṣyati/. Here tathāgatagarbha is translated into Tibetan as de bžin gšegs pa'i sñiṅ po can; and since the particle can often renders a bahuvrīhi compound, the meaning may be ‘born from the Tathāgata’ which answers to dharmarājakulodita ‘arisen in the family of the Dharmarāja (= Tathāgata)’. Cf. buddhagarbha (p. 482, 1. 26)Google Scholar = saṅs rgyas sras. See now Jikidō, TakasakiNyōraizō shishō no keisei Tokyo, 1974, 17f.Google Scholar

¹⁵ It is to be noted that the Abhidharmakośabhāṣya associates gotra = dhātu = ākara and gotra:jāti in 1.20: gotrārtho dhātvarthaḥ/ yathaikasmin parvate bahūny ayastāmrarūpyasuvarṇādigotrāṇi dhātava ucyante evam ekasminn āśraye saṃtāne vā aṣṭādaśa gotrāṇi, aṣṭādaśa dhātava ucyante/ ākarās tatra gotrāṇy ucyante/ ta ime cakṣurādayaḥ kasyâkarāḥ/ svasyā jāteḥ/ sabhāgahetutvāt/. And Yaśomitra comments (Wogihara, , p. 45)Google Scholar: ākara iti prakṛtam/ … ãkaro dhātuḥ, yato hi suvarṇādyutpattis te teṣām ākarāḥ/.

Thus gotra has the meaning of element/cause (dhātu; cf. below, n. 30; RGVV 1.149–52) or source (ākara ‘mine’), and it is then the source of its class (jāti) by reason of being a homogeneous cause.

Compare also the dhātugotra ‘mine of mineral elements’ in the Vinaya of the Mūlasarvāstivādins (ed. Dutt, N., Gilgit manuscripts, 111, pt. 1, p. 106)Google Scholar and in the Divyāvadāna (8, pp. 111–12).Google Scholar

Concerning the philosophical and soteriological use of the word gotra, the Gaṇḍavyūha contains some especially revealing passages. Thus, in the expressions (prajñā-)ākaragotra (-sambhava) (p. 40, 1. 6)Google Scholar and (sarvaratnasambhavotpatti-)gotrākara(-mūlajñāna) (p. 451, 1. 2)Google Scholar, the words gotra and ākara seem to reinforce each other (of. below, n. 78). In the same text is found the collocation of the expressions gotra and ākara (p. 495Google Scholar: … sarvabodhipakṣyadharma-ratnākaratayā gotrabhūtaṃ sarvaśukladharmasambhāvatayā ākārabhūtaṃ [read ākarabhūtaṃ ?] … bodhicittam). We also meet in it the parallel expressions (sarvatathāgatasambhava-)jñānākaragarbha (p. 282, 1. 25)Google Scholar, jñānagarbha (p. 483,11. 1–2)Google Scholar, and jñānadhātu (p. 484,1. 16)Google Scholar. Cf. also sarvatathāgatakulagotrasambhavagarbha (p. 366,1. 19, pp. 368–9)Google Scholar and kulagotrasambhava (p. 503,1. 26)Google Scholar. On the Gaṇḍavyūha as a forerunner of the tathāgatagarbha doctrine see now Takasaki, , Nyōraizō shishō no keisei, oh. iiiGoogle Scholar. And on the use of dhātu in the Akṣarāśisūtra see Ruegg, , Théorie, 145.Google Scholar

¹⁶ On ‘cat's eye’ as the meaning of vaiḍūrya see Finot, L., Lea lapidaires indiens, Paris, 1896, pp. xlv–xlviiGoogle Scholar; Vogel, C., IIJ, IX, 4, 1966, 270.Google Scholar

¹⁷ Dośobhūmikasūtra 11D (= p. 204): tadyathâpi nāma bho jinaputra mahāmaṇiratnaṃ yadā daśa ratnagotrāṇy atikramyâbhyutkṣiptaṃ ca bhavati kuśalakarmārasuparitāpitaṃ ca …, evam eva bho jinaputra yadā bodhisattvānāṃ sarvajñatāratnacittotpādo daśāryagotrāṇy atikramyotpanno bhavati dhutaguṇasaṃlekhaśīlavratatapaḥsuparitāpitaś ca …/.

