Some Remarks on Sthiramati and his Putative Authorship of the Madhyāntavibhāgaṭīkā, the *Sūtrālaṃkāravṛttibhāṣya and the Triṃśikāvijñaptibhāṣya

Jowita Kramer

University of Munich

jowita.kramer@lmu.de

Abstract

The present paper focuses on the commentaries attributed to the Indian Yogācāra scholar Sthiramati (sixth century). So far Sthiramati’s work has received far less attention from modern scholars than the texts of other Yogācāra authors like Asaṅga or Vasubandhu, possibly because of the erroneous view that as a commentator he has not been an original author in his own right. However, commentators like Sthiramati have shaped the doctrinal development of Yogācāra thought by introducing new concepts and reorganizing previous teachings to a similar extent as ‘independent’ authors. In total seventeen works are ascribed to the author Sthiramati in Indian, Tibetan, Chinese and modern Western sources. The main concern of this study is to show some aspects of the relations between three of these works, namely the Madhyāntavibhāgaṭīkā, the *Sūtrālaṃkāravṛttibhāṣya and the Triṃśikāvijñaptibhāṣya, and to question the common authorship of these texts.

Keywords

Sthiramati, Yogācāra, Indian commentaries, svabhāva, āśrayaparāvṛtti, Madhyāntavibhāgaṭīkā, *Sūtrālaṃkāravṛttibhāṣya, Triṃśikāvijñaptibhāṣya

1 Introduction

The starting point for the present study has been an edition of the Sanskrit text of Sthiramati’s Pañcaskandhakavibhāṣā (PSkV) which I published recently (Kramer 2014). In connection with the work on the PSkV I have investigated the dependence of this text on earlier, mainly Abhidharmic, material, and additionally compared Sthiramati’s commentary with two other commentaries on Vasubandhu’s Pañcaskandhaka (PSk), namely the *Pañcaskandhavivaraṇa and the

• Pañcaskandhabhāṣya (Kramer 2012, 2013a, 2013b, 2015). Besides analyzing differ©

ent commentaries on the PSk and investigating their sources, I am also working on a comparative study of other commentaries attributed to Sthiramati, that is commentaries seemingly composed by the same author but having differing root texts. The main aim of this project is to examine the texts systematically with regard to compositional techniques, the amount of explicit and unmarked quotes as well as parallels and divergences in style and contents. Through this investigation I hope to contribute to our understanding of the functions of Indian Buddhist commentaries as well as of the motives of their authors. A better knowledge of the commentarial procedures is of essential importance when one approaches the question of how to analyse style, content and compositional techniques in order to determine authorship. While I have already published a paper on the forms of intertextuality and innovation to be found in the PS and its three commentaries (Kramer 2015), the main concern of the present study is to probe into

the relations between three of the commentaries attributed to Sthiramati: the Madhyāntavibhāgaṭīkā, the *Sūtrālaṃkāravṛttibhāṣya and the Triṃśikāvijñaptibhāṣya. As already illustrated in the aforementioned paper, various forms of textual reuse, intertextuality and also of creativity and originality are found in Sthiramati’s PSkV. The latter particularly shares a great number of parallels and closely related text passages with Sthiramati’s Triṃśikāvijñaptibhāṣya. Taking into consideration the fact that most of the remaining commentaries attributed to this author do not seem to be interrelated (even in cases in which similar topics are treated in the texts), one may wonder what conclusions are to be drawn from this with regard to the authorship of the works. Is the non-usage of a textual passage from one commentary in another commentary on the same topic attributed to the same author an indication of different authorship? Or might it be more appropriate to assume that Indian commentators were following a particular set of rules or a certain ‘methodology’ when they composed their comments and that this procedure could lead to substantial differences in the writings of one and the same author? While definitive answers to these questions require a thorough investigation of a great number of Indian commentaries, I hope to provide some preliminary insights in the present paper.

2 Sthiramati’s works: an overview

In total, seventeen works are ascribed to the author Sthiramati in Indian, Tibetan, Chinese or modern Western sources. Most of these texts (but not all of them) are listed and described in some detail in Cuong tu Nguyen’s study of the ninth chapter of Sthiramati’s *Sūtrālaṃkāravṛttibhāṣya (1990, 23–83). The seventeen works cover a wide range of topics and the commentaries among them refer to Yogācāra as well as non-Yogācāra root texts:

1. Commentaries on Yogācāra works (ascribed to Sthiramati in Indian, Tibetan or Chinese sources):

• Pañcaskandhakavibhāṣā • Triṃśikāvijñaptibhāṣya (TrBh) • Madhyāntavibhāgaṭīkā (MVṬ) • *Sūtrālaṃkāravṛttibhāṣya (SAVBh) • Abhidharmasamuccayavyākhyā © Equinox Publishing Ltd 2017

Some Remarks on Sthiramati and his Putative Authorship 49 2. Commentaries on non-Yogācāra works (ascribed to Sthiramati in Indian, Tibetan or Chinese sources): • Abhidharmakośaṭīkā Tattvārthā • *Kāśyapaparivartaṭīkā • Dasheng zhongguan shilun 大乘中觀釋論 (commentary on the Mūlamadhyamakakārikā)

3. A commentary on a non-Yogācāra work (ascribed to Sthiramati in Braarvig 1993, cxxviii):

• Akṣayamatinirdeśaṭīkā 4. A lost commentary (ascribed to Sthiramati in Matsuda 2013, 52): • Commentary on the Pramāṇasamuccaya 5. Tantric works (ascribed to Sthiramati in Tibetan sources):1 • Rgyan dam pa sna tshogs rim par phye ba bkod pa (*Paramālaṃkāraviśvapaṭalavyūha) • Skabs ’grel bye brag rnam par bshad pa (*Prakaraṇaṭīkāviśeṣavyākhyā) • Byang chub sems kyi ljon shing (*Bodhicittadruma) • Dam tshig nyi shu rtsa brgyad pa rtsa ’grel (*Samayāṣṭaviṃśamūlavṛtti) • Dam tshig rin chen gter (*Samayaratnanidhi) • Mdo’i dka’ gcod (*Sūtrapañjikā) • Theg pa gsal byed sgron ma (*Yānoddyotanapradīpa)

