Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? “Dependent Origination = Emptiness” —Nāgārjuna’s Innovation? An Examination of the Early and Mainstream Sectarian Textual Sources Shì Hùifēng Abstract It is claimed that one of the innovative contributions of Nāgārjuna in his Madhyamaka thought was establishing the equivalence of emptiness (P: suññatā, Skt: śūnyatā; kōng 空, kōngxìng 空性) and dependent origination (P: paṭicca­samuppāda, Skt: pratītya­samutpāda; Ch: yīnyüán 因緣, yüánqĭ 緣起). This study examines early and mainstream Buddhist textual sources to discover what relationship between emptiness and dependent origination was established before Nāgārjuna. In Part 1, we broadly outline the near paradigmatic modern Buddhist studies discourse on the teachings of emptiness. We then focus on the role of Nāgārjuna’s Madhyamaka within this discourse. Lastly, this study rounds off with a literature review of studies on emptiness and dependent origination before Nāgārjuna. Part 2 covers the early teachings found in the Pāli Nikāyas and (Chinese translations of) the Āgamas. It finds that the term emptiness was sometimes used independently to refer to the process of dependent arising as saṃsāric dissatisfaction and cause, and also as dependent cessation into nirvāṇa. Emptiness as the “profound” also described these two complementary processes as a whole. Part 3 continues with the broad range of mainstream sectarian sūtra and śāstra literature. Here, the previous relationships are made more firm and explicit. There is greater association with the two doctrines as rejection of extreme views based on a “self”. The two are also brought within the Abhidharma methodology of analysis into conventional 1 Published Page Number: 000175 JCBSSL VOL. XI or ultimate truths, and classification as conditioned or unconditioned phenomena. Part 4 concludes, that while already nascent in the early literature, the relation or equation of emptiness with dependent origination, along with related terms, was quite well developed in pre-Nāgārjunian sectarian literature, and is strongest in the Sarvāstivādin literature. We recommend that aspects of the academic discourse on emptiness should be rectified as a result of these findings. Keywords: emptiness, dependent origination, Nāgārjuna, Madhyamaka, early Buddhism, sectarian Buddhism 2 Published Page Number: 000176 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? 1. Emptiness, Dependent Origination & Nāgārjuna Modern Buddhist studies has an already established academic discourse and narrative with regard to the teaching of the “empty” or “emptiness” (P: suñña, suññatā, Skt: śūnya, śūnyatā; Ch: kōng 空 / kōngxìng 空性). Although this general position on emptiness is seldom stated as a “discourse” or “narrative”, and is moreover not a position held universally by all scholars, it is nonetheless fairly common, a default position as it were. The fact of the lack of need to express it as an academic discourse or narrative is perhaps further indication of its paradigmatic status. Its lines of development and argument often begin from middle to late Mahāyāna, and then work backwards through history showing how each stage differed from the previous. This is because it tends to (quite erroneously) argue that “emptiness” is largely a Mahāyāna doctrine. This form of teleological approach, whereby earlier forms are only investigated in as much as they are relevant for later forms, obviously leads to numerous distortions. Here, however, we shall use the more natural diachronic presentation in a natural forward historical sequence, noting some of the modern scholars who support the respective elements of the discourse: 1. It is stated that early Buddhism as found in the Nikāyas and Āgamas also did not consider the doctrine of emptiness as particularly important. 1.1. Time period is from the Buddha to the schism of the Saṅgha, approximately 5th to 3rd century bce. 1.2. More often than not, the sutta (and to a much lesser extent the vinaya) canon of the Pāli Theravādin tradition alone is used to represent the doctrines of “early” Buddhism. 1.3. What little early Buddhism so defined does say regards emptiness, is by and large merely a synonym for not self. 2. Slightly later, mainstream sectarian Buddhism, as typified by the Abhi­dharma, referred to with the polemic term “Hīnayāna”, also did not consider the teaching of emptiness as particularly important. 2.1. The alleged time period is from the schism of the Saṅgha to the dominance of the Mahāyāna, approximately 3rd century bce to 2nd century ce. 3 Published Page Number: 000177 JCBSSL VOL. XI 2.2. It is shown that the Abhidharma dharmavāda system considers phenomena as a plurality of really existent entities. Although rejecting a “self” (ātman) in the person (pudgala), known as “pudgala­nairātmya” or “pudgala­śūnyatā”, they considered the dharmas to in fact exist. 2.3. The Northern Abhidharma­kośa and Southern Visuddhi­magga are perhaps the most commonly referred to texts of this period, despite their providence of c. 5th cty, quite some centuries after the advent of the Mahāyāna. 3. In contradistinction to the above two historical stages, the notion of emptiness is an extremely important doctrine in Mahāyāna Buddhism, especially the Prajñāpāramitā. 3.1. Spans from the turn of the millennia 0 ce on, for several centuries. 3.2. As opposed to the “Hīnayāna” view espoused above, it is claimed that the Mahāyāna notion of emptiness encompasses both the “emptiness of the person” (pudgala­śūnyatā, pudgala­ nairātmya) and also the “emptiness of the phenomenal” (dharma­śūnyatā, dharma­nairātmya). 3.3. This is considered at least a direct refutation against the Abhidharma systems, if not mainstream sectarian Buddhism as a whole. 3.4. Standard explanations of the Prajñāpāramitā Sutras are often based on much later Indian commentaries which have been preserved in Sanskrit and Tibetan, particularly the Abhisamayālaṃkārāloka of Haribhadra. 4. The definitive Mahāyānic meaning of emptiness is usually sourced from the Madhyamaka texts of Nāgārjuna and his doctrinal heirs. 4.1. From the time of Nāgārjuna, 2nd to 3rd century ce. 4.2. Sometimes Nāgārjuna is given the status of a founder or inspiration for the Mahāyāna as a whole, or at least its principal systematizer. 4.3. In particular, Nāgārjuna’s seminal text, the Mūlamadhyāmaka­ kārikā, is taken as definitive, if not exclusively. 4.4. Often his expression of emptiness is considered “revolutionary” when compared to the “realist” or “substantialist” positions of Abhidhamma Buddhism, or more specifically, the teachings of 4 Published Page Number: 000178 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? the Pudgalavādins and the Abhidharma of the Sarvāstivāda. This was the “emptiness doctrine” (śūnyatā­vāda) against the “own-nature doctrine” (svabhāva­vāda). Some consider the “early” teachings to also be his target. 4.5. Later commentaries on the Mūlamadhyamaka­kārikā, especially those preserved in Sanskrit and Tibetan, such as Candrakīrti, are used to explain Nāgārjuna’s key śāstra in turn. 5. It is commonly said that Nāgārjuna received the inspiration for his formulation on emptiness from the Prajñā­pāramitā Sūtras. 5.1. The relationship between these sūtras and śāstras is sometimes explained as Nāgārjuna’s Madhyamaka emptiness being a systematic philosophical expression of the Prajñā­pāramitā Sūtras religious teaching of emptiness. It was mainly a number of earlier studies, by the likes of Murti (1955), Stcher­batsky (1968) and Conze (1962), who first established this modern “discourse of emptiness” through detailed scholastic studies focusing on the source materials available at the time. Recently, Westerhoff has provided a survey of “The Philosophical Study of Nāgārjuna in the West”, laying out three phases: “First is the Kantian phase, then the analytic phase, and finally a post-Wittgensteinian one” (Westerhoff 2009: 9ff). It is noteworthy that Stcherbatsky and Murti, for instance, fall within the Kantian phase, and that Conze has strong Hegelian credentials. The above “discourse” is an outline of some of the key points, as exemplified in major writings by a number of Buddhist studies scholars, far too many to list individually here. For now, we would like to point out that as a whole, the discourse of emptiness so established is a very useful general outline which may serve as is when laying out the broad picture of the development of Indian Buddhist thought. However, it is not without some problems. Over the last few decades, several scholars have shown more nuanced approaches and proposed amendments to various parts of the discourse, which are worth noting. 1.1. Nāgārjuna’s Madhyamaka as Emptiness and Dependent Origination We shall not here attempt to discuss the discourse on emptiness in its entirety, which would be a huge project encompassing much 5 Published Page Number: 000179 JCBSSL VOL. XI of Buddhism in India and beyond. However, we would like to draw attention to certain aspects of it, which while being commonly accepted as paradigmatic, are still nonetheless somewhat problematic. Of the five main elements of the discourse outlined above, we would like to reflect on the general position of Nāgārjuna’s Madhyamaka. In the last few years a number of excellent studies in this area have been published, such as those by Westerhoff (2009), Seyfort Ruegg (2010) and Siderits & Katsura (2013). However, focusing mainly on Nāgārjuna’s philosophy, rather than as historical development of Buddhist thought, the Madhyamaka role within the broader “discourse” is only briefly touched upon. But despite the progress in finding more appropriate ways to read and understand Nāgārjuna, in particular his Mūla­ madhyamaka­kārikā, certain approaches have remained unchanged. Candrakīrti, for example, remains the default commentator of choice, and the use of Tibetan sources far exceeds that of the Chinese sources (see “Discourse” point 4.5). More importantly for our essay here, is the continued great emphasis placed upon Kārikā verse 24:18: yāḥ pratītyasamutpādaḥ śūnyatāṃ tāṃ pracakṣmahe| ­ sā prajñaptir upādāya pratipat saiva madhyamā|| That dependent origination, we declare it is emptiness; It is designation on a basis, it is indeed the middle way. With emptiness already the accepted core of Nāgārjuna’s Madhyamaka thought, this verse now directly associates emptiness with “dependent origination” (P: paṭicca­samuppāda, Skt: pratītya­samutpāda; Ch: yīnyüán 因緣, yüánqĭ 緣起). A large number of scholars indicate that this verse was not only crucial for a number of classic Buddhist traditions, but also themselves continue to consider it as central to the text itself. For example, Garfield states that “It is generally, and in my view correctly, acknowledged that chapter 24, the examination of the Four Noble Truths, is the central chapter of the text and the climax of the argument” (Garfield 2002: 26). Similar claims have been made by Kalupahana (1986: 28f, 31-7), Garfield (1994; 2002: 69-85), Westerhoff (2009: 91-127), and Siderits & Katsura (2013: 13ff), among others. The four noble truths are themselves a locus classicus of the principle of dependent arising. The import of this fundamental law of arising in dependence is further strengthened with reference to the very opening verses of the Kārikās, which state: 6 Published Page Number: 000180 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? anirodham anutpādam anucchedam aśāśvatam| ­ anekārtham anānārtham anāgamam anirgamam|| ­ yaḥ pratītyasamutpādaṃ prapañcopaśamaṃ śivam| ­ deśayāmāsa saṃbuddhas taṃvande vadatāṃ varam|| I prostrate to the fully awakened one, the best of speakers, who has taught dependent origination, the appeasement of conceptual proliferation, the auspicious, which is not cessation, not production, not annihilation, not eternalism, not singularity, not plurality, not coming, not going. As such, the two aspects, namely emptiness and dependent origination, as well as the relationship or even equivalence between them, have long been considered by most scholars to be of central concern for Nāgārjuna and the Madhyamaka as a whole. Again, we see this attitude in a large number of studies, such as those of Kalupahana (1986: 28f, 31-7); Garfield (1994; 2002: 69-85); Westerhoff (2009: 91-127); and Siderits & Katsura (2013: 13ff). The discourse on emptiness has largely given the Prajñā­pāramitā Sūtras as the source of Nāgārjuna’s teachings on emptiness, though much less so for dependent arising (see “Discourse” point 5). This naturally fitted with his status as one of the founders of the Mahāyāna and a number of its philosophical systems, not just the Madhyamaka. In fitting with the classical “Mahāyāna vs Hīnayāna” polemic, it was not only thought that his teachings on emptiness (and dependent origination) were not derived from the so-called “Hīnayāna” sources, but must be in active opposition to them (see “Discourse” point 4.4). This notion has been challenged in the last decades, with several scholars first questioning the Mahāyāna status of either Nāgārjuna himself, or of the Kārikās. The citation of the Nikāya and Āgama teaching to Katyāyana in Nāgārjuna’s Kārikās 15:7ff is well known (eg. Kalupahana 1991: 232ff; Siderits & Katsura 2013: 159ff). Several decades ago, Warder heralded a change in English language studies when he challenged this fundamental assumption, asking “Is Nāgārjuna a Mahāyānist?” (Warder 1973). Later, he stated the accepted tradition, yet pointed out the fact of the Mūlamadhyamaka Kārikā as referencing only Tripiṭaka materials (Warder 1998: 138). Others have since followed or countered this thesis. However, over half a century ago 7 Published Page Number: 000181 JCBSSL VOL. XI in China, Yìnshùn had already considered that Nāgārjuna’s seminal work was written based on the Āgamas, not Mahāyāna sūtras. In the introduction to his book An Investigation into Emptiness (Kōng zhī tànjìu 空之探究), Yìnshùn reflected on statements that he had made decades before Warder and others (1985: i; cf. Yìnshùn 1949: 18, 24): In the last few years, when I was reading the Prajñāpāramitā Sūtras for writing my book The Origins and Development of Early Mahāyāna Buddhism [1980], I thought of my views in Modern Discussion on the Madhyamaka over thirty years before [1949]: “The *Madhyamaka Śāstra is a treatise to elucidate the Āgamas”; and “the *Madhyamaka Śāstra takes the perspective of the Mahāyāna scholars … and selectively develops the profound meaning of dependent origination in the Āgamas, and firmly establishes the right view of (Mahāyāna) Buddha Dharma on the key-stone of dependent origination, the middle way.” This pithy summary again highlights scholars’ perceived relationship between emptiness and dependent origination in Nāgārjuna’s thought, in particular verse 24:18. Since such recent revisions, some efforts have been made to seek sources for Nāgārjuna’s thought from earlier sources. However, as our outline of the discourse shows, these have sometimes been fraught with source bias issues, in both the early Buddhism and mainstream sectarian Buddhism periods. Most studies and conclusions regarding “early” Buddhism rely solely on the Pāli canon of the Theravāda, and seldom investigate the Āgamas. The Pāli Theravāda here referred to is that of the Śrī Lankan Mahāvihāra, located some distance from the “central lands” of Gaṅgetic North India. Seldom are the Āgamas of other schools used, such as those Chinese translations