The meaning of the word ratnagotra is here clearly established by the context. (However, in the Laṅkāvatārasũtra [p. 1, 1. 7]Google Scholar, in the expression nānāratnagotrapuṣpapratimaṇḍita, ratnagotra seems to mean not ‘mine of jewels ‘but’ various kinds of jewels’, as translated by Edgerton s.v. gotra (4), gotra here being then equivalent to jāti. But even there it would perhaps be possible to understand the expression also as ‘adorned with flowers made of jewels’.)

For the idea set out in the Daśabhūmikasūtra compare the parable of the (bodhisattvacittotpādasarvajagadvyūhagarbha- and sarvajñatācittotpādendranīla-)mahāmaniratna in the Gaṇḍavyūha (53, p. 499)Google Scholar, where the term gotra however does not appear; there the sarvajñatācittotpādātyanta-vimalaviśuddhaprabha-maṇiratna also surpasses the ratnākāras (read: ratnākara ‘jewel-mine’ ?) of qualities of the worldlings, disciples (śaikṣa, presumably the Śrāvakas) and Pratyekabuddhas.

¹⁸ Dhāraṇīśvararājasūtra (= Tathāgatamahākaruṇānirdeśasūtra in the bKa'·'gyur, fols. 176b–177a of the Peking ed.) quoted in RGVV 1.2: tadyathā kulaputra kuśalo maṇikāro maṇiśuddhi-suvidhijñaḥ/ sa maṇigotrād aparyavadāpitāni maṇiratnāni gṛhītvā tikṣṇena khārodakenotkṣālya kṛṣṇena keśakambalaparyavadāpanena paryvadāpayati/ na ca tāvanmātreṇa vīryaṃ praśrambhayati/ tataḥ paścāt tīkṣṇenâmiṣarasenotkṣālya khaṇḍikāparyavadāpanena paryavadāpayati/ na ca tāvanmātreṇa vīryaṃ praśrambhayati/ tataḥ sa paścān mahābhaiṣajyarasenotkṣālya sūkṣmavastraparyavadāpanena paryavadāpayati/ paryavadāpitaṃ câpagatakācam abhijātavaiḍūryam ity ucyate/evam eva kulaputra tathāgato 'py apariāuddhaṃ sattvadhātuṃ viditvânityaduḥkhānātmāśubhodvegakathayā saṃsārābhiratān sattvān udvejayati/ … tataḥ paścād avivartyadharmacakrakathayā trimaṇḍalapariśuddhikathayā ca tathāgataviṣaye tan sattvān avatārayati nānāprakṛtihetukān/. utkṣālya, is Johnston's emendation for unmīlya of the MS.

¹⁹ For khārodaka the bKa'·'gyur translation has bul thog gi chu ‘soda solution’, while rṄog·Blo·ldau·šes·rab in his translation of the RGVV has lan tshva'i chu rnon po ‘sharp salt solution’. khāra is Prakritic for kṣāra ‘caustic’, etc.

²⁰ For āmiṣarasa the bKa'·'gyur translation has dṅul chu ‘mercury’, while the Tibetan translation of the RGVV has two different readings: zaṅs kyi khu ba in the sDe·dge ed. and zas kyi khu ba in the Peking ed. The reading zas corresponds to āmiṣa in Johnston's ed. of the Sanskrit, but zaṅs could be based on ariṣṭa (?).

²¹ Compare the mahāvaiḍūryamaṇiratna recognized by the king in the passage quoted above from the Daśabhūmikasūtra.