It can be stated with some certainty that the tantric works attributed to Sthiramati have been composed by a different author than the philosophical commentaries. The second supposition that appears very probable is that the PSkV, the TrBh and the Abhidharmakośaṭīkā Tattvārthā are very closely related and have been composed by the same person. The interdependence of the three texts seems obvious because of the great number of common passages shared by them. Thus, for instance the whole section on the mental (caitasika) factors of the PSkV is reused in the TrBh (see Kramer 2015, 330–347). In some cases, Sthiramati presents the definitions in an abbreviated form in the TrBh, probably extracting only those phrases from his PSk commentary that appeared most relevant to him. But even though the text is not identical in these instances, it is still obvious that there is a direct relation between the comments (see also Kramer 2015, 300ff.). As for the Tattvārthā and its close similarity with the PSkV, Kazunobu Matsuda and other scholars working on the edition of the Sanskrit manuscript of the text have found a great number of parallels between the PSkV and the first chapter of the Tattvārthā.2

Moreover, the same person may have authored a no longer extant Pramāṇasamuccaya commentary, as has been recently suggested by Kazunobu Matsuda (2013, 52). According to Matsuda, Sthiramati refers the reader to 1. Cordier 1909, 138f., 151 and 158.

2. This statement is based on information received from the Japanese scholars working on the edition of the Tattvārthā and on Matsuda 2014, 16–17.

a Pramāṇasamuccaya commentary in the Tattvārthā, using the expression pramāṇasamuccaya-upanibandha. As a similar term, namely pañcaskandhaka-upanibandha, appears in the TrBh when Sthiramati makes reference to his PSkV, Matsuda suggests to read pramāṇasamuccaya-upanibandha in a parallel way, that is in the sense of a commentary on the Pramāṇasamuccaya composed by Sthiramati himself. It also appears likely that the same author compiled the Abhidharmasamuccayavyākhyā. This commentary seems to be mainly a compilation of the root text, the Abhidharmasamuccaya, and the Abhidharmasamuccayabhāṣya. While both commentaries, the Bhāṣya and the Vyākhyā, are ascribed to *Jinaputra in the Tibetan tradition, Chinese sources attribute the compilation of the Vyākhyā to Sthiramati.3 Neither of the colophons of the Sanskrit versions of the Bhāṣya and the Vyākhyā seems to mention the author.

The idea that Sthiramati was involved in the production of the Vyākhyā appears possible, insofar as the Abhidharmasamuccayabhāṣya (or -vyākhyā) plays a central role in his PSkV (see Kramer 2015, 295–298). However, it is not very likely that he has composed the Bhāṣya itself since the latter appears to be older than the PSkV. As I have tried to show in a previous paper, the Bhāṣya seems to have been resorted to by Guṇaprabha in his *Pañcaskandhavivaraṇa which with some certainty can be dated earlier than the PSkV (Kramer 2015, 309–315). Apart from this, none of the traditional sources ascribes the Bhāṣya to Sthiramati. The authorship of the remaining five works remains doubtful and is to be clarified by further investigation of their contents and style. The present study aims at contributing to this clarification by illustrating some aspects of the relations between the MVṬ, the SAVBh and the TrBh.

The authorship of the *Kāśyapaparivartaṭīkā and the Dasheng zhongguan shilun 大乘中觀釋論 is another issue that requires more detailed examination. Nguyen presents several arguments that from his point of view make it plausible that only one Sthiramati composed all but the tantric works attributed to him in traditional sources (Nguyen 1990, 23–83). The *Kāśyapaparivartaṭīkā, which is available only in its Tibetan and Chinese translations, is attributed to the author Sthiramati in its Tibetan colophon. The commentary’s obvious Yogācāra perspective and its structural similarity with the Viniścayasaṃgrahaṇī may be counted in favour of Sthiramati’s authorship. However, the fact that comments on the Kaśyapaparivartasūtra which are found in Sthiramati’s MVṬ differ substantially from parallel explanations in the *Kāśyapaparivartaṭīkā4 as well as the

• Kāśyapaparivartaṭīkā’s rather early translation into Chinese (between 508 and 535

by Bodhiruci)5 make the scenario of a single Sthiramati as the common author of both texts appear doubtful. The Dasheng zhongguan shilun 大乘中觀釋論 has hardly been studied so far and therefore nothing can be said about its authorship. The Akṣayamatinirdeśaṭīkā (composed by Vasubandhu according to the Tibetan tradition) has been ascribed to Sthiramati by Jens Braarvig. Braarvig’s attribution 3. See Li 2015, 275. The Sanskrit manuscript of the Vyākhyā is currently being edited by Xuezhu Li (see Li 2012, 2013, 2014, 2015 as well as Li and Kano 2015). 4. This has recently been pointed out by Jonathan Silk in an unpublished lecture delivered at the University of Hamburg (on 28 August 2015, ‘Sthiramati and the Question of the Authorship of the Commentary to the Kāśyapaparivarta [[[Ratnakūṭa]]]: A Comparison with Passages in the Madhyāntavibhāgaṭīkā Citing the Sūtra’).

5. See Silk 2009, 382.

Some Remarks on Sthiramati and his Putative Authorship 51

(1993, cxxiii and cxxviii) is based on some passages in the Akṣayamatinirdeśaṭīkā which are similar to passages found in the SAVBh as well as in the PSkV and the TrBh. All the passages Braarvig cites with regard to the latter consider the definitions of various mental factors, which indeed the three commentaries seem to share. Whether this fact confirms a common authorship of the Akṣayamatinirdeśaṭīkā and the two other commentaries remains uncertain (as the possibility cannot be excluded that the author of the former simply copied the definitions from the latter). As for the relation of the Akṣayamatinirdeśaṭīkā to the SAVBh, it should be noted that both commentaries quote a great number of Mahāyāna sūtras.

3 Comparison of the commentaries

3.1 Introductory remarks

Summarizing some of my observations regarding the authorship of the PSkV, the TrBh and the SAVBh the following can be stated.6 The person who produced the PSkV and the TrBh gives the impression of a well–educated and innovative scholar. He usually provides philosophically complex explanations of high quality and not merely definitions of single terms. His comments are not repetitive, but concise, and lack the ‘ornamental’ use of redundant sentences and phrases. It is also notable that he only rarely gives specific examples for his statements or the terms and concepts he is commenting on.