of Āgamas from the Sarvāstivāda, the Dharmagupta and Mahāsaṃghika. Yìnshùn (1985) and Choong (1999) are exceptions here, though Lamotte (2001) also makes mention of Āgama sources. Given the relative dominance of the latter schools in mainstream Indian Buddhism, their texts and doctrines are more relevant to the greater picture of mainstream Indic Buddhist, whereas the Theravāda was geographically more removed. Yet even using these Nikāyas or Āgamas outright is problematic, for they belong to a given Buddhist mainstream school. The issue of using these to identify “early Buddhism”, if not the teachings of the “historical Buddha” himself, or even a repre­sentation 8 Published Page Number: 000182 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? of those teachings and tenets held before the various schools split apart, is a very complex and difficult matter. As for such problems in early Buddhism, likewise too for the mainstream period Abhi­dharma and commentarial literature. Often it is the 5th cty Visuddhimagga of Buddhaghosa and Vasubandhu’s 5th cty Abhidharma­kośa and Bhāṣya which are cited as representing the non-Mahāyāna traditions. But neither the Northern or Southern Abhidharma traditions began as the alleged fully fledged “reified” or “substantialist” positions that may be found in these two works. The substantialist theories which are the supposed target of the Mahāyāna according to the discourse on emptiness only reached this state perhaps even after the early Mahāyāna and Prajñā­pāramitā around the turn of the millennium was established. How can one anachronistically argue that the earlier Prajñā­pāramitā and / or Madhyamaka is a refutation or reaction against later Abhidharma substantialism, when the sources cited have such an historical relationship? Another important point is that such Abhidharma theories, in either early or later form, do not represent the entirety of pre- and non-Mahāyāna Buddhist thought. The Theravāda and Sarvāstivāda were originally closely related as Sthavīra schools, and likewise too, the Pudgala­vādins. The plethora of schools at this time period show a huge range of positions on a wide range of subjects. This includes Mahāsāṃghika influenced works such as the *Satyasiddhi Śāstra, and other non-Mahāyāna content from the Mahā­prajñā­pāramitā Upadeśa (see Yìnshùn 1985: 132f; 92ff). By studying the precedents to Nāgārjuna’s Madhyamaka thought with these problems unresolved, it is little wonder that claims of revolutionary thought and innovation are made with respect to his teachings on emptiness and dependent arising. But as we have shown, for examining these ideas in both early and mainstream Buddhism, the Pāli canon alone will simply not suffice, even if their Abhidhamma and commentarial texts are included, nor will the Abhidharma­kośa as sole representative of the Northern traditions. A deeper understanding of the teaching of emptiness across a broad range of mainstream schools is required. For this, the large number of Chinese sources need be utilized to their full extent. Moreover, closer care to the historical sequence and relationships of texts and doctrines is essential. Only then will we be in a position to ascertain the significance of Nāgārjuna’s 9 Published Page Number: 000183 JCBSSL VOL. XI Madhyamaka teachings on emptiness and dependent origination as a stage of Buddhist doctrinal development within their own historical context. 1.2. Literature Review Despite the aforementioned problems of source biases and anachronisms, even sole use of the Pāli canon is already sufficient to show that “empty” and “emptiness” were not first coined by the Mahāyāna, or Nāgārjuna. One of the earliest demon­strations of this can be seen in Karunaratne, who as per the academic discourse equated early Buddhist emptiness with absence of self or what pertains to self (1956, 1988: 169). While written in 1956, which would have made it a perfect foil for Murti as much as against Stcherbatsky’s earlier writings, this was unfortunately only published in 1988. As such, it has been largely overlooked. For early Buddhism, also working exclusively from the Pāli sources for emptiness in “early Buddhism”, we also have shorter essays such as Gómez’s “Proto-Mādhya­mika in the Pāli Canon” (1976), and Vélez de Cea’s “Emptiness in the Pāli Suttas and the Question of Nāgārjuna’s Orthodoxy” (2005), both of which show scholars’ great interest in Nāgārjuna’s Madhyamaka. They largely take emptiness as a philosophy, rather than as a matter of meditation, which is the main feature of emptiness in the early texts (see Choong 1999: 43-84; Yìnshùn 1985: 1-78). Though these attempts to trace earlier sources and contexts for Nāgārjuna are commendable, by relying only on the Pāli canon and neglecting other schools contemporary with it, the results are limited. In the end, though Vélez de Cea very clearly sees the problems of modern scholarship in this area, he really only shows that Nāgārjuna would probably not disagree with some basic tenets of the Nikāyas and Āgamas. This would be greatly assisted by broadening the source material from which comparisons are made. One of the few more comprehensive studies in English directly related to emptiness in early Buddhism is The Notion of Emptiness in Early Buddhism, by Choong (1999). Although the title explicitly states “early Buddhism” as the scope, he curiously states in his aims that he “shall argue that the teaching of emptiness is not a creation of early Mahāyāna, but that it has clear antecedents in early Buddhism”, and his very last 10 Published Page Number: 000184 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? words on the matter are: “… ‘emptiness’ is genuinely a teaching of the Buddha himself, and not simply a creation of the Mahāyāna” (pg. 2, 88). It has rather the tone of a Mahāyāna apologetic in the face of modern Buddhist studies’ text-historical criticisms. Another, different kind of apologetic, is found in Dhammajothi’s The Concept of Emptiness in Pāli Literature (2008), which seeks to “examine the relation between the concept of suñña in early Buddhism and emptiness in Madhyamaka” (pg. iii). He appears to wish to deflect the criticisms of Nāgārjuna and others away from the Theravāda, and towards other mainstream groups (pg. iii, 163). However, again, due to being “specially focused on the Pāli Tipiṭaka, and the Visuddhi­ magga” (pg. iii; pp. 100-120), this quite disorganized study can only be seen as representing the Theravāda, and neither “early Buddhism” or the entirety of mainstream Buddhism. Thus, while the author wishes to deflect Madhyamaka criticisms against the Theravādin tradition, those same criticisms remained unexamined as to their true target and precedents, and the “Hīnayāna” bogeyman remains hidden. It is to be admitted, lamentably, that there are even fewer systematic and dedicated studies into the doctrine of emptiness within the mainstream Buddhist period. This in turn reflects the general status of studies in this period, wherein there has been little research into specific doctrinal issues. However, let us examine these few relevant studies in order. Given the supposed “Mahāyāna vs Hīnayāna” polemical claims on emptiness within the discourse, it is not at all surprising that there are even fewer studies on this topic centered on mainstream sectarian material (source biases aside). In one of the many sub-essays within his epic five volume translation of the Mahā­prajñā­pāramitā Upadeśa (Dàzhìdù lùn 大智度論), Lamotte breaks the main­stream Buddhist schools into three types: 1. The “personalists”, such as the Pudgala­ vādin Vātsī­putrīyas and Saṃmittīyas; 2. The “realists”, namely the Thera­vāda and Sarvāsti­vāda Ābhidharmikas; and 3. The “nominalists”, for instance, the Mahā­sāṃghika Prajñaptivādins, and possibly nonAbhidharma Sthavira groups. Concerning their respective positions on emptiness, Lamotte establishes two basic positions, which reflect an earlier statement in the Upadeśa on the meaning of prajñā: “The teaching of emptiness is the emptiness of beings (pudgala­śūnyatā) and 11 Published Page Number: 000185 JCBSSL VOL. XI the emptiness of dharmas (dharma­śūnyatā)” (Lamotte 2001: 836f, 1650f). Due to the focus of this work, there is barely any discussion on these texts themselves. In his first two volumes of translation (1944 & 1949), Lamotte considered what he at the time called the “Śāstra” to be a work of Nāgārjuna. Though he changed his position on this in the third volume (1970), now considering it an “Upadeśa” and authored by anonymous Indic Madhyamaka convert scholar, he still thought it a Madhyamaka work doctrinally faithful to that of Nāgārjuna. Working directly with the philosophy rather than the text of the Upadeśa, Ramanan’s (1966) Nāgārjuna’s Philosophy: As Presented in the MahāPrajñā­pāramitā-Śāstra, also considered Nāgārjuna the author of this work, following Lamotte’s first two volumes. Regarding emptiness in mainstream schools, he outlined three groups, albeit different from those of Lamotte (Ramanan 1966: 53ff): 1. The pluralist line—the Sarvāstivāda. He states that: “The Vibhāṣā admits dharm­ātmā while it denies pudgal­ātmā”, and: “The Sarvāsti­vādins admit śūnyatā. For them this means that among the basic elements of existence there is no ātman, no eternal substantial entity called “I”. They interpret the Middle Way so as to make it agree with their doctrine of elements”. 2. The line in between—Vātsī­putrīyas, Sāṃmitīyas, Sautrāntikas (Saṃkrantivādins) and Mahīśāsakas. (Note: all Sthavira groups.) 3. The absolutist line— the Mahāsāṃ­ghikas: “The credit of having kept alive the emphasis on the ultimacy of the unconditioned reality by drawing attention to the non-substantiality of the basic elements of existence (dharma­śūnyatā) belongs to the Mahāsāṃghikas”; “[t]hese were the teachings of … śūnyatā of the composite as well as the incomposite, the absence of self-being in things and the peace of Nirvāṇa”. We would like to the point out, however, that this last point was also accepted by most of the Sthaviras too (Yìnshùn 1985: 104, 113), though the Pudgalavādins may be an exception (Bareau 2005: 116 #38). This brings us finally to Yìnshùn’s work An Investigation into Emptiness (Kōng zhī Tànjìu 空之探究) (1985), and excellent study unfortunately almost unknown outside of the sphere of Buddhist studies in Chinese language.1 It covers the doctrine of emptiness as it developed through four stages. 1. Early Buddhism of the Āgamas and Nikāyas; 2. The Abhidharma or Nikāya period; 3. The early Mahāyāna period of the Prajñā­pāramitā literature; and 4. The Madhyamaka philosophy of 12 Published Page Number: 000186 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? Nāgārjuna, including the Mahā­prajñā­pāramitā Upadeśa, which is traditionally attributed to him. As such, it is one of the most complete modern studies on the theme of emptiness. Structurally, it does follow the broad historical categories of the discourse, but avoids the anachronistic use of material, and draws on numerous lines of textual tradition. For early Buddhism, Yìnshùn uses both classic Chinese translations and modern Chinese translations of the Pāli canon, and for the mainstream period, has mastery over the entire range of sectarian literature well preserved in Chinese, in addition to modern translations of Pāli Abhidhamma and para­canonical material. As can be seen, previous studies into the broader question of preMahāyāna teachings on emptiness are few and often apologetic in nature. Very little has been at all on the more specific question of how emptiness relates to dependent origination. As such, the notion within the academic discourse on emptiness, which itself works backwards from the accepted idea that Nāgārjuna’s teachings were rather innovative in the light of the early and mainstream “Hīnayāna” positions, remains unexamined and unchallenged. The next two sections of this essay will thus examine the pre-Nāgārjunian literature on emptiness and dependent origination in two general historical periods, namely “early Buddhism” (Section §2) and “main­stream Buddhism” (Section §3). The distinction between these periods is as much for convenience as representing a clear cut historical division, and itself partly follows the modern academic discourse attitude toward phases of Buddhist doctrinal qua historical development. 2. Emptiness & Dependent Origination in Early Buddhism We will first examine the canonical texts of early Buddhism to understand the relationship between empty / emptiness and dependent origination. By “early” we do not mean to imply “original”, the philological holy grail of establishing the “original words of the historical Buddha” now being somewhat out of vogue. We do believe that by cross comparison of parallel texts from a number of early Buddhism schools will reveal in their commonalities those basic teachings that existed before the division into such schools occurred. Taking the first basic schism of the Buddhist community into the Sthaviras and Mahāsāṃghika to have occurred during the time of Aśoka, ie. circa 268-232 bce, with later 13 Published Page Number: 000187 JCBSSL VOL. XI splits continuing subsequently, we can give an approximate date for “early Buddhism” from the time of the Buddha himself up to the 3rd or perhaps even the 2nd century bce. This is almost universally regarded as well before the start of the Mahāyāna, and many centuries before the time of Nāgārjuna himself. The extant texts of the earliest sūtra discourses are the five Nikāyas of the Pāli canon, and also the Chinese translations of the four Āgamas. The five Pāli Nikāyas are: 1. The Samyutta Nikāya (SN); 2. The Majjhima Nikāya (MN); 3. The Dīgha Nikāya (DN); 4. The Anguttara Nikāya (AN); and 5. The Khuddhaka Nikāya (KN), which itself contains a range of material, wherein the Dhamma­pada and Sutta­nipāta are considered the oldest strata. This tradition is what is now commonly known as the Theravāda. The Chinese Āgama translations are:2 1. The Saṃyukta Āgama (SĀ, T99), translated 435-443 ce (two fascicles are missing). There is also the Alternative Translation Saṃyukta Āgama (AltSĀ, T100), possibly from the Kāśyapīya school. 2. The Madhyama Āgama (MĀ, T26), translated 397-398 ce. These Āgamas are considered to belong to the Sarvāstivāda traditions. 3. The Dīrgha Āgama (DĀ, T1), translated 413 ce, of Dharmagupta origins. 4. The Ekottara Āgama (EĀ, T125), translated 397 ce and later revised. This is from a late sect of the Mahāsāṅghika and already contains some Mahāyānic philosophy. 3 It must thus be used cautiously in any context of “early” Buddhism. Before we examine the texts, a note on the topic of dependent arising, the Buddhist law of causality, is in order, for it is a theme which strongly underlies much of early Buddhist texts. In fact, there is much evidence to suggest that the entirety of the Buddha’s teachings can be subsumed under this general principle.4 From the SN in particular, we find terms such as “stability of Dhamma” (dhamma­ṭṭhitatā), “certainty (or law) of Dhamma” (dhamma­niyāmatā), “specific conditionality” (idap­ paccayatā) and “Dhammic nature” (dhammatā) to describe dependent origination, with the SĀ providing an even greater number of such synonyms.5 Therefore, here we shall only focus on this doctrine as far as it is directly related to emptiness and related doctrines. 