²² The tathāgatanetrī (de bžin gšegs pa'i tshul) or buddhanetrī is the prajñāpāramitā. Cf. Haribhadra, , Abhisamayālaṃkārālokā, pp. 218, 267, 939.Google Scholar

²³ These are the three vimokṣamukhas, which are thus characteristic of the second stage of the teaching of the Buddha.

²⁴ trimaṇḍalapariśuddhikathā. On the meaning of trimaṇḍala (which usually refers to the three aspects of the action, the agent and the beneficiary of the action, as in RGV 5.14) see Tathāgatamahākaruṇānirdeśa, fol. 137b.

²⁵ Sāgaramatiparipṛcchā quoted in RGVV 1.68. This parable is to be found also in the *Anuttarāśrayasūtra according to Paramārtha's version of the Mahāyānasaṃgrahabhāṣya (pp. 259c–260aGoogle Scholar; of. Jikidō, Takasaki, A study on the Ratnagatravibhāga, Rome, 1966, p. 249, n. 379).Google Scholar

²⁶ v. Ruegg, , Théorie, 411f.Google Scholar

²⁷ Read probably tathāgato.

²⁸ This verse is quoted in RGVV 1.2, immediately after the passage from the Dhāraṣīśvararājasūtra cited above.

²⁹ The reading viśuddhagotraṃ tathāgatadhātum abhisaṃdhāya is presumably to be corrected to viśuddhigotram … in accordance with RGVV 1.40 buddhadhātuviśuddhigotra (cf. 1.41: āgantukamalaviśuddhigotra). See below, n. 44.

³⁰ On gotra: dhātu: hetu: nimitta see RGVV 1.149–52; Madhyāntavibhāgabhāṣya 1.15; Abhidharmakośabhāṣya 1.20 (above, n. 15).

Cf. also the use of dhātu in MSA 3.2 (where the Bhāṣya refers to the Akṣarāsisūtra; cf. Ruegg, , Théorie, p. 145, n. 3)Google Scholar and nānāprakṛtihetukāḥ sattvāḥ in the Dhāraṇīśvararājasūtra quoted in RGVV 1.2 (p. 6, 1. 6).

³¹ The pronoun tad refers back to what is said in RGV 1.144.

³² The translation ‘lineage’ is also applicable here because in RGV 1.151–2 the reference is to the production of ‘Bodies’, and the biological connotation thus coexists with the mineral and botanical images of 1.149 which are developed in the sequel.

³³ These two similes are based on examples V and VI of the TGS (cf. RGV 1.112–17 and above, p. 345).

³⁴ v. Bodhisattvabhūmi 1.1. These verses are probably commentary-verses (cf. Takasaki, Study, introduction; Sohmithausen, , WZKS, XV, 1971, 125).Google Scholar The epithet uttara can also refer to the fact that the samudānīta is ‘subsequent’ to the prakṛtistha-gotra.

³⁵ Or: ‘by nature’ (raṅ bžin gyis). This comparison is based on example VII of the TGS (cf. RGV 1.118–20).

³⁶ This comparison is based on example VIII of the TGS (cf. RGV 1.121–3).

³⁷ This comparison is based on example IX of the TGS (cf. RGV 1.124–6).

³⁸ On the two possible interpretations of the compound tathāgata-garbha, either as an adjectival (bahuvrīhi) compound—‘all sentient beings have the tathāgata as then- embryo-essence’ —or as a noun (tatpuruṣa compound)— ‘all sentient beings are essential embryos of the tathāgata’, see above, p. 353.