In contrast, the author of the SAVBh offers a great number of short examples. Moreover, his comments are philosophically less independent or elaborate and often much shorter than the comments in the TrBh and the PSkV. The SAVBh offers a rather strict word by word commentary, explaining almost every single word (or phrase) of the root text and often also of Vasubandhu’s Bhāṣya. This results in the fact that some of the comments are repeated as the author first comments on the root text and then on the same phrase in Vasubandhu’s text. Thus, in general, the author of the SAVBh appears as a diligent or ‘dutiful’ but less inspired commentator. It was probably this commentarial style that led Paul Griffiths to characterize Sthiramati as not ‘an especially original or exciting thinker’ (Griffiths 1990, 46). Griffiths must be referring here to the impression he got from reading the SAVBh, since, as mentioned above, the PSkV and the TrBh definitely contain original and philosophically sophisticated thought.

Another difference between the three commentaries is to be found in the fact that the author of the PSkV and the TrBh only rarely mentions the titles of his sources and very often does not indicate at all that he is quoting a passage. In contrast, the SAVBh cites other works frequently and most quotations are ascribed to a specific work. The cited works include the Mahāyānasaṃgraha, the Madhyāntavibhāga and the Triṃśikā (Tr) as well as a great number of Mahāyāna sūtras, e.g. the Akṣayamatinirdeśasūtra, the Buddhabhūmisūtra, the Brahmaparipṛcchāsūtra, the Daśabhūmikasūtra, the Gayāśīrṣasūtra, the 6. One of the main difficulties when comparing the SAVBh with the other commentaries results from the fact that the SAVBh is available only in a Tibetan translation and that this translation was not produced by the famous team of Ye shes sde, Jinamitra etc., but by a ‘Municandra’ and ‘Bkra shis’ and thus the text is partly presented in an unusual style with regard to grammar and vocabulary.

Laṅkāvatārasūtra and the Vimalakīrtinirdeśasūtra. None of these sūtras is mentioned in the PSkV and the TrBh. The few works that are quoted explicitly in the PSkV and the TrBh are the Saṃdhinirmocanasūtra, the Abhidharmasūtra, the Abhidharmasamuccaya and the Pratītyasamutpādavyākhyā. The TrBh seems to contain quotes from the Madhyāntavibhāga, too, but it does not mention the title explicitly (TrBh 124,2). In contrast to the SAVBh, the PSkV tends to cite passages from early sūtras (however, without specific titles: see Kramer 2015, 295). As for the MVṬ, my first impression is that it neither shows an obvious affiliation with the PSkV and the TrBh nor with the SAVBh. However, this statement is based on the comparison of only a limited number of passages, most of which are presented below. A thorough investigation of further passages is required to come to a final conclusion.

In the following the three commentaries under discussion are compared on the basis of their explanations of two important concepts. In the first part of the comparison passages concerned with the three natures (svabhāva), as found in TrBh 122,5–124,20, MSABh 64,13–65,13, SAVBh 102,14–108,4 and MVṬ 110,7– 113,22, are presented. The second part provides a comparative examination of passages dealing with the process of liberation as discussed in TrBh 137,7–140,11, MSABh 66,3–7 and 66,18–67,15 as well as SAVBh 115,23–119,2.

3.2 First example: three svabhāvas

The explanation of the first svabhāva, the ‘conceptualized nature’ (parikalpitasvabhāva), in the root text of the TrBh, the Tr, starts as follows: yena yena vikalpena yad yad vastu vikalpyate, parikalpita evāsau svabhāvo na sa vidyate (‘Through whichever thought, whichever thing is conceptualized, this is only the parikalpitasvabhāva. This does not exist.’)

Sthiramati7 paraphrases yena yena vikalpena as ādhyātmikabāhyavikalpyavastubhedena vikalpānām ānantyam (‘the indefinite number of thoughts [which arise] because of the various internal and external things that can be conceptualized’). The following phrase of the root text, parikalpita evāsau svabhāvo (‘this is only the parikalpitasvabhāva’), Sthiramati explains as na hetupratyayapratibaddhasvabhāvam (‘[this means that] it is not a nature based on causes and conditions’). Then Sthiramati adds that while it is possible to have various contradicting thoughts with regard to one and the same thing (or its non-existence), it is not appropriate to assume that one and the same thing has various contradicting svabhāvas.8 Finally, Sthiramati concludes: ‘Therefore all this is thought only because its object is of a conceptualized nature’ (tasmāt sarvam idaṃ vikalpamātram eva tadarthasya parikalpitarūpatvāt).

The definition of the ‘conceptualized characteristic’ (parikalpitalakṣaṇa) provided in verse 38 of the Mahāyānasūtrālaṃkāra (MSA) differs substantially from that in the Tr. The conceptualized characteristic is subdivided into three different aspects in the text (MSABh 64,13f): 7. Here and in the following I refer, for the sake of convenience, to the author(s) of the

• Sūtrālaṃkāravṛttibhāṣya, the Madhyāntavibhāgaṭīkā and the Triṃśikāvijñaptibhāṣya simply as

‘Sthiramati’, although, as mentioned below, I question the common authorship of these texts. 8. TrBh 122,14–16: tathā hy ekasmin vastuni tadabhāve ca parasparaviruddhānekavikalpapravṛttir dṛṣṭā / na ca tad ekaṃ vasu tadabhāvo vā parasparaviruddhānekasvabhāvo yujyate. © Equinox Publishing Ltd 2017

1. yathājalpārthasaṃjñāyā nimittam (‘the basis of the ideation of the object corresponding to the expression’)

2. tasya vāsanā (‘its imprint’) 3. arthavikhyānam (‘appearance of the object’)

In verse 39 of the MSA, parikalpitalakṣaṇa is moreover explained as ‘the basis (nimitta) of the conceptualization of the unreal (asatkalpa), the appearance of the object and the name, corresponding to the name and the object’.9

In his comments the author of the SAVBh explains the first aspect, that is yathājalpārthasaṃjñāyā nimittam, as follows:10

‘Expression’ is saying ‘pot’ or ‘cloth’. ‘Object’ is ‘pot’ or ‘cloth’. The corresponding ideation is a mental factor, the thought ‘this is a pot, not a cloth’ or ‘this is blue, not yellow’. The basis is the object of the ideation – the pot, the cloth, yellow, blue etc. – the object based on which ideation appears. Those who are familiar with the conventional usage [of language] in relation to the objects think ‘this is a pot, this is a cloth’ with regard to the pot, the cloth etc.