2.1. “Profound, Connected with Emptiness” 14 Published Page Number: 000188 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? One of the more pithy and explicit connections between emptiness and dependent origination, is that found in SN 20:7 Aṇī or SĀ 1258. In it, the Buddha states that his disciples should study “those discourses taught by the Tathāgata that are profound, profound in meaning, transmundane, connected with emptiness”, rather than those texts which are “mere poetry composed by poets, beautiful in words and phrases, created by outsiders”.6 The term “profound” or “deep” (gambhīra) was originally imbued with a simple prosaic sense, such as the “deep ocean”. Here, and elsewhere in the Pāli texts, it is used to describe the Dhamma realized by the Buddha, as expressed in SN 6:1 and MN 26 and 27:7 Profound, hard to see, hard to understand, peaceful and sublime, not within the sphere of reasoning, subtle, to be experienced by the wise. … For such a generation this state is hard to see, that is, specific conditionality, dependent origination. And this state too is hard to see, that is, the stilling of formations, the relinquishing of all acquisitions, the destruction of craving, dispassion, cessation, nibbāna. This is a complement to the statement that seeing dependent origination is seeing the Dharma itself. The Āgama equivalent of these Pāli texts either do not have this passage, or show only an abbreviated form.8 However, this two-fold meaning of “profound” as both the law of conditionality and also pacified liberation does appear in SĀ 293, which states that “these two dharmas are known as the conditioned (saṃskṛta) and the unconditioned (asaṃskṛta)”.9 The sense of the law of dependent arising as profound is further emphasized in DN 15 Mahā­nidāna Sutta, and corresponding MĀ 97 and DĀ 13, where Ānanda on contem­plating dependent co-arising states that it “is wonderful and marvelous, … how this dependent origination is profound and appears profound. Yet, to myself, I see it as clear as clear can be,” and was thus berated by the Buddha for under-estimating the profundity of this Dhamma.10 One way to understand this dual aspect of the “profound”, is to consider it in terms of dependent origination and also dependent cessation, the former as the arising which is saṃsāra or dissatisfaction—the conditioned, the latter as its cessation or nibbāna—the unconditioned. These correspond to the two complementary aspects of the standard 15 Published Page Number: 000189 JCBSSL VOL. XI formula in Udāna 1:1-3: “When this is, that is; When this arises, that arises”; and “When this is not, that is not; When this ceases, that ceases.”11 2.2 Emptiness of Self in Dependent Cosmogenesis Why does SN 20:7 Aṇī and SĀ 1258 thus make such an emphatic statement on what seems to imply the identity of dependent arising and emptiness? A clear candidate to answer this derives from some recent research by Jurewicz into the idea of the Buddha’s twelve-fold dependent origination as a response to Vedic cosmogenetic theories of the Ṛg Veda (X 129), and other pre-Buddhist Brāhmaṇic literature, such as the Śatapatha Brāhmaṇa, Bṛhad­āraṇyaka, Aitareya, Taittirīya and Chāndogya Upaniśads (see Jurewicz 2000: 171). The Vedic cosmogonies all revolve around what we may call the transformations of the ātman, whereas for the Buddhist twist to this theory, Jurewicz considers that “the Buddha chose those cosmogonic descriptions which met two conditions: first, they explicitly express the cosmogony as transformations of the ātman; second, they preserve their cognitive meaning, even if they are taken out of the Vedic content” (Jurewicz 2000: 80). Likewise says Schulman (2008: 297): “Rather than relating to all that exists, dependent origination related originally only to processes of mental conditioning. It was an analysis of the self, not of reality, embedded in the Upaniśadic search for the ātman.” For the Vedic tradition, “nidāna” referred to the connection between the world at large and the microcosm represented in the sacrificial fire, and this connection was the ātman. But for the Buddha’s “nidāna”, “there is no ātman … The negation of the ontological nidāna constitutes the Buddha’s mahānidāna” (Jurewicz 2000: 100). The various twelve links are broken down by Jurewicz into small consec­utive groups, each of which is closer to some or other passage in the various Brāhmaṇic texts listed above. Despite this, the ātman idea is seen throughout them all, whereas other phenonema such as the puruṣa, fire and so forth, have more or less importance at different stages of the dependent arising process. Jurewicz envisages the Buddha using the terminology of the Vedic cosmogony, only to conclude: “That’s right, this is how the whole process develops. However, the only problem is that no one undergoes 16 Published Page Number: 000190 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? a transformation here!” This is “to deny the ātman as the metaphysical basis of all cosmogonic transformations as well as its final forms” (Jurewicz 2000: 101). Rather than the creative power of the soul, it was an absurd and pointless cycle of death and rebirth. As Gombrich rightly concludes, Jurewicz’s interpretation through the light of ancient pre-Buddhist theories, which were no doubt known to both the Buddha and much of his more learned audiences, shows its function of “adding substance and detail to the Buddha’s ‘no soul’ doctrine” (Gombrich 2003: 14; also 2009: 133ff). Thus, references to either not self or emptiness in the many expositions on dependent origination, would have simply been unnecessary for the Buddha at that time. It would have been implicitly understood that he thus taught saṃsāric arising without a soul. The process of dependent arising is empty. 2.3. Causality and the Middle Way We find other examples of how the Buddha expressly rejects an actor agent in favour of the doctrine of dependent origination in, for example, SN 12:12 or SĀ 372. Here, when asked about agency, as “Who consumes the nutrient of cognition? … makes contact? … feels? … craves?”, the Buddha declares that the question is invalid. Rather than “who?”, the question should be “For what is the nutriment cognition [a condition]? … etc.”12 In the teaching given by Ānanda to Channa (SN 22:90 Channa, SĀ 262 *Chanda 闡陀), which brought about the latter’s break­through into the Dhamma, is for all purposes a verbatim repetition of that famous discourse given by the Buddha himself to Kaccānagotta (SN 12:15)— avoiding the two extreme views of existence and non-existence, and pursuing the middle way which is the dependent origination and cessation of dissatis­faction.13 The connection in SĀ 262 between emptiness and absence of (extreme) views is oblique, apart from emptiness as mere “absence” itself, and is via the doctrine of dependent origination. The implication being that the views of either existence or non-existence, of body and soul as identical or different, etc. are incompatible with the Buddha’s unique teaching of conditionality. 17 Published Page Number: 000191 JCBSSL VOL. XI The rejection of arising from any one or other of the four categories of self, other, both or neither (non-causality), all types of extremes to be avoided, is a recurring theme throughout SN 12 Nidāna­samyutta.14 Rejection of arising from “self” can be seen as further expressions of emptiness as not self (or what pertains to self), as the usual “selfview” predominant in non-Buddhist Indian religious-philosophical systems was one of “existence”. For example, in “saccid-ānanda”, and the uncaused cause. Rejection of arising from “other” is in effect just “another self”. With these two rejected, naturally arising from “both” is also out of the question. Yet the Buddha’s strong emphasis on causality also meant that an outright rejection of all causality, things being uncaused, was also totally out of the question. 2.4. Nirvāṇa as “Empty Pacification” Where the cosmogenetic causality of the preceding sections represents the world and its coming into being—dissatisfaction (duḥkha), the goal is the cessation of this world—the liberated release entailing extinguishing the fire of afflictions (P: nibbāna, Skt: nirvāṇa). A well known passage in SN and SĀ states that “the destruction of desire, … aversion, … delusion—this friend is called nibbāna”, “… the unconditioned” (asaṅkhata) or “… the fruition of a worthy one” (arahanta­phala).15 The second definition here, that of the “unconditioned”, elsewhere has another similar statement again in SĀ 262, which gives a list of definitive terms as follows:16 … the empty pacification (*śūnya­śamatha) of all conditionings, their non-apprehension (*anupalabhyate), the destruction of craving, the fading away of desire—nirvāṇa. Given that the equivalent sutta in SN has merely “the pacification (samatha) of all formations …”,17 one may question both the Chinese translators here and their use of the term “空寂”, and also our English translation as “empty pacification”. Should not the Chinese phrase really just be “pacification” or “appeasement” (*śamatha)? We think not. As Yìnshùn astutely notes, the Yogācāra­bhūmi Śāstra on citing this text verifies that in fact the Sarvāstivādin SĀ did indeed use the term “empty pacification” (śūnya­*śamatha) here. It explains this very sūtra passage as “The term ‘empty’ means the forsaking of all the 18 Published Page Number: 000192 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? defilements” (see Yìnshùn 1985: 118).18 Thus while there is the implied sense of “absence” of afflictions and their resultant dissatisfaction in unconditioned release, the SĀ, at least, explicitly refers to this as “empty” (śūnya). 2.5. The Emptiness Samādhi as Absence of Causes Here, a term in early Buddhism which later became almost synonymous for the path to liberation as a whole, is that of “mental release” (P: ceta­ vimutti) or “mental meditation” (Skt: *ceto­samādhi; 心三昧). There is also a connection here between emptiness and causation. The Pāli Nikāya term “mental release”, which appears in the Āgamas as “mental meditation”, is the common name given to a set of four practices analyzed by several of the Buddha’s great disciples in several texts.19 Also, at times the Pāli also does use ceto­samādhi, rather than just ceto­vimutti, but only for the signless, and not for the other three immeasurables.20 For example, in MN 43, when the question is asked as to “What is the signless mental release?” (°-vimutti), the actual answer is given in terms of the “signless mental concentration” (°-samādhi).21 The Āgamas, on the other hand, do not seem to use the equivalent cetovimukti (*心解脫) for the four immeasurables, though this term itself is used in other contexts. According to SN and SĀ, the four are: 1. Immeasurable (appamāṇa-°, apramāṇa-°; 無量) mental release / meditation; 2. Nothingness (ākiñcañña-°, ākiṃcanya-°; 無所有) mental release / meditation; 3. Emptiness (suññatā-°, śūnyatā-°; 空) mental release / meditation; and 4. Signless mental release / meditation (animitta-°; 無相).22 In response to a question, they are all explained in two senses, firstly as “different in meaning and also different in phrasing”, and secondly as “one in meaning and different only in phrasing”. There are some slight differences between answers in the Pāli and Āgama readings, but we wish to draw attention to the second set of answers, where they are alike in meaning and differ only in expression. As for the four being synonyms, “one in meaning and different only in phrasing”, it is explained that: 1. Desire, aversion and ignorance are “makers of measurement (or limit)” (pamāṇa­karaṇa), their absence 19 Published Page Number: 000193 JCBSSL VOL. XI (suñña) is the “immeasurable (or unlimited) release”; 2. These same three defilements are causes for “something” (kiñcanaṃ), a synonym for saṃsāric becoming, the absence (suñña) of “something-ness” is hence the “nothingness release; 3. The emptiness mental release is not explained here, but is a generic term for all three (see below); 4. The three defilements are also “makers of signs” (nimitta­karaṇa), their absence (suñña) is the “signless” release. The unifying theme through #1, #2 and #4, is actually #3, the “emptiness mental release”. While its unifying function is not explained explicitly, it is obviously referred when the other three are described in terms of being empty (suñña) of defilements. That is to say, of the other three, their ultimate culmination is the “immovable mental release” (P: akuppa; Skt: akopya) in SN, or “non-conflict” (*araṇa; 無諍) in SĀ.23 This is the absence (suññatā) of the three root defilements which act as makers of measure-limits and signs, and are the basis for becoming. These three terms—limit (pamāṇa), somethingness (kiñcana) and sign (nimitta)—all refer to defilements and their causes, and “emptiness”, here given as “nothingness”, is their forsaking. 2.6. The Three Dharma Seals and Characteristics At the end of the early period, when sectarian doctrines already start to make their appearance in the Nikāyas and Āgamas, we shall turn to material that quickly developed into standard criteria for the authentic Dharma, namely the Dharma seals (mudra) and characteristics (lakkhaṇa). In many ways, the colloquial use of the term “seal” (mudrā) is similar to both “nimitta” and “lakkhaṇa” in the objective sense, the special mark or sign, often of an important official, a classic example being the “royal seal” (rājā­muddā) (see PTSD 570). There are several early texts which provided the implicit principles behind the later systematized and explicit formation of the three Dharma seals. We shall deal with the Sarvāstivādin SĀ, as it is the only one to use the term in an early text—SĀ 80, the *Ārya­dharma­ mudrā­jñāna­darśaṇa-viśuddhi Sūtra.24 This is perhaps the only early sūtra which discusses the “dharma seals” (dharma­mudrā) in direct relation to the three samādhis, headed by the śūnyatā­samādhi. Indeed, the Pāli tradition uses the term “three characteristics” (ti­lakkahaṇa) 20 Published Page Number: 000194 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? for a slightly different set of statements with similar overall intention, albeit without the direct association to emptiness. SĀ 80 also deals with the three samādhis of emptiness, nothingness and the signless. It explains that the emptiness samādhi is required before proceeding to nothingness and the signless, and that each of the three has a specific contemplation: One, in the emptiness samādhi they contemplate that each of the aggregates “is impermanent and subject to cessation” and “not solid or stable, but subject to change”, they then “become detached from desire”. This is actually more of a contemplation of impermanence than not self, which is the more common gloss for emptiness. Two, in the signless samādhi they forsake the signs of the six sensory objects, form, sound, etc.. This is exactly the same as the earliest idea of “non-attention to all signs”, as found in the exegetical sūtra MĀ 211.25 This conforms to the position of SĀ 80 here, as it does not reify the signless into an object to which one can direct attention, unlike parallel text MN 43 Mahāvedalla.26 Three, in the nothingness samādhi they forsake the signs of the three root defile­ments of desire, aversion and delusion. Again, the early explanation of the defile­ments as “somethings”, causes for existence in saṃsāra. Thus, up to this point in SĀ 80, the signless and nothingness basically match SN 41:7 and SĀ 567, which were the precursors to the exegeses in MN 43 and MĀ 211. However, and more pertinent for our discussion of emptiness and dependent arising here, after these three contemplations,27 one investigates … that [notions of] “I” and “mine” arise from either what is seen, or heard, or smelt, or tasted, or touched, or cognized. … Moreover, they investigate in this manner: By whatever cause or whatever condition that cognition arises, those causes and those conditions are all impermanent. Moreover, when those causes and those conditions of that [cognition] are all impermanent, how could the cognition itself which has arisen from them be permanent? Whatever is “impermanent” is conditioned (*saṃskṛta), a form­ation (*abhi­saṃskāra), arisen 21 Published Page Number: 000195 JCBSSL VOL. XI from conditions (pratītya­saṃutpanna); is subject to decay (*vaya­dharma), subject to cessation (*kṣaya­dharma); subject to fading away (*virāga­dharma), subject to cessation (*nirodha­ jñāṇa­dharma). This is known as the purity of gnosis and vision of the āryan dharma seals. So, what exactly here does the phrase “dharma seals” in SĀ 80 refer to? Here, “dharma” is used in the sense of being “subject to” some or other principle, as per SN 12:34,28 for which Bodhi has: “… subject to destruction, vanishing, fading away, and cessation” (Bodhi 2000: 573). Parts of the Chinese Āgama, when read alone, are slightly ambiguous.29 These principles mainly refer to cessation, but this is in turn one aspect of causal conditionality. Thus, SĀ 80 is using a fairly standard set of terms used to describe conditioned phenomena, almost implying the realized goal as the natural state of conditioned phenomena. The list is headed by phenomena being dependently originated, showing the underlying principle behind the arising and ceasing of dharmas, their conditionality. The remaining four basically synonymous terms indicate the impermanence of conditioned phenomena. These are contemplations used to eradicate the view of a self or what pertains to self. Thus, the purity of gnosis and vision of the āryan dharma seals is largely about eliminating internal and external defilements, including self view and self conceit, through various forms of contemplation. 