³⁹ RGV 1.27–8: buddhajñānāntargamāt sattvarases tannairmalyasyâdvayatvāt prakṛtyā

bauddhe gotre tatphalasyôpacārād uktāḥ sarve dehino buddhagarbhāḥ ‖

samtmddhakāyaspharaṇāt tathatāvyatibhedataḥ

gotrataś ca sadā, sarve buddhagarbhāḥ śarīriṇaḥ ‖

It is no doubt partly on the basis of 1.270 that Bu·ston distinguished between the gotra as the causal level and the tathāgatagarbha as the resultant level. Moreover, according to some authorities, the term buddha-gotra (: bauddha-gotra) refers to the samudānītagotra rather than to the prakṛtistha, and this would also place it on the resultant level. Ruegg, V., Le traité du tathāgatagarbha de Bu ston, pp. 33–4, note.Google Scholar

⁴⁰ trividhabuddhakāyotpattigotraavabhāvārtham adhikṛtya tathāgatadhātwr eṣāṃ garbhaḥ sarvasattvānām (saṅs rgyas kyi sku rnam pa gaum bskyed pa'i rigs yod pa'i dbaṅ du byas te/ de nžin gšegs pa'i khams sems can 'di dag thams cad kyi sñiṅ por batan pa yin no//). On the basis of the reading of MS B, gotrasadbhāvārtham (corresponding to the Tibetan translation), Schmithausen proposes emending Johnston's text (v. WZKS, XV, 1971, 157)Google Scholar, as well as the reading gotra-sambhavārthena in RGVV 1.27–8 (p. 26, 1. 9)(?). (RGVV 1.144, however, reads gotrasvabhāva [p. 70,1. 1] parallel to dharmakāyasvabhāva and tathatāsvabhāva, as does RGVV 1.149–52 [p. 73, 1. 10]. Svabhāva is also found in RGV 1.144.)

⁴¹ RGVV 1.144, 149–52.

⁴² RGVV 1.149–52 (on garbha = dhātu = hetu of the tathāgata, the state of tathāgata being the state constituted or informed by the three Buddha-Bodies: trividhabuddhakāyaprabhāvitatvaṃ hi tathāgatatvam).

⁴³ Bead probably na tad yataḥ (see above, n. 6).

⁴⁴ RGVV 1.41 (p. 37, 1. 3) reads prakṛtiviśuddhagotra in Johnston's edition, but the MS reads prakṛtiviśuddhigotra according to Schmithausen (WZKS, xv, 1971, 146); the Tibetan translation has raṅ bžin gyis rnam par dag pa'i rigs. Cf. above, n. 29.

⁴⁵ Tib. rnam par dag par ruṅ, ba ñid.

⁴⁶ Or, perhaps, to example IV, which relates not to the gotra but to the tathatā.

⁴⁷ Tib. de rigs de bžin thob pa.

⁴⁸ RGVV 1.86: tadgotrasya prakṛter acintyaprakārasamudāgamārthaḥ = de'i rigs raṅ bžin gyis bsam gyis mi khyab pa'i rnam pa thob pa'i don.

⁴⁹ This definition corresponds to the definition of the (prakṛtistha) gotra found in the Šrāvakayānist literature, and occasionally quoted in the Māhāyānist sources; v. Śrāvakabhūmi, fol. 2b (and Bodhisattvabhūmi 1.1, p. 2Google Scholar: yad bodhisattvānāṃ ṣaḍāyatanaviśeṣaḥ); cf. Yaśemitra ad Abhidharmakośa 6.58–9, p. 583: prikagjanāvasthām ārabhyêndriyabhedaḥ; Sthiramati, Madhyāntavibhāgaṭīkā 4.15d–16: indriyabhedo gotram.

⁵⁰ MSA 9.37: sarveṣām aviśiṣṭâpi tathatā śuddhim āgatā

tathāgatatvaṃ tasmāc ca tadgarbhāḥ sarvadehinaḥ ‖

‘Though without differentiation for all, when Thusness has come to purity it is tathāgata-hood. And hence all embodied beings have this [tathatā: tathāgatatva] as their embryonic essence’. The MSABh. explains: sarveṣāṃ nirviśiṣṭā tathatā tadviśuddhisvabhāvaś ca tathāgataḥ/ ataḥ sarve sattvās tathāgatagarbhā ity ucyate/. Cf. above, n. 6.