With regard to the second aspect of parikalpitalakṣaṇa, the vāsanās, Sthiramati states that the imprints are also (an aspect of) the imagined characteristic because they provide the basis for the arising of the ideation of the object corresponding to the expression if the vāsanā immersed in the ālayavijñāna has not been removed.11

As for the third aspect of parikalpitalakṣaṇa Sthiramati explains that the object is also conceptualized if a person does not know the name of the perceived object:12

Those fools who are not familiar with the conventions do not know that it is a pot with regard to [an object] that appears like a pot in the mind, the ideation of the object […] not existing in them, nonetheless the thought arises ‘What is this?’ when they see a pot.

Finally, verse 39 of the MSA is paraphrased in the following manner by Sthiramati:13

9. MSABh 64,21f.: yathānāmārtham arthasya nāmnaḥ prakhyānatā ca yā / asatkalpanimittaṃ hi parikalpitalakṣaṇam (asaṃkalpa corrected to asatkalpa following Nagao 2007, 91). 10. SAVBh 102,19–25: de bum pa dang snam bu zhes brjod pa ni smra ba’o // bum pa dang snam bu ni don to // ji bzhin gyi ’du shes ni sems las byung ba’i chos te / ’di ni bum pa nyid yin gyi ’di ni snam bu ma yin pa dang / ’di ni sngon po nyid yin gyi ser po ma yin zhes rtog pa’o // mtshan ma ni ’du shes pa’i yul du gyur pa bum pa dang snam bu dang ser po dang sngon po la sogs pa ste / yul gang la brten nas ’du shes ’jug pa’o // don du na tha snyad la mkhas pa rnams bum pa dang snam bu la sogs pa la ’di ni bum pa ’di ni snam bu zhes rtog pa’o.

11. SAVBh 102,29f.: de’i bag chags kun gzhi la rlan pa yang na ma zhig na smra don ji bzhin gyi ’du shes ’byung ba’i rgyu byed pas na bag chags la yang kun brtags kyi mtshan nyid ces bya ste. 12. SAVBh 103,4–7: … tha snyad la mi mkhas pa byis pa rnams la […] don gyi ’du shes med par yang sems las bum pa lta bur snang la / bum pa nyid yin par ni mi shes kyi bum pa mthong na ni ’di ci zhig yin no snyam par rtog pa ’byung na yang …. 13. SAVBh 104,21–27: bum pa dang snam bu la sogs pa’i ming ji ltar brjod pa bzhin du bum pa dang snam bu la sogs pa’i don kyang de bzhin du yod do zhes rtog pa ni kun brtags kyi mtshan nyid do … bum pa dang snam bu la sogs pa’i don ji ltar gzhag pa bzhin du bum pa dang snam bu’i ming yang yod do zhes rtog pa ni / … ’di yang kun brtags kyi mtshan nyid do … nam ming gi sgo nas don yod par rtog pa na ni ming yang dag pa ma yin pa’i kun tur tog pa’i rgyu yin no / nam don gyi sgo nas ming yod par rtog pa

Thinking that objects like a pot or a cloth exist according to how the names ‘pot’ or ‘cloth’ are expressed is the parikalpitalakṣaṇa. … Thinking that names like ‘pot’ or ‘cloth’ exist according to how objects like a pot or a cloth are established … is also parikalpitalakṣaṇa. … If by means of the name the object is thought of as existent, the name is the basis of the conceptualization of the unreal. If by means of the object the name is thought of as existent, the object is the basis of the conceptualization of the unreal. In this way thinking of the name as existent and thinking of the object as existent are parikalpitalakṣaṇa.

Coming back to the Tr, the second svabhāva, the ‘dependent nature’ (paratantrasvabhāva), is characterized as ‘the thought that has arisen from a cause’ (vikalpaḥ pratyayodbhavaḥ; TrBh 122,21). In his commentary Sthiramati explains paratantra as follows:14 [[[Paratantra]] is used in] the sense that [something] arises in dependence, meaning that it is produced by other causes and conditions. It comes into existence through other causes and conditions than itself. The author of the root text of the Mahāyānasūtrālaṃkāra, again, takes a very different perspective on the three svabhāva/lakṣaṇa concept when defining the paratantralakṣaṇa: ‘[[[Paratantra]]] is of a threefold threefold appearance, being characterized by the perceived and the perceiver.’15 In Vasubandhu’s commentary these six (i.e. ‘threefold threefold’) categories are explained as ‘foundation’ (pada), ‘object’ (artha), ‘body’ (deha), ‘afflicted notion [of “I”]’ (kliṣṭamanas), ‘[[[five sense]]] perceptions’ (udgraha) and ‘thought’ (vikalpa), i.e. ‘mental perception’ (manovijñāna). While the first three are said to constitute the object (grāhya), the last three are aspects of the subject (grāhaka). Vasubandhu concludes his explanations with the statement that this conceptualization of the unreal (abhūtaparikalpa) is the dependent characteristic (MSABh 65,1–5).

In his commentary Sthiramati offers some additional information on the six categories by categorizing pada as the ‘store mind’ (ālayavijñāna) that appears like the ‘inanimate world’ (bhājana), the ‘great earth’ (mahāpṛthivī), artha as the six objects, visible matter (rūpa) up to the ‘factors’ (dharma) and deha as the six faculties (indriya), obviously understanding manas, referred to as mana-indriya, as one of them. Notably, Vasubandhu’s last statement regarding the abhūtaparikalpa as the dependent characteristic is explained by Sthiramati as referring to the ālayavijñāna and its accompanying factors (’khor ba dang bcas pa; SAVBh 106,5–24). As for the ‘perfect nature’ (pariniṣpannasvabhāva), it is defined in the following way in the Tr:16

The perfect [[[nature]]] is the [paratantrasvabhāva’s] being permanently free of the first [the parikalpitasvabhāva]. Therefore it is neither different nor not different from the paratantra[[[svabhāva]]]. [It] is to be explained as in the case of impermanence etc. When this is not seen, that is not seen. ni don yang dag pa ma yin par kun tu rtog pa’i rgyu yin te / de ltar ming yod par rtog pa dang / don yod par rtog pa ni kun brtags pa’i mtshan nyid yin no. 14. TrBh 124,3f.: parair hetupratyayais tantryata iti paratantra utpādyata ity arthaḥ / svato ’nyahetupratyayapratibaddhātmalābha iti.