2.7. Summary Between the aforementioned Nikāya and Āgama explanations, emptiness and dependent origination were related as a key part of the Dharma from its inception. The previous six sections can perhaps be divided into three broad groups. In the first category, sections §2.1 and §2.6, emptiness relates to both the process of dependent arising and also cessation. We have seen that the notion of “profound” or “deep” (gambhīra) as referring to both dependent origination and also dependent cessation—nibbānic release—was present though not overly strong in the very early canon. However, the Sthavira traditions considered it important enough at quite an early date, texts and statements which appear to be specific to both 22 Published Page Number: 000196 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? the Theravāda and Sarvāstivāda schools. Our examination of the three dharma seals is from the later stages of the early tradition. The texts still explicitly connect this to dependent origination, that phenomena are subject to arise and cessation. As “seals” of conditioned dharmas, there is a gradually movement toward universality. The second group, in sections §2.2 and §2.3, we see how the Sarvāstivāda in particular used the term “emptiness” to refer to dependent arising in a broad sense. While this mainly focuses on its forward saṃsāric process, there is still a connection with the reversal into nirvāṇa. Recent studies by Jurewicz provide a key connection between the implied sense of lack of self, ie. emptiness, within the Buddha’s standard twelve limb form of dependent origination, and other expressions thereof, as a parody of Brahmanic cosmogenesis. Included within this broader sense of causality, was the middle way of the absence of extreme views, which were considered counter to the position of the Dharma as dependent origination itself. Lastly, a third group consisting of sections §2.4 and §2.5 makes stronger the relationship between dependent arising and emptiness as nirvāna. Again, it is the Sarvāstivādin tradition that makes the explicit connection of release as “empty pacification”. But both this school and the Theravāda use the emptiness medita­tion or mind release (respectively) as a catch all term for the practices that bring about the total elimination of afflictions as causal factors. 3. Emptiness & Dependent Originationin Mainstream Buddhism We can now move from the early period, to that of the subsequent mainstream sectarian period. From the last section (§2.6) in particular, it is important to recall that this historical distinction is a simple heuristic rather than a hard delineated fact. As we underscored in our criticisms of the modern academic “discourse of emptiness” in the introduction, there are several methodological problems with regards to the mainstream period, in particular source biases and anachronist explanations of doctrinal relations and development. Regarding source biases, there is the heavy usage of later Abhidharma literature, particularly that of Vasu­bandhu’s Abhi­dharma­kośa­bhāṣya 23 Published Page Number: 000197 JCBSSL VOL. XI and Buddhaghosa’s Visuddhi­magga, as repre­sentative of this period as a whole. This reflects a more general bias towards viewing only the extant Pāli and Sanskrit literature as of value in the study of Indian Buddhism. With respect to historical anachron­isms, the Mahāyāna position is typified by the doctrine of emptiness from the Prajñā­pāramitā and systematized by Nāgārjuna, and the aforementioned Abhi­dharma texts represent the mainstream period. However, the former texts both hail from the 5th century, whereas the latter are from the 1st to 3rd centuries— how can the latter be a critical response to the former? Therefore, in this section, we must pay close attention to textual sources that are deemed to date from before Nāgārjuna’s time (2nd—3rd cty ce), and even then, be aware of their relative historic relationships. The previous section on early Buddhism spanned up to the 3rd, or at the latest, the 2nd century bce, as noted at the start of the previous section. Our material here begins with the latter end of this period, when the Nikāyas and Āgamas were still being compiled by each of the schools. Therefore, though we shall naturally cite śāstra sources which are more obviously affiliated with a given school, we shall first cite several sūtras. These are texts which appear in the Āgamas of some schools, but are not necessarily found in or approved by other schools, thus lying outside of our basic criteria for textual sources of “early Buddhism”. In particular, this includes several sūtras for which we have Chinese translations in the Sarvāstivādin SĀ and the later Mahāyāna influenced Mahāsāṃghika EĀ (both discussed previously in §2), in addition to the Śālistamba Sūtra. For the mainstream sectarian period, it is admittedly more natural to refer to the Abhidhamma or Abhidharma literature. The various schools disagreed as to whether or not it was recited at the first convocation. The Mahāsāṃghika and two Vibhajyavāda schools, the Mahīśāsaka and Theravāda, did not mention its recitation there in their respective Vinayas, but only spoke of it as the third Piṭaka in their later commentarial traditions. The Sarvāstivāda, Haimavata, Dharmagupta and Mūla­sarvāstivāda did include it, but differed in their details. In the commentarial literature, both the Theravāda and Sarvāstivāda stated that the Abhidhamma / Abhidharma was in fact the “word of the Buddha” (see Yìnshùn 1968: 9-11). However, the Theravāda meant 24 Published Page Number: 000198 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? this in the literal sense, whereas the Sarvāstivāda had a more figurative explanation. The core Abhidharma literature thus became even while the Nikāyas and Āgamas were being compiled and finalized, up to the 2nd century bce. This is still a century or two before the early Mahāyāna, giving it sufficient time to be propagated broadly across the Indian subcontinent. Of the two Abhidharma schools, the Pāli Theravāda has seven core texts: 1. Dhamma­saṅgaṇī; 2. Vibhaṅga; 3. Dhātu­kathā; 4. Puggala­paññatti; 5. Yamaka; 6. Paṭṭhāna; and 7. Kathā­vatthu. 30 The other Abhidharma school, the Sarvāstivāda, also had seven Abhidharma Śāstras, albeit different to the above: 1. Dharma­skandha­pāda Śāstra (T1537 法蘊足論); 2. Saṃgīti­paryāya Śāstra (T1536 集異門足論); 3. Prajñapti Śāstra (T1538 施設論); 4. Vijñāna­kāya Śāstra (T1539 識身 足論); 5. Jñāna­prasthāna Śāstra (T1543 八犍度論) and (T1544 發智 論); 6. Prakaraṇa­pāda Śāstra (T1541 眾事分) and (T1542 品類足論); and 7. Dhātu­kaya­pāda Śāstra (T1540 界身足論). 31 Additionally, there is the Śāriputra Abhidharma Śāstra (T1548 舍利弗阿毘曇論). This is also a Vibhajyavādin work, exhibiting clear structural parallels with the Vibhaṅga and Dharma­skandha­pāda Śāstra, and also the Dhamma­ saṅgaṇī and Prakaraṇa­pāda, from the other two Abhidharma traditions. It is believed that the Śāriputra Abhidharma Śāstra was probably shared with the early Vātsīputrīyas, Dharmaguptas, and other more central Indian Sthavira Vibhajya­vādin schools. 32 What we now have of the Theravāda commentarial tradition by Buddhaghosa in the Aṭṭha­kathās is a summary of earlier material. We still have the Northern Sarvāsti­vādin commentarial Vibhāṣa literature in Chinese translation, most notably the Abhidharma Mahāvibhāṣa Śāstra (T1545 大毘婆沙論), with variants Abhi­dharma Vibhāṣa Śāstra (T1546 毘婆沙論) and Vibhāṣa Śāstra (T1547 鞞婆沙論). This Vaibhāṣika standard was likely compiled by a large number of Abhidharma scholars over the course of centuries, in order to establish orthodoxy within their own ranks, as well as counter the views of other schools, and reached its basic final form in the mid 2nd cty ce (see Yìnshùn 1968: 209-220; Dhammaj­oti 2007: 65). 25 Published Page Number: 000199 JCBSSL VOL. XI As such, the Mahāvibhāṣa is the last pre-Nāgārjunian material that we shall examine here. Citing material later than Nāgārjuna would endanger our goal here by potentially falling into anachronistic arguments as discussed earlier in the discourse on emptiness above. However, this time period coincides with the formative period of the incipient Mahāyāna movement, and we can feel the currents of mainstream school notions of the bodhisattva—here referring to Śākyamuni—influencing the new movement, even while these mainstream Ābhidharmika commentaries did not seem to be aware of the Mahāyāna as a distinct school in its own right. We shall see how mainstream notions of the bodhisattva, and how they involve both emptiness and dependent origination, may have developed in these texts in Section §3.4, below. 3.1. Ultimate Emptiness—“Neither Coming Nor Going” The first sectarian sūtra under consideration is the Paramārtha­śūnyatā Sūtra, extant as SĀ 335 and EĀ 37:7. 33 If we only compared with the Pāli to find no equivalent of this sūtra, we may be tempted to classify it as a text of mainstream school origins. However, the fact that the Sthavira Sarvāstivādin version is nearly word for word identical with the Mahāsāṃghika EĀ version—with the exception of some juxta­ posing of paragraphs which is negligible in terms of content—suggests that the text could also possibly be considered “early”. It is possible that the Theravāda tradition may have lost their own version of this text at some point in time. The Paramārtha­śūnyatā Sūtra (SĀ 335) states:34 … when the eye arises, it does not come from any location; when [the eye] ceases, it does not go to any location. In this way, the eye is unreal, yet arises; and on having arisen, it ends and ceases. There is action (karma) and result (vipāka), and yet no actor agent (kāraka). On the cessation of these skandhas, another set of aggregates continues elsewhere (anyatra). There are merely dharmas classified as conventional, ie. the ear, nose, tongue, physical body and mind, are also declared as such. Here, the parallel text EĀ 37:7 also adds “These six sense faculties are also not created by a ‘person’ (*puriṣa, *pudgala).”35 This further emphasizes the absence of an agent. According to Karunadasa, such statements on neither coming nor going are also found in the later Theravādin commentarial literature: “There is no store (sannidhi) from 26 Published Page Number: 000200 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? which they come and there is no receptacle (sannicaya) to which they go”;36 and “If they appear it is not that they come from somewhere (na kuto ci āgacchanti); if they disappear it is not that they go anywhere (na kuhiñci gacchanti)” (Karunadasa 2010: 30-31). 37 This is the same basic gloss the commentaries also give to SN 35:246 Vīṇā (= SĀ 1169), which uses a simile of how the music of a lute is dependent on a number of parts and factors. 38 The phrase in translation “除俗數法” (“dharmas classified as ‘conventional’”) poses some problems for interpretation. We must examine the various versions of this sūtra and also parallels of the passage in other texts. The phrase appears in the Bimbisāra Sūtra, which is in turn cited in the Mūla­sarvāstivādin Vinaya. 39 Subsequently, in the Kośa and Bhāṣya.40 we conclude that the original term was “*dharma­ saṅkheta”, which we translate as “dharmas classified as conventional”. This is akin to phenomena being “nominal” here, though semantically at least, it differs from the Abhidharma usage of “prajñapti” as “designation”, and the implications of it be vis-à-vis “paramārtha”, the “ultimate sense”. The SĀ version of the sūtra then continues, stating:41 Dharmas classified as ‘conventional’, that is to say—‘When this exists, that exists; when this arises, that arises … the arising of this sheer great mass of dissatisfaction.’ … Moreover—‘When this does not exist, that does not exist; when this ceases, that ceases … the cessation of this sheer great mass of dissatisfaction.’ O monks! this is named the Dharma Discourse on Ultimate Emptiness (Paramārtha­śūnyatā­dharma Sūtra). It is thus clear that “dharmas classified as conventional” refers to the dharmas which comprise the limbs of dependent arising itself. This is not only in the for­ward order of arising, but also the reversal into cessation. The cessation of dis­satisfaction is nirvāna, the unconditioned. Is it any particular one of these aspects which is “ultimate emptiness”? It is difficult to say, but the overall sense appears that it is the totality of this situation, the Dharma law of dependent arising and cessation of conventional or nominal dharmas as phenomena, all of which takes place without recourse to an agent or actor, that is “ultimate emptiness”. It could be possible to then read this through the “two truths” system 27 Published Page Number: 000201 JCBSSL VOL. XI so popular in the mainstream sectarian period, utilized by both the Abhidharma systems as well as the Madhyamaka. In that way, the “ultimate” could be juxtaposed against the “designated” (prajñapti), giving the underlying principle of dependent origination as the former, whereas the phenomena are the latter. 