⁵¹ For the term tādi(n) see Suttanipāta 522 f., which establishes the connexion with tathatta. And for tādiso as an epithet of the Sage cf. Aṅguttaranikāya, 1, p. 150.Google Scholar On the use of words derived from demonstrative pronouns and from words meaning ‘thus’ to designate a sage, of. Ruegg, D. S., ‘Védique addh et quelques expressions parallèlesà tathāgata’, JA, CCXLIII, 2, 1955, 163–70Google Scholar; Roth, G., Śri Mahāvīra Jaina Vidyālaya suvarṇamahotsava grantha, Bombay, 1968.Google Scholar

⁵² Cf. RGVV 1.40: buddhadhātu-viśuddhigotra, and RGVV 1.2 viśuddhigotra tathāgatadhātu.

⁵³ That the bauddha-gotra of RGV 1.27 refers in particular to the samudānītagotra is at least possible in view of the fact that emphasis is placed on its being only metonymously identifiable with the phala = tathāgata; while the prakṛtisthagotra (and the tathāgatadhātu = tathāgatagarbha) is at least essentially identical with the phala (cf. above, n. 39). Compare Bu·ston's position on the status of the gotra in relation to the tathāgatagarbha mentioned above, n. 39.

The suvarṇabimba example in RGVV 1.41 may also be connected with the samudānītagotra (see above, p. 346).Google Scholar

⁵⁴ Or perhaps threefold in so far as it is the source of the three Buddha-Bodies (although these in their turn correspond to a twofold scheme dharmakāya/rūpakāya).

⁵⁵ These four factors are also mentioned in the *Anuttarāśrayasūtra (Taishō, no. 669), ch. 2–5, a text which Takasaki considers to be based on the RGV (Study, 49 f., p. 186, n. 3).

⁵⁶ pada. Cf. vajrapada in RGV 1.1–2.

⁵⁷ Other enumerations of these four components are found elsewhere in the RGV.

1.1: dhātuḥ, bodhiḥ, guṇāh, bauddhaṃ karma.

1.3: garbhaḥ, agrabodhiḥ, dharmāḥ, sarvasattvārthakṛt.

5.1: buddhadhātuḥ, buddhabodhiḥ, buddhadharmāḥ, buddhakṛtyam.

[5.7: āśrayam, tatparāvṛttiḥ (parīvṛttiḥ ?).…]

5.25: vyavadānadhātuḥ, bodhiḥ, guṇāḥ, karma.

⁵⁸ The seven vajrapadas are found in the Dhāraṇīśvararājasutra; see RGV 1.2.

⁵⁹ v. Ruegg, , Théorie, 113, 283.Google Scholar

⁶⁰ E. Obermiller has translated Tib. rigs = gotra in the RGV and RGVV as ‘Lineage’, ‘germ’ (1.27–8, 41, 86) or ‘source’ (1.24) (‘The sublime science of Maitreya’, Acta Or., IX, 2–3Google Scholar, 1931, 81 f.). E. Frauwallner has translated gotra as ‘Keim’, and Ratnagotravibhāga as ‘Erläuterung des Keimes der (drei) Juwelen’ (Die Philosophie des Buddhiamus, second ed., Berlin, 1958, 255–6Google Scholar; he has also used ‘Keim’ to translate garbha in MSA 9.37 on p. 318 of this same work). Takasaki has translated the title as ‘Analysis of the Germ of the Jewels’ (Study, 141)Google Scholar, and he regularly uses ‘Germ’ to translate gotra. A. K. Warder has pointed out that the gotra ‘is the “clan” (… 1.24, 28 etc., which might also be translated in this context as “mine” or “quarry”)’ (Indian Buddhism, Delhi, 1970, 405).Google Scholar Thus no clear distinction has been made between the uses of the term in the different sections of the RGV and commentary.