15. MSABh 64,27f.: trividhatrividhābhāso grāhyagrāhakalakṣaṇaḥ. 16. TrBh 124,7 and 12, 125,1 and 8: niṣpannas tasya pūrveṇa sadārahitatā tu yā // ata eva sa naivānyo nānanyaḥ paratantrataḥ // anityatādivad vācyo nādṛṣṭe ’smin sa dṛśyate //

Some Remarks on Sthiramati and his Putative Authorship 55 In his comments Sthiramati provides the following explanations (only the parts relevant for the present comparison are quoted):17 It is perfect because it is perfection without change … [If it were different from paratantrasvabhāva], paratantra would not be free from the conceptualized … [If it were not different], it would not be the object of purification because it would have contamination as its nature in the same way as paratantra.

The main aspects of pariniṣpannalakṣaṇa presented in the MSA are ‘non-existence’ (abhāvatā), ‘existence’ (bhāvatā), ‘sameness of existence and non-existence’ (bhāvābhāvasamānatā), ‘non-pure and pure’ (aśāntaśānta) and ‘non-conceptualized’ (akalpa) (MSABh 65,6f.). Although the root text mentions ‘non-existence’ and ‘existence’ in the context of pariniṣpannalakṣaṇa, none of the commentators relates these two aspects to the idea (advocated in the Tr commentary) that the perfect nature is the dependent nature without the superimposition of the parikalpitasvabhāva/–lakṣaṇa. Instead abhāvatā is understood as the non-existence of ‘of all conceptualized factors’ (sarvadharmāṇāṃ parikalpitānām) and bhāvatā as existence ‘through the non-existence of these [factors]’ (tadabhāvatvena) by Vasubandhu. The concept of bhāvābhāvasamānatā is explained as ‘[[[sameness]]] because existence and non-existence are indivisible’ (tayor bhāvābhāvayor abhinnatvāt), aśāntaśānta as ‘impure because of the incidental contaminations, pure because of natural pureness’ (aśāntā cāgantukair upakleśaiḥ śāntā prakṛtipariśuddhatvāt), and akalpa as ‘non-conceptualized because it is not the object of thinking due to [its] non-proliferation’ (avikalpā vikalpāgocaratvāt niṣprapañcatayā; MSABh 65,8–11).

In the SAVBh Sthiramati provides some additional comments on the terms mentioned in the root text. The term pariniṣpanna, according to him, is used with regard to the characteristic (lakṣaṇa) of the ‘ultimate reality’ (dharmadhātu), the ‘emptiness’ (śūnyatā) and the ‘true reality’ (tathatā; in Vasubandhu’s commentary (MSABh 65,8) only the tathatā is mentioned as constituting the pariniṣpannalakṣaṇa). Non-existence (abhāvatā) refers to the non-existence of grāhya and grāhaka and bhāvatā is the existence of śūnyatā that is free of grāhya and grāhaka. The concept bhāvābhāvasamānatā concerns the non-existence of grāhya and grāhaka in the existence of emptiness on the one hand and the existence of emptiness in the non-existence of grāhya and grāhaka on the other. With regard to śānta Sthiramati adds that it refers to the fact that pariniṣpanna has inherently light and pureness as its nature (rang bzhin gyis ’od gsal zhing dag pa’i rang bzhin yin pas; SAVBh 106, 27–107,14).

Taking finally a brief look at Sthiramati’s comments on the three lakṣaṇas in the MVṬ, we are confronted with yet another alternative way of explaining the relations between the three natures. The definition of the lakṣaṇas in the root text, the Madhyāntavibhāga, consists of the following three phrases (MVBh 38, 1f.): 1. asac ca nityam (‘[The parikalpitalakṣaṇa] is eternally non-existent.’) 2. sac cāpi atattvataḥ (‘[The paratantralakṣaṇa] exists, but not as true reality.’) 3. sadasat tattvataś ca (‘[The pariniṣpannalakṣaṇa] exists, does not exist and [this] according to true reality.’) 17. TrBh 124,8–20: avikārapariniṣpattyā sa pariniṣpannaḥ … na parikalpitena paratantraḥ śūnyaḥ syāt … na viśuddhālambanaḥ syāt paratantravat saṃkleśātmakatvāt.

Before commenting on each of these three explanations, the commentator Sthiramati offers a number of general comments on the three lakṣaṇas, none of which is implied in the root text itself. In this context he mentions various alternative ways of understanding the lakṣaṇas propagated by other scholars. The following list illustrates the five ways of classifying the lakṣaṇas provided by Sthiramati (MVṬ 112,1–14):

1. a. vyavahāra (‘convention’) b. paramārtha (‘highest reality’) c. tadāśraya (‘its basis’) 2. a. viparyāsa- (‘[[[object]] of] error’) b. tannimitta- (‘[[[object]] of] its cause’) c. tatpratipakṣālambana (‘object of its antidote’) 3. a. prahāṇa- (‘abandoning’) b. parijñā- (‘insight’) c. sākṣātkriyā- (‘realization’)18 4.

prakṛtigāmbhīryasya prajñāpāramitārūpasya trisvabhāvadvāreṇābhrāntipratipattyartham (‘[The three lakṣaṇas are taught] for the sake of the faultless understanding of the nature of the perfection of insight, which is naturally profound, by means of the three natures.’) Sthiramati explains this statement further by way of a quotation from the Abhidharmasūtra and subsequently provides the threefold categorisation of: 1. māyādideśanā (‘teaching of a magical creation etc.’) as paratantrasvabhāva 2. nāstideśanā (‘teaching of non-existence’) as parikalpitasvabhāva 3. pariniṣpannadeśanā (‘teaching of the perfect’) as the four purifications (viśuddhi)19 5.

a. laukika- (‘[[[object]] of] the mundane’) b. lokottara- (‘[[[object]] of] the supramundane’) c. tatpṛṣthalabdhajñānaviṣaya (‘object of the subsequently attained knowledge’) Later in his commentary Sthiramati also comments directly on the threefold definition of the lakṣaṇas given in the root text (MVṬ 113,3–22). These explanations may be summarised as follows:20 18. The whole sentence reads bodhisattvānām āvaraṇavisaṃyogārthaṃ prahāṇaparijñāsākṣātkriyāvastupradarśanārtham (‘[The three lakṣaṇas are taught] in order to demonstrate the basis of abandoning, insight and realization for the sake of the bodhisattvas’ separation from the hindrances.’). 19. This passage appears only in the Tibetan translation of the text. For further details of the four viśuddhis, see MVṬ 112, n. 3.