3.2. Great Emptiness—“Neither Identity Nor Plurality” The second sectarian sūtra is another Sarvāstivādin text, SĀ 297 Mahā­śūnyatā Dharma­paryāya.42 Without a Pāli equivalent, Yìnshùn (1971: 651) locates it in the Nidāna­saṃyukta, which is supported by the content and surrounding sūtras. It has a similar teaching format to SĀ 262, above. As we would thus expect, it explicitly uses the standard twelve link formula of dependent arising. However, it also ties dependent origination in with emptiness:43 What is the Dharma Discourse on Great Emptiness? It is this— ‘When this exists, that exists; when this arises, that arises.’ Which is—‘Formations are conditioned by ignorance; cognitions are conditioned by formations; and so forth, up to; the amassing of this sheer mass of dissatisfaction.’ … Then follows a refutation of several positions which appear in the ten or fourteen unanswered questions (avyākṛta), all of which are forms of “self-view” (ātma­dṛṣṭi), such as “life (jīva) and body are different” or “identical”. A very similar statement occurs in SN 12:35,44 lending further support to placing SĀ 297 in the Nidāna­saṃyukta. The sūtra stresses dependent origination as the “middle path” between these extremes of view based around the idea of a self (as jīva).45 The connection between these questions and the rest of the text is most likely that such questions are basically predicated on the notion of a “self”, though often under the guise of the term “tathāgata” (Karuna­ dasa 2005). As we have already seen from Jurewicz’s studies above, this is obviously one of the key points that the Buddha’s formulation of dependent origination seeks to reject. The reverse process, the destruction of the dependent cessation process in reverse sequence (pratilomika), is given as the sūtra ends with:46 With the fading away (*viraktā) of ignorance (avidyā), knowledge (vidyā) arises; on the cessation of ignorance, there 28 Published Page Number: 000202 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? is the cessation of conditionings; up to; the cessation of the sheer, great mass of dissatisfaction. This is named the Dharma Discourse on Great Emptiness. Together, these show that existence is dependent arising in forward order (anuloma) from self view and desire, and also dependent cessation through the reverse order (pratiloma) of this process. In the text it is not altogether clear whether the term “emptiness” applies to both, or merely one of, these two processes. However, we have shown earlier how “empty” was used as a description of conditioned phenomena, along with impermanence, dissatisfactoriness and not self, and also, that the absence of the defilements is given as the transmundane meaning of “emptiness” elsewhere. Therefore, it seems fair to conclude that the terms “empty” and “emptiness” refer to both processes. 3.3. Seeing Dependent Origination as Dharma, as Buddha A third sectarian sūtra of note is the Śālistamba Sūtra. This is often claimed to be a Mahāyāna sūtra. While the titles of the earliest recensions in Chinese are simply “Śālistamba”, the later Sanskrit recension title is prefixed with “Madhyamaka-°”. However, as Reat clearly shows, content and structure “much of which is found scattered through the Pāli suttas”, all suggest “as the date of the Śālistamba Sūtra as a whole, 200 bce plus or minus 100 years” (Reat 1993: 4-5).47 It thus actually pre-dates the already self-identifying form of the Mahāyāna, but was later widely cited by Mahāyāna scholars, and formed an important basis for their presentation of dependent origination. Note that unlike the previous two sectarian sūtras, however, it is not associated with the SĀ or the Sarvāstivāda in general. The sūtra famously states that: “Whoever sees dependent origination sees the Dharma. Who sees the Dharma, sees the Buddha.”48 This is most likely a combination of two statements concerning “seeing the dharma” found in the Āgamas.49 The text has a four-fold structure, considering “cause” (hetu) and “condition” (pratyaya) in relation to internal and external phenomena. The predominant simile is that of the growing of a seed, hence the name “śāli-stamba” which means “rice-stalk”, which also hints at a distinction between seed as cause, and other factors as conditions. 29 Published Page Number: 000203 JCBSSL VOL. XI The difference between cause and condition is somewhat akin to that between the causal nature of “own nature” (sabhāva) and “other nature” (parabhāva) in the Theravādin Peṭak­opadesa and Nettip­pakaraṇa (see Warder in Ñāṇamoli 1982b: xvii). It is a distinction also found in the Sarvāstivāda. According to the earliest version of the sūtra,50 living beings do not go from this life to another life … but there is action (karma) and result, causally conditioned retribution without any loss. There are other paraphrases of statements found in the early sūtras. There are passages in the earlier recensions with a variety of terms as direct adjectives for dependent arising, which include “not conditioned, not abiding, unconditioned, not an object of mind, pacified, cessation, signless”.51 In the later Sanskrit, this is expanded to include “impermanence, dissatisfaction, emptiness, absence of self”.52 The terms of this common pericope are often very closely associated with emptiness, in fact, emptiness is often the first term on the list, indicating its predominance over the others.53 This means that the correct contemplation of dependent origination ensures that the practitioner will not arise various deluded views about their existence in the past or future. They will remove all the heterodox views which are based on theories of a soul (ātma­vāda), living being (sattva-°), life principle (jīva-°), person (pudgala-°), and so forth.54 The implied sense is that of the emptiness of the Self. The sūtra finally concludes with statements that one who “correctly enters into receptivity of the Dharma of dependent origination” will in fact become a fully awakened buddha.55 As this statement is found in most recensions of the text, it is at present difficult to assess whether this statement is a later addition or original. Similar to statements in the Mahā­vibhāṣa Śāstra indicate that dependent arising was the object of contemplation for the bodhisattva(s), a point which we shall examine in further detail below, via the comments of the Vaibhāṣika Master Parśva. 30 Published Page Number: 000204 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? 3.4. Abhidharma Śāstra and Mahāvibhāṣa Exegesis In addition to the obvious connection that the Sarvāstivādin sūtras place upon considering dependent origination as emptiness, their Abhidharma also makes this connection. We shall examine this from the ancient Dharma­skandha­pāda Śāstra, and the later orthodox standard, the Mahāvibhāṣa Śāstra. The Dharma­skandha­pāda is perhaps the earliest of the Northern Abhidharma literature, hailing from around 300 bce, contemporary with the Pāli Vibhaṅga. The text cites a sūtra which discusses both the Dharma law of dependent origination (緣起[法]) as well as dependently originated dharmas as phenomena (緣起已生法).56 The sūtra cited, *Pratītya­samutpāda­dharma (因緣法), appears to be our previously encountered SĀ 296, a parallel of the Pāli SN 12:20 Paccayo. Dealing with “causality” (nidāna), it suggests a text from the oldest strata of SĀ and SN, the Nidāna­saṃyuktā. This same sūtra is also cited by perhaps the oldest of all the Abhidharma literature, the Śāriputra Abhidharma, but without elaboration.57 After elucidating both the Dharma principle of the causal relationships between the phenomen­ological dharma events, and the actual dharmas themselves, it is stated that for one who knows and sees this as it really is, it is impossible that they will fall into various forms of thought such as “Did I exist in the past?”, “Will I exist in the future”, and so forth. Both the SĀ sūtra itself and also the Dharma­skandha­pāda usage of it, (but neither the Pāli sutta itself nor the Śāriputra Abhidharma Śāstra citation,) then expressly state that this is because for one who knows and sees, they have totally removed all these various conceptualizations which are “connected with the view of a self, … a living being” and so forth.58 This is thus akin to the points of view rejected in the Mahā­ śūnyatā Dharma­paryāya and the Śālistamba Sūtra (above). Together, this is a very strong indicator that vision of dependent arising, and thus not self, was the considered the factor which made one an āryan. This is supported by earlier material such as SĀ 347 and SN 12:70 Susīma, which explains that one first realizes gnosis of the “stability of Dharma” (dharma­sthitatā), and then gnosis of nirvāṇa. But despite this seemingly important statement, the Dharma­skandha­ 31 Published Page Number: 000205 JCBSSL VOL. XI pāda commentary resigns itself to merely mechanically explaining the twelve links, and remains silent as these various ātman based views. Of approximately the same historical period, the Theravādin Paṭisambhidā­magga also explains this gnosis of stability of Dhamma (dhamma­ṭṭhitatā) as knowledge of the various dependently related conditions.59 The same sūtra cited by Dharma­skandha­pāda also provides us with a number of terms for dependent origination which indicate its position as a kind of natural and immutable law. The SĀ version adds several terms over and above the Pāli, including “suchness of Dharma” (法如; *dharma­tathatā), “Dharmic nature” (法爾; *dharmatā), “factuality” (審諦; *bhūtatā), and others.60 Elsewhere in the Dharma­skandha­ pāda, a strikingly similar description which indicates a kind of eternal principle, hinting at an unconditioned nature, is also given for the āryan truths.61 A shortened list featuring only *dharma­sthititā and *dharmatā appears in the Mahāvibhāṣa. Here the authors wish to refute the Vibhajya­vādin view that dependent origination is an unconditioned (asaṃskṛta) dharma, by stating that unlike the unconditioned cessation, etc., dependent origination is still within the sphere of the past, present and future, and thus conditioned, despite it being a fixed and eternal principles.62 All this quite possibly has its roots in the Dharma­skandha­ pāda passages cited above. As shown in the Dharma­skandha­pāda, although both the Dharma of dependent arising itself and dependently originated dharmas are “of one substance”, they are still “different objects”.63 The Vaibhāṣika master Vasumitra, in presenting the statement “Whosoever sees … dependent origination sees the Dharma”, where dependent origination is explicitly given in the twelve causal links (nidāna) format, claims that “some [masters] state” that “realization through the emptiness entrance to release” is a case of “seeing dharmas but not seeing … dependent origination”; whereas “realization through the intentionless entrance to release” is “seeing both dharmas and dependent origination”.64 This would mean that the emptiness release described by Vasumitra refers to the various dharmas themselves, but not their mutual causal relationships. 32 Published Page Number: 000206 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? Moving ahead several centuries we arrive at the critical Northern Abhidharma commentarial tome, the Mahāvibhāṣa. Completed in the middle of the 2nd century ce, it certainly precedes Nāgārjuna (2nd—3rd century ce), and would doubtlessly have been a major doctrinal force across much of mainland Indian Buddhism during his lifetime. In the Mahā­vibhāṣa, the Susīma Sūtra was again cited for discussion, but now the above position of the Dharma­skandha­pāda became just one of many explanations, to be finally supplanted by an explanation more in tune with the developing Vaibhāṣika system.65 In the Mahāvibhāṣa, Parśva analyses various personality types, and states that the bodhisattvas who are “followers by desire” (*tṛṣṇānusārin) who “take the result as the entrance”, and also the “followers by views” (*dṛṣṭy­anusārin) who “take the cause as the entrance”, both “contemplate the Dharma of dependent origination”, and “based on the emptiness samādhi, they enter into certitude of perfection”.66 This is further shown by a verse which states:67 The fully awakened ones (saṃbuddha) of the three periods of time, break the poison of sorrow, they all emphasize the true Dharma (*saddharma), always abiding in the nature of Dharma (*dharmatā). While this appears to be specifically just for those on the bodhisattva path to eventual full awakening, elsewhere other statements seem to imply that āryans as a whole all “realize *dharmatā”.68 But despite this, the connection between realiza­tion of this dependent arising and the āryan stages appears to have become gradually weaker over time. Perhaps it was more the case that this sort of release through the emptiness samādhi which contemplates dependent origination as per *dharmatā as their object, was a special case for the bodhisattva. Statements in the Mahāvibhāṣa Śāstra which explicitly connect the emptiness samādhi to dependent arising are few,69 when compared with the extremely common format of this practice being contem­plation on the not self and emptiness aspects of the four truths. In the opening passages of the Mahāvibhāṣa, the connection between knowledge of not self (emptiness is not mentioned) and dependent origin­ation is made explicit, as perception of the former gives rise to the latter.70 Rather than a vision of dependent origination, insight into those fixed 33 Published Page Number: 000207 JCBSSL VOL. XI laws that indicated how specific dharmas arise and cease due to specific conditions as being the first stage of āryan status, there was a shift towards vision of just the dharmas themselves. Both Parśva’s position vis-à-vis the bodhisattva(s), which is no doubt based on the notion that it was just such a contemplation that led to Śākyamuni’s own awakening universally extended to all bodhisattva candidates, and also Vasumitra’s older source for dependent arising, hint at the antiquity of these two scholars who have a tendency towards the earlier works over the later commentarial śāstras. Both would have been approximately 100 years before Nāgārjuna’s own time (circa 2nd— 3rd centuries ce), around the period of the newly forming Mahāyāna. 