⁶¹ cf. Takasaki, , StudyGoogle Scholar, 11 f.; Sohmithausen, , WZKS, XV, 1971, 123 f.Google Scholar

⁶² Taishō, no. 1611, was translated by Ratnamati in about 511. This Chinese text consists of two parts: (1) the ‘Kārikā text’, a collection of 300 verses (including 18 not found in the Sanskrit text of the RGV), and (2) the verses of the RGV (excluding, however, some found in the Chinese ‘Karika text’) together with the prose-commentary. See. Z. Nakamura, introduction to his edition of the Sanskrit and Chinese texts of the RGV (Tokyo, 1960)Google Scholar; Takasaki, , StudyGoogle Scholar, 9 f. (The Chinese verse-text thus differs from the Tibetan verse-text handed down separately in the bsTan·'gyur.)

Although it is not possible here to go into the question of the authorship of the RGV, it may at least be noted that the division between the basic verses (as preserved in the first part of Taiehō, no. 1611), the verse-commentary and the prose-commentary would make it possible to assign the basic verses to one author, and the verse and prose-commentary to another author or other authors. Thus, Nakamura assigns (pp. xxiv–xxvi) the basic verses to *Sāramati (So-lo-mo-ti) and both the verse and prose commentaries to Vasubandhu, while Tsukinowa had earlier suggested that the basic verses are *Sāramati's, the verse commentary Maitreya's, and the prose-commentary Asaṅga's. More recently, Takasaki has suggested that the basic verses are Maitreya's (Study, 9 and 62Google Scholar) and the commentary is *Sāramati's (pp. 46, 62), thus in a way reconciling the Tibetan tradition attributing the RGV to Maitreya and the Chinese tradition on *Sāramati going back to Devaprajña, the translator of the *Dharmadhātunirviśeṣaśāstra (Taishō, nos. 1626–7) also attributed to *Sāramati, and handed down by Fa-tsang (643–712) and Yüan-ts'ê (613–96). However ingenious these attempts to interpret the tradition handed down by Fa-tsang and Yüan-ts'ê may be, the differences in the results these three scholars have reached only serve to underline the obscurity of that tradition.

⁶³ v. Takasaki, , Study, 10–19.Google Scholar

⁶⁴ Study, 18, 393.Google Scholar

⁶⁵ Study, 18.Google Scholar

⁶⁶ Takasaki also includes in his basic text RGV 1.30, which implicitly refers to the gotra.

⁶⁷ Study, 14.Google Scholar

⁶⁸ In its treatment of the gotra the Fo hsing Iun (Taishō, no. 1610) does not make use of the four aspects mentioned above (v. Takasaki, , Study, 47–8Google Scholar), and it is therefore questionable whether fo hsing (= buddha-gotra) in the title can have a meaning similar to ratnagotra in the title of the RQV.

As mentioned above, the *Anuttaraāśrayasūtra (Taishō, no. 669) does refer to the four aspects of the gotra, which are treated in its ch. 2–5 (cf. Study, 49 f.).Google Scholar

⁶⁹ Mahāpariniroāṇasūtra translated from Sanskrit, IHa-sa ed., fol. 195 a 6, and colophon, fol. 222b.

⁷⁰ The original form of the text known as the RGV has also been recently discussed briefly by L. Schmithausen (‘Philologische Bemerkungen zum Ratnagotravibhāga’, WZKS, XV, 1971, 123–30Google Scholar). Following Takasaki he has included RGV 1.23 in the second or ‘B’ category of verses as an ‘Inhaltsangabe-Vers’ (pp. 126–7).Google Scholar And he then proceeds to show that Takasaki's basic text of 27 verses can be further reduced by excluding further verses. But the resulting fragment he rightly considers to be only one of the elements composing the original RGV. He concludes that the original RGV was essentially identical with the separately preserved Chinese verse-text (p. 129).