20. Passages missing in the Sanskrit manuscript (and reconstructed by Yamaguchi) are provided © Equinox Publishing Ltd 2017 Some Remarks on Sthiramati and his Putative Authorship 57 1. parikalpitalakṣaṇa a. to which the fools (bāla) have attachment (abhiniveśa) of grāhya and grāhaka and vācya (‘the one to be expressed’) and vācaka (‘the one who expresses’)

b. appears like in a dream (svapna iva pravartate) c. ‘It is a conceptualized nature because it does not have an own nature.’ (sa svarūpābhāvāt kalpitasvabhāvaḥ) d. ‘Since it has the non-existent as its self and is without an error the parikalpitalakṣaṇam is non-existence.’ (de’i phyir med pa’i bdag nyid yin pa dang / ’khrul pa med pa’i phyir med pa nyid kun brtags pa’i mtshan nyid do) e. ‘Only non-existence is said to be [its] true reality.’ (asattvam eva tattvam ucyate)

2. paratantralakṣaṇa

a. ‘It exists because it is the basis for conventional conceptualization of grāhya and grāhaka etc.’ (gzung ba dang ’dzin pa la sogs pa brtags pa’i tha snyad kyi gnas yin pas yod do)

b. ‘It does not really exist as such because it does not exist having the self of grāhya and grāhaka etc.’ (gzung ba dang ’dzin pa la sogs pa’i bdag nyid du med pas, na tathā tattvataḥ)21

c. ‘What appears, like the basis of an illusion, that exists. But in the way as it appears, in the form of grāhya and grāhaka, it does not [[[exist]]], like [in the case of] a magical person etc. Therefore it is taught to be a fault.’ (yat khyāti māyopādānavat tad asti / ji ltar gzung ba dang ‘dzin pa’i rnam par snang ba de ni de lta ma yin te / sgyu ma’i skyes bu la sogs pa bzhin no // de’i phyir ‘khrul pa zhes bstan par ‘gyur ro)

d. ‘It is the appearance of something that exists differently in the form of something that does not exist intrinsically.’ (anyathā vidyamānasya svātmany avidyamānenākāreṇa prakhyānam)

3. pariniṣpannalakṣaṇa a. ‘It exists because it has the nature of the existence of non-existence of the two [[[grāhya]] and grāhaka].’ (gnyis po med pa yod pa’i bdag nyid yin pas yod pas so)

b. ‘It does not exist because it has the nature of non-existence of the two.’ (dvayābhāvātmakatvād asattvam) c. ‘It is reality because it is the object of purification.’ (viśuddhyālambanatvāt tattvam)

If we assume that the same person authored all the commentaries under discussion, it seems difficult to find a plausible explanation for the fact that none of the definitions mentioned in the TrBh and the SAVBh are included in the extensive collection of alternative interpretations of the three lakṣaṇas provided in the MVṬ. As already noted above, it is also remarkable that Sthiramati does not in their Tibetan translation (see Peking 5534, 84b5–85a1). 21. The first part of this sentence is preserved only in Tibetan.

explain the relationship between paratantra– and pariniṣpannasvabhāva/–lakṣaṇa in the SAVBh and the MVṬ in a similar way as in the TrBh. In the SAVBh Sthiramati characterises non-existence (abhāva) with regard to pariniṣpannasvabhāva as the non-existence of the object (grāhya) and subject (grāhaka) and existence (bhāva) as the existence of emptiness (śūnyatā) that is free of grāhya and grāhaka. In the MVṬ the two parallel terms sat and asat are interpreted as dvayābhāvabhāvātmakatva– (‘having the nature of the existence of non-existence of the two’) and dvayābhāvātmakatva– (‘having the nature of nonexistence of the two’) respectively. Although the ideas expressed in both texts are very similar, the wording does not seem to be related (whereas it is almost identical in similar cases in Sthiramati’s Tr and PSk commentaries). As for parikalpitalakṣaṇa/-svabhāva, it is characterized as not existent in both, the MSA (parikalpitalakṣaṇaṃ nityam asat) and the Tr (na sa vidyate). Nonetheless Sthiramati’s comments on these phrases differ from each other, as is visible from the comparison provided above.

Another striking feature of the MVṬ is the frequent usage of terms like ‘others’ (anye), ‘some’ (eke) and ‘another’ (aparaḥ) in order to introduce alternative interpretations of a concept. In contrast, the other commentaries under discussion contain only few references to ‘other’ views and only very rarely use specific terms to introduce them. If the commentaries were composed by the same author, why did he not provide a comparable amount of alternative views in all his commentaries?

3.3 Second example: Liberation

The second example for the different style and contents of the commentaries under discussion concerns the explanation of the process of liberation. The latter is discussed in a number of verses scattered over chapters nine and eleven of the MSA and in three verses of the Tr. Below I present an exemplary comparison of some of Sthiramati’s comments on verses 47–49 of the MSA and on verses 27–29 of the Tr. A study of all the passages related to liberation in the MSA goes beyond the scope of this paper and would hardly add any further insights to the topic under discussion (as the amount of parallels and differences is similar in the passages not treated here).

According to the MSA the mind is associated with ‘badness’ (dauṣṭhulya), caused by the false view of the self (ātmadarśana), before liberation (MSABh 67, 10f.). In the Tr the description of the process of liberation is generally less complex than in the MSA, but the sequence of steps is similar. Verse 29 also mentions dauṣṭhulya, which is of two kinds and which is removed at a certain stage of the liberation process (TrBh 138,7).

Notably the author(s) of the SAVBh and the TrBh follow(s) differing lines of thought when explaining dauṣṭhulya. While in the SAVBh dauṣṭhulya is subdivided into that of the body and that of the mind (citta),22 in the TrBh it is mainly explained as the inflexibility (akarmaṇyatā) of the basis (āśraya), being the seed (bīja) of ‘the hindrances [consisting in] the contaminations’ (kleśāvaraṇa) and ‘the hindrances of the objects of knowledge’ (jñeyāvaraṇa; TrBh 140,10f.). In the SAVBh 22. SAVBh 118,13f. SAVBh 118,14–16 specifies the bodily dauṣṭhulya as killing, stealing etc. and the mental as desire, ill will or the conceptualization of the non–existent.

dauṣṭhulya is specified as consisting of the imprints (vāsanā) of the two kinds of hindrances (SAVBh 118,16f.).