3.5. Un/conditioned Status of Dependent Origination At this point, we would like to leave aside the citation of individual texts, and turn to a broader issue. Up to this point, we have seen that the Abhidharma method­ology of classification of dharma(s) as either conditioned (saṃskṛta) or unconditioned (asaṃskṛta) dharma gradually became more significant as time progressed. While this “dharma­vāda” approach should be more narrowly confined to dharmas (plural) as phenomena, it appears that the use of dharma (singular) as a law or principle, was unable to escape such analysis. It thus came to be that the question of whether or not the very principle of dependent origination (pratītya­samutpāda­dharmaḥ) was conditioned or unconditioned, and not just those things that were dependently originated (pratītya­ samutpanna­dharmāni), came about. As we have already seen in the Nikāyas and Āgamas themselves, there were a number of terms commonly taken as synonyms for dependent arising, such as “stability of Dharma” (dharma­sthititā) and “suchness” (tathatā). These were also drawn into the range of this debate, though as we shall see below, they were not necessarily seen as exact equivalents. Drawing from the excellent study of Bareau on the “jungle of views” that makes up sectarian Buddhism, we may tabulate the attribution of conditioned or uncondi­tioned status to these three notions—dependent origination, the stability of Dharma, and suchness—as advocated by various schools (from Bareau 2005: 287):71 34 Published Page Number: 000208 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? Conditioned (saṃskṛta) Dependent Origination Stability of Suchness Dharma Vātsīputriya ✓ (#38) Theravāda ✓ (#55) ✓ (#21) ✓ (#186) ✓ (#6, #28) ✓ (#6) ✓ (#6) Sarvāstivāda Śāriputra Abhidharma Mahāsāṃghika ✓ (#43) Mahīśāsaka ✓ (#19) Uttarāpathaka Unconditioned (asaṃskṛta) Dependent Origination ✓ (#10) ✓ (#10) ✓ (#43) ✓ (#43) ✓ (#19) Stability of Suchness Dharma ✓ (#10) ✓ (#19) ✓ ✓ (#32*) (#32*) Dharmagupta ✓ (#13) Pūrvaśaila ✓ (#9) Vibhajyavāda ✓ (#8) On one side, there were the schools which tended towards considering dependent origination, etc., as conditioned: These are notably the earlier Sthavira schools with Ābhidharmika tendencies, such as the Theravāda, Vātsīputriya and Sarvāstivāda. We would probably expect to see the Śāriputra Abhidharma Śāstra among this group, but out of the three they only considered suchness as conditioned. For the conditioned status of the stability of Dharma and suchness, there was also support from the Mahāsāṃghikas and Mahīśāsakas, and a qualified affirmation from the Uttarāpathakas. On the other side, there were also those who inclined to an interpretation of these as unconditioned: The Mahāsāṃghikas and Śāriputra Abhidharma Śāstra had a shared list of nine unconditioned, including: “8. The self-nature of the members of conditioned production (*pratītya­ samutpādāṇga­svabhāva)”; and “9. The self-nature of the factors of the Path (*mārgāṇga­svabhā).”72 However, according to Bareau’s sources at least, both the Mahā­sāṃghikas and the Śāriputra Abhi­dharma Śāstra otherwise considered that suchness itself is conditioned. Were there different forms of suchness, such that some were conditioned 35 Published Page Number: 000209 JCBSSL VOL. XI and others not? The Mahīśāsaka also had nine unconditioned, on a slightly different list, although also including: “8. The suchness of the Path (mārga­tathatā)”; and “9. The suchness of dependent origination (pratītya­samutpāda­tathatā).”73 The Dharma­guptakas upheld unconditioned status regards dependent origination alone, but there is no mention of their position vis-à-vis stability of Dharma or suchness.74 The Pūrvaśailas agreed, simply on the basis of a sūtra, which was most probably their equivalent of SN 12:20 / SĀ 296.75 They also considered that the four āryan truths themselves were unconditioned, for basically the same reasons.76 The otherwise unelucidated agreement of so-called Vibhajyavādins,77 makes them appear to be any group other than the three early Sthavira Abhidharma schools. In addition to the above points, we also find some other related doctrinal positions of the Sthavira traditions. The Katthāvatthu states a position of the Theravādins against the Andhakas, namely that the former consider that emptiness (along with the signless and intentionless) “is not included in the aggregate of the mental formations”.78 On the grounds that whatever is a formation (saṃskāra) is also conditioned (saṃskṛta), this would seem to mean that the Andhakas considered even emptiness itself to be conditioned. The category of the unconditioned has always been a standard of Ābhidharmika analysis based on the early sūtras. However, the very notion of anything being “unconditioned” has always been problematic for Buddhists who were loathe to run up against their core doctrine of not self or non-self, ie. emptiness. Here we have seen how it related to dependent arising, and its synonyms. But it obviously also extended to the notion of emptiness, as some form of equivalence had already been established. This difficulty can be seen, for example, in Nāgārjuna’s under­standing of “the emptiness of emptiness”, the argument that to turn emptiness itself into a reified—unconditioned, not dependent— phenomena, is to make perhaps the greatest mistake of all. Much of the discourse on “the emptiness of emptiness” has been an ontological one, which has often failed to look to its earlier precedents. It appears to have its roots in seeing as empty the very insight contemplation which perceives the emptiness of phenomena. This is quite a different matter altogether, but unfortunately beyond our scope here. 36 Published Page Number: 000210 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? 3.6. Summary The first two subjects for discussion of emptiness and dependent origination in mainstream sectarian Buddhism involved citing two Saṃyuktāgama sūtras, the Paramārtha­śūnyatā (Section §3.1) and Mahā­śūnyatā (Section §3.2) sūtras. Both of these two sūtras not only directly related emptiness — described as “ultimate” or “great” emptiness — to dependent origination, in the simple format “When this exists, that exists”, etc., or in the later standardized twelve limb format. Further­more, the two texts raise the not ions of “neither coming nor going” and “neither identity nor plurality”, respectively, to describe dependent arising. These two statements would appear to be sources for four of the famous “eight negations” that make up the opening verses of Nāgārjuna’s Kārikās. This fact has been noted by Yìnshùn in his Investigations, Chp. 4.2, where he also provides early (and sectarian) sources for the other two sets of negations, “neither arising nor ceasing” and “neither eternalism nor nihilism” (Yìnshùn 1985: 210f). We have omitted detailed discussion of these sources here, as while highly implying a relation to emptiness, they do not mention it explicitly.79 Yìnshùn concludes that “We have no doubts that teaching dependent origination based on the eight negations has its source in the Samyukta Āgama sūtras” (1985: 210f). Most of these texts have Pāli equiva­lents, though EĀ parallels are available. As such, they are seldom if ever raised in the usual discourse on emptiness which relies excessively on Pāli material. While sūtras, they still most likely represent notions from the early sectarian period, rather than any so-called primitive or early Buddhism. The Śālistamba Sūtra provides further early sectarian material from other than Sarvāsti­vādin sources (in Section §3.3). We know that this text was later fully adopted by the Madhyamaka and other Mahāyāna scholars. Similar to “Ultimate Emptiness” and “Great Emptiness”, much attention is given to describing dependent origination in terms that avoid the extremes that result from views of a Self, or other synonymous entity. The system of causality distinguishing primary cause from supportive conditions began to develop, whereas earlier these terms were for all purposes simply synonymous. This development also suggests just how early the systems that appear as Nāgārjuna’s 37 Published Page Number: 000211 JCBSSL VOL. XI model for analysis in Chp. 1 of the Kārikās may be. This system may thus not necessarily be restricted to the Vaibhāṣikas. By the actual Abhidharma śāstras themselves, and the later fully developed Vaibhāṣika system, we see the continued reference and citation to the Āgamas, albeit interpreting them through the Abhidharma method (Section §3.4). In particular, the analysis into real existent dharmas and designations (prajñapti), and their being condi­tioned or unconditioned (Section §3.5). This basic method posed problems, however, when inter­preting both the principle (dharma—singular) or dependent origination and those phenomena (dharmas—plural) which were dependently originated. Various schools were split along lines where the Ābhidharmikas generally considered both conditioned, but others, most notably the Mahāsāṃ­ghikas and those they influenced, considering the principle itself as unconditioned. 4. “Emptiness = Dependent Origination” — Nāgārjuna’s Innovation? In Section §1, we began our examination from the broad context of the academic discourse on emptiness. This fairly standard and often paradigmatic scholastic narrative of the teaching of emptiness usually starts from Nāgārjuna’s Mahāyāna Madhyamaka which it extols as a novel and creative if not revolutionary teaching, only then moving back through mainstream sectarian “Hīnayāna” schools, and finally to early Buddhism. But the earlier historical stages are often only there to explain the Madhyamaka, giving the approach something of a teleological flavor. But besides this difficulty, source biases and anachronistic arguments are also highly problematic. The source biases are mainly the heavy if not sole use of Pāli and Sanskrit material, ignoring translations of important Indic texts preserved only in Chinese translation. The anachronistic arguments are that lines of doctrinal development often use post-Nāgārjunian classic Theravādin and Sarvāstivādin cum Sautrāntika commentarial literature to represent the targets of the Madhyamaka critique. That Madhyamaka critique is, in a word, emptiness (śūnyatā). Or, in a verse, that emptiness is dependent origination, designation on a basis, and the middle way (as per Mūla­ madhyamaka­kārikās, v. 24:18), and also the opening verses of praise of the Kārikās, upon which the academic discourse places such a heavy 38 Published Page Number: 000212 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? weight. While the discourse of emptiness is a very broad topic, we have chosen to focus upon this latter particular aspect of it. That is, we have asked: is Nāgārjuna’s direct relationship of equivalence or near equivalence between emptiness and dependent origination, an innovation on his part? In the light of the source biases and anachronistic arguments mentioned above, we have chosen a relatively simple approach here. The stronger that equivalences or relationships between emptiness and dependent origination can be shown in the literature prior to Nāgārjuna (circa 2nd to 3rd century ce), the weaker the argument that this is an innovation on his part, is clear. Thus we have examined the material utilizing two heuristically functional historical periods of “early” Buddhism (up to circa 3rd century bce), and “mainstream sectarian” Buddhism (3rd century bce to mid 2nd century ce), respectively. Now, to what degree has a pre-Nāgārjunian relationship between these two doctrines been shown? In short, a direct and explicit connection between emptiness and dependent origin­ation has been present in Buddhist literature from the early period. While this may not be particularly strong in the Pāli texts, it can be seen more clearly in the Chinese translations of the Āgamas, in particular the Sarvāstivādin Saṃyuktāgama and (probably) Mahāsāṃghika Ekottarāgama. The connection also continues into the sectarian period, as can be seen from examination of translations of the Sarvāsti­vādin Abhidharma śāstras and commentarial literature. Such use of early and mid-period Indian Buddhist literature outside of the usual Pāli and Sanskrit recensions should be standard practice for Buddhist studies. We can no longer justify the source biases that often occur when such alternate material is ignored. The relationship between the notions of emptiness and dependent arising found in the early texts that we have examined here works on a number of levels: 1. As demonstrated in Sections §2.1 and §2.6, the first is where emptiness refers to a lack of a “Self” or “what pertains to Self”, the usual gloss of the discourse for early teachings on emptiness. The process of the arising and coming into being of phenomena, saṃsāric becoming, is one which takes place with recourse to such a Self entity. This finding is quite to be expected, a core thread 39 Published Page Number: 000213 JCBSSL VOL. XI that winds from most likely the Buddha himself through the texts and commentaries, to Nāgārjuna, and beyond. 2. But in addition to this is the sense wherein emptiness refers to the state of nirvāṇa itself, the absence of affliction, as shown in Sections §2.3 and §2.3. The complementary aspect is not dependent arising, but, if we may coin a neo­logism, dependent cessation. 3. A position tending toward universalization of the aforementioned two levels begins to emerge, as per Sections §2.4 and §2.5. Both the principle dependent arising and the state of appeasement, two otherwise radically different sense of “emptiness”, are “profound”. Thus this becomes almost a catch phrase for the totality of the Buddha’s teachings, a notion which develops into the idea of “dharma seals” or “dhamma characteristics” during the end phase of the early texts and the start of the mainstream schools. That “all phenomena are empty”, be they conditioned or unconditioned, saṃsāra or nirvāṇa, is a hall mark of not just the phenomena (dharmas) themselves, but the principle (Dharma) that connects them, and the teaching (Dharma) that elucidates both aspects. The mainstream period continues this trend, where Abhidharma methodology such as analysis into conditioned and unconditioned, conventional and ultimate truths, and so forth, further clarifies the ever more significant role of “emptiness”. 1. First, in the Saṃyuktāgama texts on “Ultimate Emptiness” (Paramārtha­śūnyatā) and “Great Emptiness” (Mahā­śūnyatā), there is the equation wherein emptiness is explicitly described in terms of both aspects of dependent origination and cessation (in Sections §3.1 and §3.2). The Sarvāstivādins, at least, elevated emptiness to have a similar status as imperma­nence, dissatisfaction and absence of Self. By making it one of the sixteen fundamental aspects (ākāra), a characteristic shared by all phenomena (samānya­lakṣaṇa), this was further strengthened. 2. Outside this school, the Śalistamba Sūtra (see Section §3.3), which was later to be adopted by the Madhyamaka Mahāyāna, further pushed dependent origination to the fore by identifying it as a universal truth. This occurred both abstractly, as the “Dharma”, and anthropo­morphi­cally as the “Buddha”. 3. While the Abhidharma had its own methodology, it was still bound to a greater or lesser degree, to conform to the early texts. When faced with early sūtra (and probably vinaya) that described the Bodhisattva Gautama’s awakening in terms of insight into dependent origination, the notion that this was a particular practice for cultivation for those 40 Published Page Number: 000214 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? on the bodhisattva path (as opposed to śrāvakas and pratyeka­buddhas), became more formalized (Section §3.4). An incipient Mahāyāna tradition, drawing much from the institutionally large and philosophical powerful traditions such as the Sarvāstivāda, would no doubt be influenced by such doctrinal lineages. The influence need not be one of mere acceptance and continuation thereof, and could just as likely be one of dialectical opposition and transformation. As can now be clearly seen, the connection between emptiness and dependent origination is not an innovation or creation of Nāgārjuna. In the still narrower sense of the relationship being an equation of the two, namely that emptiness is just dependent origination itself, this has also been well established before Nāgārjuna’s time. In addition, there are numerous other doctrinal positions that associate the two one way or another. But while we have shown in the introduction of this essay how influential the notion of Nāgārjuna’s equation between emptiness and dependent origination is on academic narrative, we have not detailed what Nāgārjuna actually meant by this. But this has been amply discussed elsewhere, so we need not cover old ground here. What we have aimed to do is provide earlier material, letting the texts speaks for themselves, as they say. Although this textual analysis demonstrates that relating and connecting emptiness and dependent origination is not an innovation of Nāgārjuna, no doubt new doctrinal interpretations of these core ideas have been made by him. The next task would then be to compare the material presented here with that of Nāgārjuna to further understand what his specific developments were exactly. 