⁷¹ cf. Ruegg, , Théorie, 13.Google Scholar

⁷² As supposed by Takasaki, , Study, 22; of. p. 18 and p. 141, n. 1.Google Scholar

⁷³ cf. RGV 1.149 and 1.86. This identification, however, would not be accepted by a commentator who holds the tathāgatagarbha to be the Fruit ('bras bu = phala) when the gotra is, according to RGV 1.27, the cause on to which the name of the Fruit is metonymously transferred (of. RGVV 1.149–52). See above, n. 39, and n. 53.

⁷⁴ These difficulties were indeed not without their consequences in the later period when the Buddhist doctors came to attempt a systematic exegesis of the gotra and tathāgatagarbha doctrines.

⁷⁵ However, if the reference in RGV 1.41 to agotra were to be taken as an indication that the author accepted such a category in his own system, the reference could only be, in the framework of the tathāgatagarbha theory as presented elsewhere in the RGV, to the acquired (samudānīta) or developed (paripuṣṭa) gotra. However, the reference seems to be to the agotra concept of other schools, which is purely hypothetical for the author of the RGV or which is to be understood as alluding to a temporary incapacity only (of. kālāntarābhiprāya of RGVV 1.41). See above, n. 6.

⁷⁶ See above.

⁷⁷ cf. in general Hōbōgirin s.v.Byō. See also Ruegg, , Théorie, 516.Google Scholar

⁷⁸ For sattvadhātu referring to the ‘constitution’ of sentient beings cf. Gaṇḍavyūha 45, pp. 450–1Google Scholar: sattvadhātucikitsābhaiṣajyasaṃyogajñāneṣu dhātutantrasaṃyogaprayogeṣu suvarṇamaṇimuktāvaiḍūryaśaṅkhaśilāpravāḍalohitakāmusāragalvakeśaraśrīgarbhāśmagarbhasarvaratnasambhavotpattigotrākaramūlyajñāneṣu.…

^In addition, the concept of the tathāgatadhātu seems at a certain stage of its history to have also been closely linked with the ideas associated with the precious relic-deposit (dhātu) in the stūpa, which of course is not necessarily a mineral.

⁷⁹ Also: ‘containing the tafhāgata’. This meaning appears also of course when it means ‘born of the Tathāgata’, as it seems to do in the passage of the Gaṇḍavyūha where it is parallel to dharmwājakulodita and serves to describe Sudhana as a Bodhisattva who, by definition, is a jinātmaja ‘son of the Victor’ and an aurasaḥ putraḥ ‘bodily’ son of the Buddha. See above, p. 342.

⁸⁰ v. RGVV 1.146–7 (dharmakāya, in which case the tatpuruṣa interpretation is given); 1.148 (tathatā, in which case the bahuvrīhi interpretation is given); 1.149–52 (gotra; for the interpreta tion given in the RGVV see above, n. 40).

⁸¹ Edgerton, F., Buddhist Hybrid Sanskrit dictionary, New Haven, 1953, s.v. gotra.Google Scholar

⁸² See above, n. 15, and n. 30.

⁸³ See above, pp. 341–2.

⁸⁴ Laṅkāvatārasūtra, p. 17Google Scholar: samudramalayaśikhare … nānāratnagotrapuṣpapratimaṇḍite; p. 269, 1. 12 (‘probably’): vijñaptigotrasaṃcchannam.Google Scholar

⁸⁵ See above, n. 15.

⁸⁶ As the Chinese equivalent of gotra in Buddhist usage P. Demiéville considers that hsing ‘nature’ is a faulty though current variant of haing ‘family, clan’ etc.; v. Le concile de Lhasa, I, Paris, 1952, p. 63, n. 4.Google Scholar

⁸⁷ See above, pp. 341–2.

⁸⁸ In his review of Edgerton's dictionary V. Raghavan has maintained that the meaning ‘mine’ is merely an extension of ‘mountain’: ‘In classical Skt. gotra is well-known as mountain’ (Indian Linguistics, XVI, 1955, 322Google Scholar). However, the meaning ‘mountain’ is likely to be only a secondary development in Sanskrit based on the Vedic word gotrabhid. Cf. Renou, L., JA, CCXXXI, juillet-septembre 1939, 358–9.Google Scholar For this reason Raghavan's explanation is in itself unlikely, and it seems to be completely excluded by the other Indo-Iranian evidence discussed below.