The process of liberation is described from a number of different perspectives in the MSA. In verse 47 the practitioner is said to enter true reality (tattva) through grasping (grahaṇatas),23 having understood the twofold selflessness (nairātmya). A similar concept is expressed in verse 48 in an alternative way. There it is stated that the practitioner sees ‘name only’ (nāmamātra) having the ‘basis’ (ādhāra), which is explained as ‘learning’ (śruti) in Vasubandhu’s commentary, and the ‘product’ (ādhāna), paraphrased as ‘correct contemplation’ (yoniśomanaskāra). In the following step, when the practitioner fixes the mind (manas) in true reality,24 the perception of tattva (or, according to Vasubandhu, vijñaptimātra) and nāmamātra also ceases and liberation (mukti) is attained (MSABh, 66,20–67,4). The liberating process culminates in the ‘transformation of the basis’ (āśrayaparāvṛtti), which, according to verse 44 of the MSA, includes the transformation of the ‘support’ (pada), the ‘object’ (artha) and the ‘body’ (deha), that is the transformation of the ‘seed’ (bīja), the ālayavijñāna (MSABh 66,3–5).

In the SAVBh we find a whole class of comments on this passage, which refer to various stages of the path and have no parallel at all in the TrBh. Sthiramati explains that correct contemplation (yoniśomanaskāra) is reached at the level of the ‘stage of the practice of devotion’ (adhimukticaryābhūmi; SAVBh 117,16– 18). The ‘seeing of nāmamātra’ is achieved at the level of ‘heat’ (uṣmagata) and of ‘summit’ (mūrdhan; SAVBh 117,21), which constitute the first two stages of adhimukticaryābhūmi. At the stage of ‘endurance’ (kṣānti), which is the third stage of adhimukticaryābhūmi, the (concept of an) object is given up (SAVBh 117,23– 27). It is also on the level of kṣānti that the mind is fixed in mind only according to Sthiramati (SAVBh 116,19), while on the level of ‘supreme mundane factors’ (laukikāgradharma), which corresponds to the last level of adhimukticaryābhūmi, the practitioner does not perceive ‘mind only’ anymore (SAVBh 116,22f.). Finally, on the first bhūmi liberation (vimukti) is attained, the practitioner dwelling in objectless insight (jñāna; SAVBh 118,1–6).

The only comment related to the stages of the path to be found in the TrBh is the remark that practitioners who dwell in the vijñaptimātratā attain the result (phalasaṃpatti) by gradually reaching higher special [states] (uttaraviśeṣagati), starting with the path of seeing (darśanamārga; TrBh 138,10f.). Instead of referring to any levels of adhimukticaryābhūmi in the TrBh, Sthiramati only explains that seeing ‘this is representation only’ (vijñaptimātram evedam) means seeing that there is no external object (artha) and provides some examples for the meditation objects (ālambana) the practitioner places ‘in front [of him]’ (agratas), namely various stages of a deteriorating body (TrBh 134,18–136,4). The second step within the liberation process, that is the letting go of the concept ‘this is representation only’, is expressed in the Tr as insight (jñāna) not grasping any object and the practitioner being fixed in the vijñaptimātratā (TrBh 136,7f.).

23. Vasubandhu (MSABh 66,24f.) explains that the practitioner enters true reality through grasping ‘This is grasping only’ (grahaṇamātram etad iti). 24. A similar idea is expressed in MSABh 67,11 as tadadhyātmasthiteḥ (‘through fixing the [[[mind]]] within’), which is paraphrased in Vasubandhu’s commentary (MSABh 67,14) as tasya cittasya citta evāvasthānāt (‘because of fixing this mind in the mind only’).

The TrBh comments on this verse of the Tr vary strongly from the SAVBh explanation. The statement that jñāna does not perceive objects is specified by Sthiramati as jñāna that does not perceive objects like teachings (deśanā), instructions (avavāda) and visible matter (rūpa) etc. The practitioner sees the objects as they really are (yathābhūta), not as they are by someone who is blind from birth, and is fixed in the ‘true reality of his own mind’ (svacittadharmatā; TrBh 136,9–12). Of particular interest are also Sthiramati’s comments on the transformation of pada, artha and deha in the SAVBh. Sthiramati explains that the normal appearance of pada, that is the ‘inanimate world’ (bhājanaloka), being characterized by

ravines and cliffs, is replaced by the appearance of different precious stones, like crystal and beryl (vaiḍūrya). The objects of everyday perception are replaced by objects like the ‘wish–granting tree’. Finally, with regard to deha Sthiramati states that the six faculties (visual sense up to manas) are replaced by one single indriya which is able to perceive all objects, and that 112 qualities as well as the Buddha body are obtained.25 None of these comments is to be found in the TrBh. These examples may suffice to illustrate the substantial differences between the *Sūtrālamkāravṛttibhāṣya and the TrBh. The root texts of both commentaries present the twofold structure of the path to liberation, the first step of which consists in the realization of the non-existence of objects and in having the concept that this is nāma-/vijñaptimātra, while the second stage involves the nonperception of the latter and the cessation not only of the object but also of the subject. As illustrated above, Sthiramati’s comments on these two stages differ considerably from each other. Some of the differences may certainly be explained by the varying arguments of the root texts. For example the transformation of pada, artha and deha mentioned in the MSA is not mentioned in the Tr. One may therefore argue that Sthiramati did not feel the need to refer to it in the TrBh.

This might partly explain the omissions in the respective comments. However, there are numerous examples to be found in the PSkV and the TrBh, which show that Sthiramati does not hesitate to go beyond the root text in other instances and to include concepts in his comments that are neither indicated nor in any way implied in the text he is commenting on. As already mentioned, Sthiramati also tends to quote himself (i.e. the PSkV) in the TrBh. Moreover, if we assume that the same individual has composed the comments on the MSA and the Tr, it is striking that the author explains dauṣṭhulya so differently (and uses the term vāsanā instead of bīja) and that he does not mention at all the stages of the path in his Tr commentary. One may also wonder why the idea of ‘fixing the mind in [[[true reality]]]/representation only’ (MSABh 66,20: tatra sthānān manasaḥ; MSBh 66,25: vijñaptimātrasthānān manasaḥ) or “being fixed in representation only” (TrBh 136,8: sthitaṃ vijñaptimātratve) is commented on in such a different way by Sthiramati. While in SAVBh 116,19 he explains the root text as bzod pa’i dus na phyi’i yul mi dmigs kyi sems tsam la gnas pa (‘being fixed in cittamātra, in which no external objects are perceived, at the time of kṣānti’), in TrBh 136,9–12 he states: yasmin kale … jñānaṃ bahiś cittān nopalabhate … tasmin kale … svacittadharmatānāṃ … pratiṣṭhito bhavati (‘when knowledge does not perceive any [[[object]]] outside of the mind, one becomes fixed in the true reality of one’s 25. On this transformation, see also verse 41 of the ninth chapter of the MSA.

own mind’). Although the arguments presented by Sthiramati are partly related, the wording and the style of the explanation vary substantially.