41 Published Page Number: 000215 JCBSSL VOL. XI Bibliography Primary Sources T001 T026 T099 T100 T125 《長阿含經》 Cháng āhánjīng, Dīrghāgama (Dharmaguptaka) 《中阿含經》 Zhōng āhánjīng, Madhyamāgama (Sarvāstivādin) 《雜阿含經》 Zá āhánjīng, Saṃyuktāgama (Sarvāstivādin) 《別譯雜阿含經》 Bíeyì Za āhánjīng, Saṃyuktāgama (Kaśyapīya) 《增壹阿含經》 Zēngyī āhánjīng, Ekottarāgama (Late Mahāsaṃghika) T1509 《大智度論》 Dàzhìdù lùn, Mahāprajñāpāramitā Upadeśa T1537 《阿毘達磨法蘊足論》 Āpídámó făyǜnzú lùn, Dharma­skandha­pāda Śāstra T1544 《阿毘達磨發智論》 Āpídámó făzhì lùn, Abhidharma Jñānaprasthāna Śāstra T1545 《阿毘達磨大毘婆沙論》 Āpídámó Dàpípōshā lùn, Abhidharma Mahāvibhāṣā Śāstra T1548 《舍利弗阿毘曇》 Shělìfú āpítán, Śāriputra Abhidharma Śāstra T1558 《阿毘達磨俱舍論》 Āpídámó Jǜshě lùn, Abhidharma Kośa Bhāṣyam T1564 《中論》 Zhōng lùn, Mūlamadhyamaka Kārikā Bodhi, (1984). The Great Discourse on Causation: The Mahānidāna Sutta and its Commentaries, Buddhist Publication Society: Kandy. Bodhi, (2000). The Connected Discourses of the Buddha: A Translation of the Samyutta Nikāya, Wisdom: Boston. Ireland, J D (1997). The Udāna & The Itivuttaka, Buddhist Publication Society: Kandy. Lamotte, E, Vol I (1944), Vol II (1949), Vol III (1970), Vol IV (1976), Vol V (1980). Le Traite de la Grande Vertu de Sagesse, Institut Orientaliste: Louvain-la-Neuve. Lamotte, E, trs. Boin, S (1976). The Teaching of Vimalakīrti, (translation of Vimalakīrti-nirdeśa), Pāli Text Society. Original title: L’Enseignement de Vimalakīrti, 1962. Lamotte, E, trs. Karma Migme Chodron (2001). Le Traite de la Grande Vertu de Sagesse, Vol I, Vol II, Vol III, Vol IV, Vol V, English translation (2001). Ñāṇamoli, (1956). The Path of Purification, (translation of the Visuddhimagga). Pāli Text Society. Ñāṇamoli, (1977). The Guide, (translation of the Nettippakaraṇa). Pāli Text Society. Ñāṇamoli, (1979). “The Exposition of Non-Conflict (translation of the Araṇavibhaṅga Sutta)”, in The Wheel Publication No. 269, Buddhist Publication Society: Kandy. 42 Published Page Number: 000216 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? Ñāṇamoli, (1981). “Three Cardinal Discourses of the Buddha, (translations of the Dhamma-cakka- ppavattana Sutta, the Anatta-lakkhana Sutta, and the Āditta-pariyāya Sutta)”, in The Wheel Publication No. 17, Buddhist Publication Society: Kandy. Ñāṇamoli, (1982a). “The Greater Discourse on Voidness: Mahāsuññatā Sutta”, in The Wheel Publication No. 87, Buddhist Publication Society: Kandy. Ñāṇamoli, (1982b). The Path of Discrimination, (translation of the Paṭisambhidāmagga). Pāli Text Society. Ñāṇamoli & Bodhi, (1995). The Middle Length Discourses of the Buddha: A Translation of the Majjhima Nikāya, Wisdom: Boston. Norman, K R (1984). The Rhinoceros Horn, and Other Early Buddhist Poems, Pāli Text Society: Oxford. Reat, N R (1993). The Śālistamba Sūtra, Motilal Banarsidass: Delhi. Vaidya, P L (1961). Śālistamba Sūtra, Mithila Institute: Darbhanga. Vipassanā Research Institute (1995). Chaṭṭha Saṅgāyana Tipiṭaka version 4.0. (Pāli Canon.) Walshe, M (1987). The Long Discourses of the Buddha: A Translation of the Dīgha Nikāya, Wisdom: Boston. Secondary Sources Akanuma, Chizen 赤沼智善 (1929). The Comparative Catalogue of Chinese Āgamas and & Pāli Nikāyas, Hajinkaku-Shobō: Nagoya. Anālayo & Bucknell, R (2006). “Correspondence Table for Parallels to the Discourses of Majjhima Nikāya: Toward a Revision of Akanuma’s Comparative Catalogue”, pp. 215-243, in Journal of Buddhist Studies, Vol. 4, 2006. Centre for Buddhist Studies: Śrī Lanka. Anālayo, (2008). “Reflections on Comparative Āgama Studies”, pp. 3-21, in Chung-Hwa Buddhist Journal, Vol. 21, 2008. Chung-Hwa Institute of Buddhist Studies: Taibei. Anālayo, (2009). “The Āneñjasappāya-sutta and its Parallels on Imperturbability and on the Contribution of Insight to the Development of Tranquility”, pp. 177-195, in Buddhist Studies Review, Vol. 26.2, 2009. Journal of the UK Association for Buddhist Studies. Bareau, A (2005), tr. Ani Migme Chödrön. The Sects of the Small Vehicle. PDF accessed via personal correspondence. Original title: Les Sectes Bouddhiques du Petit Véhicule (1955). Choong, Mun Keat (1999). The Notion of Emptiness in Early Buddhism, Motilal Banarsidass: Delhi. Conze, E (1962). Buddhist Thought in India, George, Allen and Unwin: London. Dhammajothi, M (2008). The Concept of Emptiness in Pāli Literature, Corporate Body of the Buddha Educational Foundation: Taibei. 43 Published Page Number: 000217 JCBSSL VOL. XI Dhammajoti, K L (2007). Sarvāstivāda Abhidharma, University of Hong Kong: Hong Kong. Garfield, J L (1994). “Dependent arising and the emptiness of emptiness: Why did Nāgārjuna start with causation?”, pp. 219-250, in Philosophy East and West, Vol. 44, No. 2, 1994. University of Hawaii Press. Garfield, J L (2002). Empty Words: Buddhist Philosophy and Cross-Cultural Interpretation, Oxford University Press: Oxford. Gombrich, R (1996). How Buddhism Began: The Conditined Genesis of the Early Teachings, Munshiram Manoharlal: Delhi. Gómez, L O (1976). “Proto-Mādhyamika in the Pāli Canon” pp. 137-165, in Philosophy East and West, Vol. 26, No. 2 (April 1976), Philosophy East and West. Jurewicz, J (2000). “Playing with Fire: The pratītyasamutpada from the Perspective of Vedic Thought”, pp. 77-104, in The Journal of the Pali Text Society, Vol. XXVI, 2000. Antony Rowe: Chippenham. Kalupahana, D J (1975). Causality: The Central Philosophy of Buddhism, University Press of Hawaii: Honolulu. Kalupahana, D J (1986). Nāgārjuna: The Philosophy of the Middle Way, SUNY Press: New York. Karunadasa, Y (2005). “The Unanswered Questions: Why Are They Unanswered? A Re-examination of the Textual Data” pp. 85-111 in the Journal of Buddhist Studies, Vol. III, 2005, Centre for Buddhist Studies, Sri Lanka. Karunadasa, Y (2010). The Theravāda Abhidhamma: its Inquiry into the Nature of Conditioned Reality, Centre of Buddhist Studies, University of Hong Kong. Karunaratne, W S (1988). The Theory of Causality, PhD thesis, 1956. Law, B C (1933). A History of Pāli Literature, Indica: Varanasi. Lĭ, Rùnshēng (1999a). 中論(上) (Zhōng Lùn (Shàng)) [Middle Treatise I], 全 佛: 新店 Qüán Fó: Xīndiàn. Lĭ, Rùnshēng (1999b). 中論(下) (Zhōng Lùn (Xìa)) [Middle Treatise II], 全 佛: 新店 Qüán Fó: Xīndiàn. Murti, T R V (1955). The Central Philosophy of Buddhism: A Study of Mādhyamika System, Munshiram Manoharlal: New Delhi. Nagao, Gadjin (1989). The Foundational Standpoint of Mādhyamika Philosophy, State University of New York Press: New York. Nagao, Gadjin (1991). Mādhyamika and Yogācāra, State University of New York Press: New York. Ramanan, V K (1966). Nāgārjuna’s Philosophy, Motilal Banarsidass: Delhi. Rhys Davids, T W & Stede, W, eds (1921). The Pali Text Society’s Pali– English Dictionary. Chipstead: Pali Text Society. Santi, Bhikkhu (2006). Three Emptiness Sūtras, Translated in Pāli and English. Accessed 2008: http://www.library.websangha.org/ earlybuddhism/Three_Emptiness_Sutras.pdf 44 Published Page Number: 000218 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? Shulman, E (2008). “Early meanings of dependent-origination”, Journal of Indian Philosophy Vol. 36, No. 2. Seyfort Ruegg, D (2004). “Aspects of the Study of the (Earlier) Indian Mahāyāna”, pp. 3-62, in Journal of the International Association of Buddhist Studies, Vol. 27, Number 1, 2004. International Association of Buddhist Studies: Vienna. Seyfort Ruegg, D (2010). The Buddhist Philosophy of the Middle: Essays on Indian and Tibetan Madhyamaka, Wisdom: Massachusetts. Siderits, M & Katsura, S (2013). Nāgārjuna’s Middle Way, Wisdom: Massachusetts. Skilling, P (1994). Mahāsūtras Vol. I. Pāli Text Society: Oxford. Skilling, P (1997). Mahāsūtras Vol. II. Pāli Text Society: Oxford. Stcherbatsky, T (1968). The Conception of Buddhist Nirvāṇa, with Sanskrit Text of Madhyamaka-Kārikā, Motilal Banarsidass Publishing. Tachikawa, Musashi (1993). “The Mādhyamika Tradition”, pp. 188-202, in Buddhist Spirituality: Indian, Southeast Asian, Tibetan and Early Chinese, Vol I, Yoshinori, Takeuchi 武內義範, ed., SCM Press. Tachikawa, Musashi (1997). An Introduction to the Philosophy of Nāgārjuna, Motilal Banarsidass: New Delhi. Vélez de Cea, A (2005). “Emptiness in the Pāli Suttas and the Question of Nāgārjuna’s Orthodoxy” pp. 507-528 in Philosophy East and West, Vol. 55, No. 4 (October 2005), Philosophy East and West. Warder, A K (1970). Indian Buddhism, Motilal Banarsidass: Delhi. Warder, A K (1998). A Course in Indian Philosophy, Motilal Banarsidass: Delhi. Westerhoff, J (2009). Nāgārjuna’s Madhyamaka: A Philosophical Introduction, Oxford University Press: Oxford. Willemen, C, Dessein, B & Cox, C (1998). Sarvāstivāda Buddhist Scholasticism, Brill: Leiden. Yìnshùn, 印順 (1944). 性空學探源 (Xìngkōngxǘe tànyüán) [Investigations into the Origins of Emptiness Teachings], 正聞: 新竹 Zhèngwén: Xīnzhú. Yìnshùn, 印順 (1971). 原始佛教聖典之集成 (Yüánshĭ fójiào shèngdiăn zhī jíchéng) [Compilation of the Early Buddhist Canon], 正聞: 新竹 Zhèngwén: Xīnzhú. Yìnshùn, 印順 (1980). 初期大乘佛教之起源與開展 (Chūqĭ dàshèng fójiào zhī qĭyüán yü kāizhăn) [The Origins and Development of Early Mahāyāna Buddhism], 正聞: 新竹 Zhèngwén: Xīnzhú. Yìnshùn, 印順 (1985). 空之探究 (Kōng zhī tànjìu) [Investigations into Śūnyatā], 正聞: 新竹 Zhèngwén: Xīnzhú. 45 Published Page Number: 000219 JCBSSL VOL. XI Notes 1 2 3 4 5 6 7 8 Full disclosure: The author here is presently working on a full English translation of this work, to be titled An Investigation into Emptiness, to be published by Noble Path Publishing, USA. Refer also Yìnshùn (1971: 89-102). See Yìnshùn (1971: 755); though Mizuno (2003: 561fff) is less sure about its sectarian affiliation. The usual generic formula is given as that in Ud 1:1-3: “Iti imasmiṃ sati idaṃ hoti, imassuppādā idaṃ uppajjati, imasmiṃ asati idaṃ na hoti, imassa nirodhā idaṃ nirujjhati”; Ireland (1997: 13ff); also SN ii 28, 65, 70, 78f, 95f, v 388; Bodhi (2000: 552, 575, 579, 585f, 596f); and MN 38 i 262f, ii 32, iii 63; Ñāṇamoli & Bodhi (1995: 355f, 655, 927); AN v 184. Bodhi (2000: 517) states: “Dependent origination is one of the central teachings of early Buddhism, so vital to the teaching as a whole that the Buddha is quoted elsewhere as saying, ‘One who sees dependent origination see the Dhamma, and one who sees the Dhamma sees dependent origination’ (MN i 190-191)”; “Vuttaṃ kho panetaṃ bhagavatā—‘yo paṭicca­samup­pādaṃ passati so dhammaṃ passati; yo dhammaṃ passati so paṭiccasamuppādaṃ passatī’ ti.” See Kalupahana (1975), Causality: The Central Philosophy of Buddhism, for an in depth examination of this teaching in early Buddhism. Such as those seen in SN 12:20, ii 25: “Katamo ca, bhikkhave, paṭiccasamuppādo? … Uppādā vā tathāgatānaṃ anuppādā vā tathāgatānaṃ, ṭhitāva sā dhātu dhammaṭṭhitatā dhamma­niyāmatā idappaccayatā…”; and SN 6:2, i 140 for “dhammatā” as the content realized by all the Buddhas, which they in turn reverence; Bodhi (2000: 550f, 235). In equivalent SĀ 296 《雜阿含經》卷 12：「云何為因緣法。謂此有故彼有。謂緣無明行。緣行識。…若佛出世。 若未出世。此法常住。法住法界。…此等諸法。法住．法空．法如．法爾。 法不離如。法不異如。審諦真實．不顛倒。如是隨順緣起。是名緣生法。… 」(CBETA, T02, no. 99, p. 84, b14-26), we see the addition of the generalized formula at the start (cf. Ud 1:1-3). SN 20:7 Āṇi ii 267: “… ye te suttantā tathāgata­bhāsitā gambhīrā gambhiratthā lokuttarā suññatā­paṭisaṃyuttā …”; Bodhi (2000: 709). Also SN 55:53 Dhammadinna v 407; Bodhi (2000: 1833f). Parallel in SĀ 1258 《雜阿含經》 卷47：「如來所說修多羅甚深明照空相應隨順緣起法」(CBETA, T02, no. 99, p. 345, b13-14), and also SĀ 293 《雜阿含經》卷12 (CBETA, T02, no. 99, p. 83, c4-6); which has no Pāli equivalent. SN 6:1 i 136: “Adhigato kho myāyaṃ dhammo gambhīro duddaso duranubodho santo paṇīto atakkāvacaro nipuṇo paṇḍita­vedanīyo, ālayarāmā kho panāyaṃ pajā ālayaratā ālaya­sammuditā ālayarāmāya kho pana pajāya ālayaratāya ālaya­sammuditāya duddasaṃ idaṃ ṭhānaṃ yadidaṃ idap­paccayatā paṭicca­ samuppādo. Idampi kho ṭhānaṃ yadidaṃ sabba­saṅkhāra­samatho sabbūpadhi­ paṭinissaggo taṇhāk­khayo virāgo nirodho nibbāṇaṃ.”; Bodhi (2000: 231). A largely identical Pāli passage in MN 26 i 167 and MN 72 i 487; see Ñāṇamoli & Bodhi (1995: 260, 593) SN 6:1, paralleled by EĀ 19:1 《增壹阿含經》卷10〈19 勸請品〉 (CBETA, 46 Published Page Number: 000220 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? 9 10 11 12 13 14 15 T02, no. 125, p. 593, a25-27), where dependent origination is not mentioned, however. MN 26 and parallel MĀ 204 which lacks this passage altogether. MN 72 and parallel SĀ 962 which merely has 《雜阿含經》卷34：「是則不然。甚 深廣大。無量無數永滅。」(CBETA, T02, no. 99, p. 246, a10): “… it cannot be fathomed, it is most deep, vast, immeasurable, incalculable, eternal cessation”, refering to the extinction of the fire—nirvāṇa.” SĀ 293 《雜阿含經》卷12：「此甚深處。所謂緣起。倍復甚深難見。所 謂一切取離．愛盡．無欲．寂滅．涅槃。如此二法。謂有為．無為。」 (CBETA, T02, no. 99, p. 83, c13-15) No equivalent in SN, though possibly around SN 12 given the relation of other adjacent sūtra / sutta. Similar to both SN 6:1, above, and also the opening of DN 15 Mahānidāna Sutta, ii 55. DN 15 ii 55: “… “acchariyaṃ bhante, abbhutaṃ bhante, yāvagambhīro cāyaṃ bhante, paṭicca­samuppādo gambhīravabhāso ca. Atha ca pana me uttānakuttānako viya khāyatī”ti. Mā hevaṃ ānanda avaca, mā hevaṃ ānanda avaca, gambhīro cāyaṃ ānanda paṭiccasamuppādo gambhīrāvabhāso ca.”; see Walshe (1987: 223); Bodhi (1984: 45). Same as MĀ 97 Mahā­nidāna Sūtra 《 中阿含經》卷24〈4 因品〉：「… 此緣起甚奇。極甚深。明亦甚深。然我觀 見至淺至淺。世尊告曰。阿難。汝莫作是念。此緣起至淺至淺。所以者何。 此緣起極甚深。明亦甚深。」(CBETA, T01, no. 26, p. 578, b13-17). Also DĀ 13 (大緣方便經) *Mahā­nidāna­upāya Sūtra and DN 15 Mahā­nidāna Sutta, are largely similar. Note that for Ānanda’s comments, Pāli has “clear as clear can be”, whereas MĀ has “shallow as shallow can be”, ie. opposite of “profound”. Udāna 1:1-3 1f: “imasmiṃ sati idaṃ hoti, imassūppādā idaṃ uppajjati, imasmiṃ asati idaṃ na hoti, imassa nirodhā idaṃ nirujjhati.”; English in Ireland (1997: 13f). See also SĀ sūtras 296-302 《雜阿含經》卷12 (CBETA, T02, no. 99, p. 84, b12 – p. 86, b23); etc. SN 12:12 ii 13f; Bodhi (2000: 541f). SĀ 372 《雜阿含經》卷15 (CBETA, T02, no. 99, p. 102, a12-b17). SĀ 262 《雜阿含經》卷10 (CBETA, T02, no. 99, p. 66, c25-p. 67, a8). SN 12:15 Kaccāna­gotta Sutta; SN ii 17; Bodhi (2000: 544). Or SĀ 301 《雜阿含經》卷 12 (CBETA, T02, no. 99, p. 85, c17-p. 86, a3). On causality, see suttas SN 12:15, 17, 18, 24, 25, 26 and 67; and for other extreme views, SN 12:46, 47 and 48. For “nibbāna” in SN 38:1f iv 251f: “Yo kho āvuso rāgakkhayo dosakkhayo mohakkhayo idaṃ vuccati nibbānanti”; Bodhi (2000: 1294f). Parallel in SĀ 490 《雜阿含經》卷18：「涅槃者。貪欲永盡。瞋恚永盡。愚癡永盡。一切諸 煩惱永盡。是名涅槃。」(CBETA, T02, no. 99, p. 126, b3-4). As a definition of the “unconditioned” (asaṅkhata) in SN 43 Asaṅkhata Saṃyutta iv 359ff: “Katamañca bhikkhave asaṅkhataṃ: yo bhikkhave rāgakkhayo dosakkhayo mohakkhayo, idaṃ vuccati bhikkhave asaṅkhataṃ”; Bodhi (2000: 1372ff). Parallel in SĀ 890 《雜阿含經》卷31：「云何無為法。謂貪欲永盡。瞋恚． 愚癡永盡。一切煩惱永盡。是無為法。」(CBETA, T02, no. 99, p. 224, b1-3). Also as a definition for the fruition of an arhat in SĀ 1129 《雜阿含經》卷41： 「何等為阿羅漢果。若彼貪欲永盡。瞋恚永盡。愚癡永盡。一切煩惱永盡。 是名阿羅漢果。」(CBETA, T02, no. 99, p. 299, a5-7); the sūtra other paralleled by SN 55:55-58 v 410-411; Bodhi (2000: 1836-7), but the Pāli lacks this phrase. 47 Published Page Number: 000221 JCBSSL VOL. XI 16 SĀ 262 《雜阿含經》卷10：「一切諸行空寂．不可得．愛盡．離欲．涅 槃。」