⁸⁹ cf. Hübschmann, H., Persische Studien, p. 96, n. 948Google Scholar, quoted by Benveniste, , OLZ, LV, 1–2, 1960, col. 7.Google Scholar

⁹⁰ cf. Renou, , JA, CCXXXI, juillet-septembre 1939, 353 f.Google Scholar, and Études védiques et pāṇinéennes, I, Paris, 1955, 10Google Scholar; Gonda, J., Old Indian (Handbuch der Orientalistik, Abt. II, Bd. I, Absohn. 1), Leiden, 1971, 163.Google Scholar

⁹¹ OLZ, LV, 1–2, 1960, col. 7 f.Google Scholar: ‘Déjà Hübsohmann, , Persische Stud. p. 96Google Scholar n. 948, jugeait “bedenklich” le rapport entre skr. gotrá- et pers. gōhar…. En moyen-perse gōhr signifie “substance fondamentale, essence propre”; c'est par spécialisation qu'il désigne aussi le “métal”, la “pierre précieuse”, comme des variétés ou des espèces de la “matière par excellence”. Il peut aussi s'appliquer à la “nature fondamentale” du caractère humain. En tout cas, nous n'atteignons dans gōhr qu'un sens abstrait; et rien, ici non plus, n'établit un lien avec gav-. En principe, la forme gōhr peut remonter à *gauθra, mais aussi bien à *gavaθra (cf. Bailey, , Zoroastrian problems, p. 83Google Scholar [but see JRAS, 1953, 3–4, p. 115, n. 1]Google Scholar). S'il faut l'analyser en iranien même, on la prendra oomme dérivé nominal en -θra- de la raoine gav- “procurer”, sans le moindre rapport avec gav- “boeuf”. Que reste-t-il alors de l'équation skr. gotrá- = mp. gōhr? A notre avis deux possibilités sont à envisager. Ou skr. gotrá- “famille” est identique à gotrá- “étable” (ce qui n'est pas démontrÉ), en ce cas mp. gōhr n'a plus rien à fake ici; ou bien mp. gōhr “matière” correspond à gotrá- au sens de “essence constitutive, caractère fondamental”; alors skr. gotrá- “étable” est un mot différent, qui restera d'ailleurs difficile à analyser’.

⁹² cf. Renou, , ‘Eléments védiques dans le Sanskrit classique’, JA, CCXXXI, juillet-septembre 1939, p. 359Google Scholar, n. 1: ‘Quant à gotra “famille, nom de famille” depuis le véd. récent (le sens est-il déjà postulé par gotrabh t-, déformation de gotrabhíd- dans un mantra de MS, Edgerton, Studies Collitz, p. 34 ?), o'est la notion sooiale du lieu de réunion de la famille, englobant l'étable ou le parc, notion qu'on retrouve sous un aspect plus général dans goṣṭhī “conversation” (aussi “connexions familiales ou oollégiales” MhBh), en regard de véd. goṣṭha “étable”’.

⁹³ For these semantic and lexicographical concepts, cf. Schlerath, B., ZDMG, CXXII, 1972, 409–11.Google Scholar

⁹⁴ Edgerton, F., Buddhist Hybrid Sanskrit grammar, New Haven, 1953Google Scholar, §1.57: ‘In principle, I have excluded from my grammar and dictionary all forms which are standard Sanskrit, and all words which are used in standard Sanskrit with the same meanings’; cf. §1.53: ‘The hall-mark which distinguishes it [the BHS tradition] is the vocabulary…’. On the Middle Indian basis of ‘BHS’, see §1.4.

I wish to thank Professor M. Dresden for assistance with some of the Iranian materials. The responsibility for the interpretations offered above is, of course, mine alone.