4 Conclusions

By means of the comparisons of the comments on the three svabhāvas and on the process of liberation in the MVṬ, the TrBh and the SAVBh, I hope to have illustrated that the texts under discussion differ fundamentally from each other and have hardly anything in common with regard to their contents and style. The fact that Sthiramati quotes himself extensively in the TrBh (copying passages from the PSkV) as well as the introduction of new (philosophical) concepts and explanations in the Tr and the PSk commentaries shed doubt on the scenario that he is the sole author of all the Yogācāra commentaries ascribed to him.

When trying to understand the nature of the commentaries, one has to consider that the root texts present different developmental stages or perspectives of the concepts commented on. The differences in the root texts as the starting points for the commentaries necessarily result in differences in the respective commentaries. However, not all of the variant comments found in the works under discussion can be explained in this way. Thus, we must either accept the possibility that not all the commentaries attributed to Sthiramati were composed by one and the same person, or we have to find other possible explanations for the divergences in the author’s comments. We may for instance assume that an Indian commentator was not only guided by the developmental stage of the root text (and by the explanations provided by the latter) but also by how the root text was understood in the milieu in which it was produced. In other words, the Indian commentator might have adjusted his own comments to the zeitgeist in which the respective root text was composed so that he commented differently on the same topic.

We do not yet know enough about the commentarial techniques of ancient Indian authors to come to a final conclusion on the circumstances under which the commentaries were produced. However, I find it improbable that commentators were operating under such constraints and aimed to present only a partial viewpoint. Rather, it seems likely to me that more than one individual was involved in the production of the commentaries ascribed to the author Sthiramati. Leaving aside the question of common authorship, it should be noted that the author(s) of all the commentaries under discussion appear(s) as creative and original thinker(s), enriching the root texts with a great number of additional terms and concepts not mentioned or even indicated in the texts commented on. Acknowledgements

I would like to thank Elisa Freschi, Ralf Kramer, Alexander von Rospatt and Lambert Schmithausen for very valuable comments on previous drafts of this paper. I am also grateful for the support received from the German Research Foundation (DFG), which enabled me to complete this article.

Abbreviations MSA Mahāyānasūtrālaṃkāra, edited in MSABh.

MSABh Mahāyānasūtrālaṃkārabhāṣya (Vasubandhu), edited by Sylvain Lévi, Paris 1907.

MVBh Madhyāntavibhāgabhāṣya (Vasubandhu), edited by Gadjin Nagao, Tokyo 1964. MVṬ Madhyāntavibhāgaṭīkā (Sthiramati), edited by Susumu Yamaguchi, Nagoya 1934. PSk Pañcaskandhaka (Vasubandhu), edited by X. Li and E. Steinkellner, Beijing/Vienna 2008.

PSkV Pañcaskandhakavibhāṣā (Sthiramati), edited by Jowita Kramer, Beijing/Vienna 2014. SAVBh *Sūtrālaṃkāravṛttibhāṣya (Sthiramati), edited by Osamu Hayashima, Bulletin of Faculty of Education Nagasaki University 27, 1978, pp. 37–70. Tr Triṃśikā (Vasubandhu), edited in TrBh. TrBh Triṃśikāvijñaptibhāṣya (Sthiramati), edited by Hartmut Buescher, Vienna 2007.

Bibliography

Braarvig, Jens. 1993. Akṣayamatinirdeśasūtra. Two volumes. Oslo: Solum. Cordier, Palmyr. 1909. Catalogue du fonds tibétain de la Bibliothèque Nationale. Deuxième partie: Index du Bstan-ḥgyur (tibétain 108–179). Paris: Imprimerie Nationale. Griffiths, Paul J. 1990. ‘Painting space with colors: Tathāgatagarbha in the Mahāyānasūtrālaṅkāra-corpus IX.22–37’. In Buddha Nature: A Festschrift in Honor of Minoru Kiyota, ed. Paul J. Griffiths and John P. Keenan, 41–63. Tokyo: Buddhist Books International.

Kramer, Jowita. 2012. ‘Descriptions of “feeling” (vedanā), “ideation” (saṃjñā), and “the unconditioned” (asaṃskṛta) in Vasubandhu’s Pañcaskandhaka and Sthiramati’s Pañcaskandhakavibhāṣā’. Rocznik Orientalistyczny 65(1): 120–139. ———. 2013a. ‘A study of the Saṃskāra section of Vasubandhu’s Pañcaskandhaka (with reference to its commentary by Sthiramati)’. In The Foundation for Yoga Practitioners: The Buddhist Yogācārabhūmi Treatise and Its Adaptation in India, East Asia, and Tibet, edited by Ulrich Timme Kragh, 920–970. Cambridge, MA: Harvard University Press. ———. 2013b. ‘Notes on the Rūpa section of the Pañcaskandhakavibhāṣā’. China Tibetology 21(2): 86–99.

———. 2014. Sthiramati’s Pañcaskandhakavibhāṣā, Part 1: Critical Edition. Part 2: Diplomatic Edition. Two volumes. Vienna/Beijing: Austrian Academy of Sciences/China Tibetology Publishing House.

———. 2015. ‘Innovation and the role of intertextuality in the Pañcaskandhaka and related Yogācāra works’. Journal of the International Association of Buddhist Studies 36/37: 281–352.

Li, Xuezhu. 2012. ‘Abhidharmasamuccayavyākhya no jobun ni tsuite’ Abhidharmasamuccayavyākhya の序文について (‘Sanskrit text of the Opening Section of Sthiramati’s Abhidharmasamuccayavyākhya’). Indogaku Chibettogaku Kenkyu インド学チベット学研究 (Journal of Indian and Tibetan Studies) 16: 1–7. ———. 2013. ‘Diplomatic transcription of newly available leaves from Asaṅga’s Abhidharmasamuccaya’. Annual Report of the International Research Institute for