(CBETA, T02, no. 99, p. 66, b17-18; c10-11) Unlike the SN version, the SĀ states that this occurred “not long after the [Buddha’s] parinirvāṇa”. Pāli is SN 22:90 Channa iii 133 (see below). 17 SN 22:90 Channa iii 133: “Atha ca pana me sabba­saṃkhāra samathe sabbūpadhi­ paṭi­nissagge taṇha­kkhaye virāge nirodhe nibbāne cittaṃ na pakkhandati nappasīdati na santiṭṭhati nādhi­muccati paritassanā upādānaṃ uppajjati, paccudāvattati mānasaṃ, atha kho carahi me attāti, na kho panevaṃ dhammaṃ passato hoti “ko nu kho me tathā dhammaṃ deseyya yathāhaṃ dhammaṃ passeyya”nti.” Refer Bodhi (2000: 946). 18 Yogācāra­bhūmi Śāstra 《瑜伽師地論》卷83：「所言空者。謂離一切煩惱 等故。無所得者。謂離一切所有相故。言愛盡者。謂不希求未來事故。 言離欲者。謂無現在受用憙樂故。所言滅者。謂餘煩惱斷故。言涅槃者。 謂無餘依故。」(CBETA, T30, no. 1579, p. 766, a22-26): “The term ‘empty’ means forsaking all the defilements. The term ‘non-obtainment’ means being separated from the signs of any existence. The term ‘destruction of craving’ means not seeking anything in the future. The term ‘forsaking of desire’ means not delighting in present experiences. The term ‘cessation’ means abandoning remaining defilements. The term ‘nirvāṇa’ means no remaining substratum.” 19 SN 41:7 Godatta; SN iv 295f; Bodhi (2000: 1325f), which has the Citta and Godatta; parallel is SĀ sūtra 567 (T02, no. 99, p. 149, c6 – p. 150, a15), which has instead Citra and Nāgadatta. Also MN 43 Mahāvedalla Sutta, MN i 292; Ñāṇamoli & Bodhi (1995: 387f); with Mahākoṭṭhita and Sāriputta; parallel is AN 6:175 Koṭṭhita; and MĀ 211; paralleled by SĀ 251. The SN and both SĀ versions are the shortest, while the MN and MĀ versions have become more extensive exegeses. 20 SN 40:9 Animitta; PTS SN iv 268; Bodhi (2000: 1308). Here, it is also given as “non-attention to all signs”. The only other Pāli uses of “ceto­samādhi”seems not directly related to any of the four immeasurables, such as DN Brahmajāla i 13ff; DN Mahāpadāna ii 100; DN Pāthika iii 30; AN ii 54, iii 51; etc. 21 MN 43 Mahāvedalla; PTS MN i 298; = Ñāṇamoli & Bodhi (1995: 395). 22 SN 41:7 Godatta; SN iv 295f; Bodhi (2000: 1325f). Parallel in SĀ 567 《雜阿 含經》卷21：「有無量心三昧．無相心三昧．無所有心三昧．空心三昧。」 (CBETA, T02, no. 99, p. 149, c13-14). 23 Possibly, as Yìnshùn states, simply shorthand for “non-conflict meditation (*araṇa­samādhi; 無諍三昧)” (Yìnshùn 1985: 22). 24 SĀ 80 *Ārya­dharma­mudrā­jñāna­darśaṇa­viśuddhi Sūtra (聖法印知見清淨經) 《 雜阿含經》卷3 (CBETA, T02, no. 99, p. 20, a25-b27). 25 MĀ 211 《中阿含經》卷58：「有二因二緣住無想定。云何為二。一者不念 一切想。二者念無想界。是謂二因二緣住無想定。」(CBETA, T01, no. 26, p. 792, b18-20). 26 MN 43 Mahāvedalla Sutta, MN i 296-297: “Tayo kho āvuso paccayā animittāya cetovimuttiyā ṭhitiyā: sabba­nimittānañca amanasikāro, animittāya ca dhātuyā manasikāro pubbeva abhisaṅkhāro ...”; English in Ñāṇamoli & Bodhi (1995: 393): “… there are two conditions for the persistence of the signless deliverance of mind: non-attention to all signs, attention to the signless element, and the 48 Published Page Number: 000222 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? prior determination [of its duration].” 27 SĀ 80 《雜阿含經》卷3：「… 復有正思惟三昧。觀察…我．我所。從若 見．若聞．若嗅．若甞．若觸．若識而生。復作是觀察。若因．若緣而生識 者。彼識因．緣。為常．為無常。復作是思惟。若因．若緣而生識者。彼 因．彼緣皆悉無常。復次。彼因．彼緣皆悉無常。彼所生識云何有常。無常 者。是有為行。從緣起。是患法．滅法．離欲法．斷知法。是名聖法印．知 見清淨。」(CBETA, T02, no. 99, p. 20, a29-b25). 28 SN 12:20, ii 26; SN 12:34, ii 60; SN 22:21, iii 24; SN 36:8, iv 214f; MN i 500: ≈ “… aniccā saṅkhatā paṭiccasamuppannā khayadhammā vayadhammā virāgadhammā nirodhadhammā” (slight variants); Bodhi (2000: 551, 573, 871, 1268f); Ñāṇamoli & Bodhi (1995: 605f): “…impermanent, conditioned, dependently arisen, subject to destruction, vanishing, fading away and ceasing” (with slight variants). 29 However, it appears that the Chinese reads “*virāga­dharma” or “*virakta-°” as literally “forsake desire dharma” (離欲法), which seems out of place in that it suggests that all formations (*abhi­saṃskāra) have thus “forsaken desire”. Also, “forsake knowledge dharma” (斷知法; *nirodha­jñāna­dharma) does not match the Pāli “subject to cessation” (nirodha­dhamma). It is hard to ascertain whether this reading is part of the Sarvāstivāda tradition, or translator error. A similar translation appears in the Śāriputra Abhidharma Śāstra 《舍利弗阿 毘曇論》卷11 (CBETA, T28, no. 1548, p. 606, a2-12; and ibid. p. 654, c9-10): = “… teaching on pacification, cessation, the dharma of forsaking desire, the true path towards nirvāṇa.” However, another sense of “dharma” as “teaching” appears in the Mahāvibhāṣa Śāstra 《阿毘達磨大毘婆沙論》卷161：「復次 有瑜伽師。雖厭欲界苦求離欲樂。而於離欲法懈怠不修故。」(CBETA, T27, no. 1545, p. 816, c3-5): = “Moreover, although some practitioners of meditation (*yogācārin) have revulsion towards the dissatisfaction of the desire realm and seek the [meditative] bliss of forsaking sensual craving, however, they are negligent and do not cultivate the ‘dharma of forsaking desire’.” There may be some conflation of various senses of “dharma” here, eg. teaching, principle or law, phenomenon, etc. 30 These are listed in accepted order of compilation (from Gethin 1998: 205). 31 These are listed in accepted order of compilation, for a full list see Yìnshùn (1968: 121-171), and using Dhammajoti’s system of abbreviations (2007: x-xii, 98ff). 32 See Yìnshùn (1968: 20f, 69ff, 127f); Willemen, Dessein & Cox (1998: 164); Bareau (2005: 289f); and Mizuno (2003: 389-416) for details of the Śāriputra Abhidharma Śāstra in this role. 33 SĀ 335 《雜阿含經》卷13 (CBETA, T02, no. 99, p. 92, c12-26); and EĀ 27:7 《增壹阿含經》卷30：〈37 六重品〉 (CBETA, T02, no. 125, p. 713, c12-p. 714, b12). Similar passages are also given in: EĀ 38:9 《增壹阿含經》卷32 (CBETA, T02, no. 125, p. 724, b12-18); and EĀ 51:8 《增壹阿含經》卷49〈51 非常品〉 (CBETA, T02, no. 125, p. 819, c5-25), (which otherwise parallel to MN 143 Anāthapiṇḍikovāda, but the Pāli lacks the passage here). 34 SĀ 335 《雜阿含經》卷13：「眼生時無有來處。滅時無有去處。如是眼不 實而生。生已盡滅。有業報而無作者。此陰滅已。異陰相續。除俗數法。 49 Published Page Number: 000223 JCBSSL VOL. XI 35 36 37 38 39 40 41 42 43 44 45 46 47 48 耳．鼻．舌．身．意亦如是說。」(CBETA, T02, no. 99, p. 92, c16-25). No Pāli equivalent. EĀ 37:7 《增壹阿含經》卷30：〈37 六重品〉「彼云何名為第一最空之 法。若眼起時則起。亦不見來處。滅時則滅。亦不見滅處。除假號法．因緣 法。…此六入亦無人造作。亦名色．六入法。六入亦無人造作。」(CBETA, T02, no. 125, p. 713, c12-p. 714, b12). From Vibhaṅga­aṭṭhakathā 68; also compare PṭsA 371. From Visuddhi­magga 410; also Abhidhammattha­vikāsinī 417. Cf. SN 35 Saḷāyatana Saṃyutta SN 35:246 Vīṇā (CBETA and Akanuma have 35:205 which is incorrect); = Bodhi (2000: 1253f); SN iv 195f. SĀ 1169 《雜 阿含經》卷43 (CBETA, T02, no. 99, p. 312, c6-16). For the Sanskrit Mūlasarvāstivāda Vinaya Saṃghabhedavastu, (Gnoli A 407b, SBV I 158): “… nānyatra dharmasaṅketāt; tatrāyaṃ dharma­saṅketo yadutāsmin satīdaṃ bhavati; …”. For the Chinese version of the Bimbisāra Sūtra: 《頻婆 娑羅王經》卷1 (CBETA, T01, no. 41, p. 826, b5-10); which has “由有法想” and “由彼法想” for “dharma­saṅkhetād”, lit. “from those dharma perceptions”. Also MĀ 《中阿含經》卷11〈6 王相應品〉：「…皆由因緣合會生苦…」 (CBETA, T01, no. 26, p. 498, b9-14), which gives a stronger sense of dependency. Abhidharma­kośa verse 3:17-18, and Bhāṣya: “asti karmāsti vipākaḥ, kārakas tu nopalabhyate ya imāṃś ca skandhān nikṣipati anyāṃś ca skandhān prati­ sandadhāty anyatra dharma­saṃketāt| tatrāyaṃ dharmasaṃketo yadutāsmin satīdaṃ bhavatīti vistareṇa pratītya­samutpādaḥ|” An English translation can be found in de la Vallèe Poussin, trs. Pruden (1991: Vol II, 399); though the meaning seems lost in translation. SĀ 335 《雜阿含經》卷13：「「除俗數法。俗數法者。謂此有故彼有。此 起故彼起。如無明緣行。行緣識。廣說乃至純大苦聚集起。又復。此無故彼 無。此滅故彼滅。無明滅故行滅。行滅故識滅。如是廣說。乃至純大苦聚 滅。比丘。是名第一義空法經。」(CBETA, T02, no. 99, p. 92, c16-25). No Pāli equivalent. This text is strictly speaking a “paryāya, and not to be confused with MN 121 / MĀ 190 Mahā­śūnyatā Sūtra. SĀ 297 《雜阿含經》卷12：「云何為大空法經。所謂此有故彼有。此起故 彼起。謂緣無明行。緣行識。乃至純大苦聚集。..」(CBETA, T02, no. 99, p. 84, c14 – p. 85, a1-9) For a Sanskrit re-construction of this text, see Trīpāṭhi in Choong (1999: 91f). SN 12:35 Avijjāpaccayā, ii 60f; see Bodhi (2000: 60f). SĀ 297 《雜阿含經》卷12 (CBETA, T02, no. 99, p. 84, c14–p. 85, a1-9). SĀ 297 《雜阿含經》卷12：「…若比丘無明離欲而生明。彼無明滅則行 滅。乃至純大苦聚滅。是名大空法經。」(CBETA, T02, no. 99, p. 84, c14 – p. 85, a1-9). Although Reat’s study is excellent, it is unfortunate that apart from the Sanskrit and Tibetan sources for the sūtra and comparisons with the Pāli, more attention was not given to the Chinese sources and extant Āgama literature. See Reat (1993: 28); Vaidya (1961: 100): “yo bhikṣavaḥ pratītya­samutpādaṃ paśyati, sa dharmaṃ paśyati | yo dharmaṃ paśyati, sa buddhaṃ paśyati iti”; Chinese Śālistamba Sūtra 《佛說稻芉經》卷1：「見十二因緣。即是見[23] 50 Published Page Number: 000224 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? 49 50 51 52 53 54 55 56 57 58 59 60 法。即是見佛。」(CBETA, T16, no. 709, p. 816, c24-25) [23]法＋（見法）【 元】【明】; and other Chinese translations. The two statements are found in MN 28 i 190-1 and SN 22:87 iii 120 respectively. Śālistamba Sūtra 《佛說稻芉經》卷1：「如是舍利弗。眾生不從此世至於後 世。不從後世復至於[9]此。然有業果因緣報應。不可損減。…」(CBETA, T16, no. 709, p. 818, b20-26) [9]此＋（世）【宋】【元】【明】. Compare with Sanskrit in Reat (1993: 64 v36); and Vaidya (1961: 105): “tatra na kaścid dharmo ‘smāllokāt paralokaṃ saṃkrāmati| asti ca karma­phala­prativijñaptiḥ, hetu­pratyayānām­avaikalyāt|”. Śālistamba Sūtra 《佛說稻芉經》卷1：「…無生無作。非有為無住無為。非 心境界寂滅無相。…」(CBETA, T16, no. 709, p. 817, a11-17). Reat (1993: 32, 70); and Vaidya (1961: 100-1, 106): “…asataḥ tucchataḥ riktataḥ asārataḥ rogataḥ gaṇḍataḥ śalyataḥ aghataḥ anityataḥ duḥkhataḥ śūnyataḥ anātmataḥ samanupaśyati.” See SĀ 259 《雜阿含經》卷10 (CBETA, T02, no. 99, p. 65, b14-15); SN 22 Khandha­saṃyutta has “aniccato dukkhato rogato gaṇḍato sallato aghato ābādhato parato palokato suññato anattato ...”. Other references include SĀ 104 (= SN 22:85 Yamaka); SĀ 259 (= SN 22:122 Sīla); SĀ 265 (= SN 22:95 Pheṇa); SĀ 291 (= SN 12:66 Sammasa); SĀ 864, 867, 869, 870; SĀ 1175 (= SN 35:204 Kiṃsukā); etc. Śālistamba Sūtra 《佛說稻芉經》卷1：「…諸見我見眾生見命見丈夫見吉 不吉見。…」(CBETA, T16, no. 709, p. 818, c6-13); see Reat (1993: 70-2); Śālistamba Sūtra (Vaidya 1961: 106): “… ātmavāda­pratisaṃyuktāni sattvavāda­ pratisaṃyuktāni jīvavāda­pratisaṃyuktāni pudgala­vāda­pratisaṃyuktāni kautuka­ maṅgalavāda- pratisaṃyuktāni …||”. This passage is from the end of SN 12:20 or SĀ 296, Śālistamba Sūtra 《佛說稻芉經》卷1 (CBETA, T16, no. 709, p. 818, c1316). See Reat (1993: 72-73); Vaidya (1961: 106): “yo bhadanta śāriputra evaṃvidha­dharmakṣānti­samanvitaḥ pratītya­samutpādaṃ samyag­avagacchati, tasya tathāgato ‘rhan samyak­saṃbuddho … samyak saṃbodhiṃ vyākaroti— samyak saṃbuddho buddho bhaviṣyasīti |”. Dharma­skandha­pāda 《阿毘達磨法蘊足論》卷11〈21 緣起品〉：(CBETA, T26, no. 1537, p. 505, a9-b13). Śāriputra Abhidharma Śāstra 《舍利弗阿毘曇論》卷12〈非問分〉：(CBETA, T28, no. 1548, p. 606, b17-25). Dharma­skandha­pāda 《阿毘達磨法蘊足論》卷11〈21 緣起品〉：「彼如是 知，如是見故。所有世間，各別見趣。謂：我論相應、有情論相應、命者論 相應、吉凶論相應；瑩飾防護，執為己有，有苦有礙，有災有熱。彼於爾 時，得斷遍知。如斷樹根及多羅頂，無復勢力，後永不生。」(CBETA, T26, no. 1537, p. 505, b6-10). The parallel passage in SĀ 296 《雜阿含經》卷12 (CBETA, T02, no. 99, p. 84, c3-7). Pṭs i 50-52; English in Ñāṇamoli (1982b: 50-52). SĀ 296 《雜阿含經》卷12：「云何為因緣法。謂此有故彼有。謂緣無明 行。緣行識。…若佛出世。若未出世。此法常住。法住法界。…此等諸法。 法住(*dharma­sthititā)．法空(à 法定 *dharma­niyāma; see Yìnshùn 1944: 18)． 法如(*dharma­tathatā)．法爾(*dharmatā)．法不離如(*dharma­viparīta­tathatā) 51 Published Page Number: 000225 JCBSSL VOL. XI 61 62 63 64 65 66 67 68 69 ．法不異如(*dharm­āviparītathatā)．審諦(*bhūtatā)．真實(*satyatā / *tattva) ．不顛倒(*avipary­ayatā)。如是隨順緣起。是名緣生法。…」(CBETA, T02, no. 99, p. 84, b14-26). The passage in Dharma­skandha­pāda 《阿毘達磨法蘊足 論》卷11〈21 緣起品〉：「苾芻當知。生緣老死。若佛出世。若不出世。 如是緣起法住(*dharma­sthititā)法界(*dharma­dhātu)。一切如來。自然通達。 等覺宣說。施設建立。分別開示。令其顯了。謂生緣老死。如是乃至。無明 緣行。應知亦爾。此中所有法性(*dharmatā)．法定(*dharma­niyāma)．法理 (*dharma­naya / °-yukti)．法趣(*dharma­gati)．是真(*tattva)．是實(*bhūtatā) ．是諦(*satyatā)．是如(*tathatā)．非妄(*avitathatā)．非虛(*amṛṣatā)．非倒 (*aviparītatā)．非異(*avipary­ayatā)．是名緣起。」(CBETA, T26, no. 1537, p. 505, a16-22). Dharma­skandha­pāda 《阿毘達磨法蘊足論》卷6〈10 聖諦品〉：(CBETA, T26, no. 1537, p. 480, c23-29; p. 481, a19-25; p. 481, b25-c2; p. 482, a14-20); four paralleled statements, one for each of the four āryan truths. Mahāvibhāṣa Śāstra 《阿毘達磨大毘婆沙論》卷23 (CBETA, T27, no. 1545, p. 116, b28-c22). Dharma­skandha­pāda 《阿毘達磨法蘊足論》卷11〈21 緣起品〉：「此中緣 起、緣已生法，其體雖一，而義有異。謂：…或有緣已生法非緣起者。謂： 無明、行、識、名色、六處、觸、受、愛、取、有、生、老死。」(CBETA, T26, no. 1537, p. 505, b14-19). Assuming, of course, that Vasumitra adheres to the Dharma­skandha­pāda. This portion at the beginning of the chapter is unfortunately not within the recently discovered and edited Sanskrit fragments of the Dharma­skandha­pāda. *Vasumitra Bodhisattva Saṃgraha Śāstra 《尊婆須蜜菩薩所集論》卷2：「… 以世間智觀。不等越次取證。頗觀法不觀十二緣起耶。空解脫門等越次取 證。頗觀法及十二緣起。以是緣觀。無願解脫門等越次取證。頗不觀法不觀 十二緣。除上爾取事則其義也。…」(CBETA, T28, no. 1549, p. 735, a27-b14); and ibid. 卷2：「問起亦是十二緣起。或作是說。若十二緣起。是十二緣起 法耶。或十二緣起法。彼非十二緣起諸起空寂法。或作是說。諸法生時。是 十二緣起諸法。已生十二緣起法。」(CBETA, T28, no. 1549, p. 736, a25-28). Mahāvibhāṣa Śāstra 《阿毘達磨大毘婆沙論》卷110： (CBETA, T27, no. 1545, p. 572, b16-c27). Mahāvibhāṣa Śāstra 《阿毘達磨大毘婆沙論》卷24：(CBETA, T27, no. 1545, p. 123, c1-p. 124, a8); and also at VbŚ 《阿毘曇毘婆沙論》卷13〈3 人品〉 ：(CBETA, T28, no. 1546, p. 97, b9-13). Mahāvibhāṣa Śāstra 《阿毘達磨大毘婆沙論》卷183：「三世三佛陀，能破諸 愁毒，彼皆重正法，恒住於法性。」(CBETA, T27, no. 1545, p. 917, b9-10). Mahāvibhāṣa Śāstra 《阿毘達磨大毘婆沙論》卷126：「…證法性故。」 (CBETA, T27, no. 1545, p. 658, b17-21). Mahāvibhāṣa Śāstra 《阿毘達磨大毘婆沙論》卷33：「無怖者。謂善通達 緣起法故。善修習空解脫門故。不畏惡趣及生死苦。」(CBETA, T27, no. 1545, p. 173, a28-29): = “‘Fearless’ is from having skillfully penetrated into the Dharma of dependent origination, from having skillfully cultivated the emptiness entrance to release, they no longer fear evil destinites and the pains of birth and death.” Also in variant Vibhāṣa Śāstra 《阿毘曇毘婆沙論》卷18〈4 愛敬品〉：(CBETA, T28, no. 1546, p. 131, b8-12). 52 Published Page Number: 000226 Shi Huifeng: “Dependent Origination = Emptiness”—Nāgārjuna’s Innovation? 70 Mahāvibhāṣa Śāstra 《阿毘達磨大毘婆沙論》卷2：(CBETA, T27, no. 1545, p. 7, a24-b2). 71 See Bareau (2005: 287) for the basic theses of each school on these three issues. The thesis numbers (#) are those given in Bareau’s lists, and are found respectively on pages: Theravāda #21, #55, #186 (pp. 222, 226, 242); Vātsīputrīya #38 (pg. 116); Sarvāstivāda #6 (pg. 137); Śāriputra Abhidharma #10 (pg. 199); Mahāsāṃghika #43 (pg. 56); Mahīśāsaka #19 (pg. 188); Dharmagupta #13 (pg. 197); Pūrvaśaila #9 (p. 94); Uttarāpathaka #32 (pg. 258)—note that they are split on both sides; and the Vibhajyavāda #8 (pg. 175). 72 Bareau (2005: 56 n130): “Vasumitra, thesis 40. Vinītadeva, thesis 25”; and ibid. (2005: 199 n1066): “Taishō 1548, 633ab.” Given that the source below for the Mahīśāsaka is from the Pāli where the sources here are all from the Chinese, Bareau’s rendering here of “*°-svabhāva” from “性” should probably be “°-tathatā” (as below), or at least “°-tā”. 73 Bareau (2005: 188 n995): “Vasumitra, thesis 17; Vinītadeva, thesis 8. Kathāvatthu, II, 11; VI, 2 and 6.” 74 Bareau (2005: 197 n1052): “Taishō 1545, p. 198a.” 75 Bareau (2005: 94 n374): “Kathāvatthu, VI, 2.” 76 Bareau (2005: 94 n375): “Kathāvatthu, VI, 3.” 77 Bareau (2005: 175 n917): “Taishō 1545, p. 116c. Cf. L. V. P.: Kośa, III p. 77. See thesis 9 of the Pūrvaśailas.” 78 Katthā­vatthu 19:2 578: “Suññatā saṅkhāra­kkhandha­pariyāpannāti? Āmantā. … Suññatā na vattabbā—‘‘saṅkhāra­kkhandha­pariyāpannā’’ ti? Na hevaṃ vattabbe…”; etc.. See Bareau (2005: 242). 79 “Neither arising nor ceasing”: See SĀ 293 《雜阿含經》卷12 (CBETA, T02, no. 99, p. 83, c13-17); no equivalent in SN, though possibly around SN 12 given the relation of other adjacent sūtras. It is similar to both SN 6:1, and also the opening of DN 15 Mahānidāna Sutta, DN ii 55. “Neither eternalism nor nihilism”: See SĀ 300 《雜阿含經》卷12 (CBETA, T02, no. 99, p. 85, c11-13); = SN 12:46 Aññatara (Another); refer Bodhi (2000: 583); SN ii 76. 53 Published Page Number: 000227 JCBSSL VOL. XI 54 Published Page Number: 000228