Candrakīrti

活跃于六七世纪的印度中观派论师月称，在《明句论》一书中，批评了清辨论师所采用的自立比量。其中，月称所用技巧与汉传因明自他共三比量之说非常接近。对比此二者，可以为我们理解月称思想与汉传因明的起源提供有益的视角与材料。本文在分析月称《明句论》梵文本的基础上，着重考察自他共三比量视角下月称思想的样态，进而得出如下结论：月称归谬论法的核心是“他比量”；他认为清辨的自立比量只是一种“自比量”，不具备能破的功能；月称所用技巧更为质朴、简洁，与汉传因明中的三比量说的复杂形成了鲜明对比。因此，月称之说可能代表了三比量发展的较早阶段，为我们了解三比量说乃至汉传因明的起源和发展提供了珍贵的原始资料。

The Three Kinds of Inferences Found in Candrakīrti’s Prasannapadā
Abstract: The 6th-7th century Buddhist scholar Candrakīrti is well-known for his justifications for the Buddhapālita’s prasaṅga argumentative method and his severe criticisms toward Bhāvi-veka’s independent probative inference, which he provided in the first chapter of Prasannapadā (PsP). Because of his great influence over Tibetan Buddhism, this section of PsP has been re-peatedly studied since the discovery and publication of PsP. However, previous studies have rarely noted that the technique Candrakīrti utilized for his justifications or critiques is highly similar to the famous “three -inferences -theory” in Kuiji’s (窥基) “Great Commentary on Nyāyapraveśa”, namely the inference accepted by oneself (自比量), by others (他比量), and by both (共比量). This paper aims to analyze Candrakīrti’s related passages using the three -inferences -theory in order to reveal why Candrakīrti abandoned Dignāga’s rigid requirement of the common acceptance (共许决定) for an inference. It is reasonable to conclude that Can-drakīrti’s justifications for Buddhapālita and critiques toward Bhāviveka can be understood through the three -inferences -theory.

Revised publication of Ph.D. dissertation providing a text-critical study and translation of the Karmaphalaparīkṣā chapter of Candrakīrti's Prasannapada with a study of early Buddhist theories of karman, relying on Sanskrit, Pali, Chinese, and Tibetan sources.

Mahāyāna streams of Buddhism have had a profound impact on the shape of Buddhist formations in India, throughout Asia, and beyond. This course examines, in a selected survey format, the classification, cultural formations, literature, and systematic thought of Mahāyāna doctrines and practices utilizing a range of interdisciplinary methods and approaches. The course focuses on the problems of classification and definition; the "new metaphysics" of emptiness; controversies in the conception of Buddhahood; the centrality of compassion; visualization and other meditation techniques; the role of faith and the practices of lay people; and the development of the bodhisattva ideal.

The Lotus Sutra, an important Mahāyāna scripture in East Asian history, influenced a great number of scholars and schools in China and Japan. What about the influence and importance of the Lotus Sutra in Indian Buddhism? This paper examines the understanding of the Lotus Sutra found in the works of the Indian Buddhist scholar Candrakīrti (ca. 570-650), an influential commentator in the Buddhist thought of India and Tibet. Candrakīrti’s use of the Lotus Sutra provides a case study for the importance of the Lotus Sutra in the history of Indian Buddhism.

The teachings of Madhyamaka (" middle way philosophy ") have been the basis of Tibetan Buddhist thought and practice since the eighth century. After the twelfth century, Tibetan scholars distinguished two branches of Madhyamaka: Autonomist (rang rgyud pa) and Consequentialist (thal 'gyur ba, *prāsaṅgika). What distinctions in Madhyamaka thought and practice did twelfth century Tibetan scholars make to differentiate these two branches? This article focuses upon a newly identified twelfth century Tibetan manuscript on Madhyamaka from the Collected Works of the Kadampas: Khu lo tsā ba's Treatise. Khu lo tsā ba, also known as Khu ston Mdo sde 'bar, was a contemporary of Jayānanda and Pa tshab Lo tsā ba Nyi ma grags and instrumental for the revitalized reception of Madhyamaka in twelfth century Tibet. Khu lo tsā ba's Treatise outlines a multifaceted understanding of Madhyamaka (" middle way philosophy ") as the work solely focuses upon the distinctions between Autonomist and Consequentialist Mādhyamikas. The paper outlines the structure and content of the Treatise and identifies six topics that the author emphasizes to distinguish Autonomist and Consequentialist branches of Madhyamaka. Based on an overview of these topics, the paper concludes that the Autonomist/Consequentialist distinction of Madhyamaka was already well developed in twelfth century Tibet and involved a differential of systems that encompassed much more than distinct styles of logical proof.

PhD dissertation on the 17th chapter of Nagarjuna's Mulamadhyamakakarika with Candrakirti's Prasannapada commentary, providing an analysis of early Buddhist karma theories of little known Abhidharma schools. The majority of the dissertation is now obsolete, since a revised version has been published in the 2006 book "Early Buddhist Theories of Action and Result". However, the dissertation contains some parts that were not included in the 2006 book and that remain unpublished. These include the critical Sanskrit and Tibetan editions of the last third of the 17th chapter of the text as well as the appendix providing an English translation of that part of the text.

The sense of the opening argument of Nāgārjuna's Mūlamadhyamakakārikā -- an argument that Candrakīrti and other commentators took to epitomize Madhyamaka -- depends on the construal of a locative construction in MMK 1.3. On one reading, the argument of this verse seems clearly fallacious, although there has been some back-and-forth about the most natural reading of Nāgārjuna's Sanskrit. Little attention has been paid, though, to Candrakīrti's extensive comments on the matter, according to which what is advanced is a sophisticated critique of causal realism. Against the  most natural reading, Candrakīrti takes the locative in question as a locative absolute; while that is the construal that yields a clearly fallacious argument, however, Candrakīrti, emphasizing the temporal implications of locative absolutes, clarifies that what is at stake is the temporality of causation. The argument, on Candrakīrti's reading, is that causation cannot make sense as ultimately real just because causation is a relation between relata (viz., cause and effect) that are, by definition, never present together. To that extent, it can only be as conceived from some perspective that causation is intelligible.

Translation of the section of Tattvasangraha dealing with the refutation of the Pudgalavada followed by implications of the negativity one can find there compared with orthodox Madhyamika conceptions of negativity.

Copyright of the sanskrit text: Eltschinger Vincent.
Translation supervised and corrected by: Eltschinger Vincent.

This seminar examines the history of Buddhist Thought in India. Buddhists like to place emphasis not on belief as such but on practicing, following a path, and knowing, directly seeing. This direct ‘seeing things the way they really are’ is held to free a person from the depths of suffering through cognitive transformation. By ‘Buddhist thought’ we
mean, in this general context, the discussions, speculations, and arguments concerning ‘how things really are.’ In this seminar we will try to understand this issue through the works of various Buddhist philosophers beginning with Śākyamuni Buddha (480-400 BCE) through to Candrakīrti (ca. 570-650 CE).

Entre os vários pontos da obra de Nāgārjuna que deram origem a análises e discussões, o tema das 'duas verdades' é um dos mais controversos. Com efeito, dentro da ampla bibliografia dedicada a essa temática, são muitas, e amiúde divergentes, as tentativas de explicar o que Nāgārjuna entendesse - no verso 24.8 das suas Mūla-madhyamaka-kārikā (MMK) - com as expressões 'verdade convencional' e 'verdade suprema'. Esses pontos de vista interpretativos, entretanto, frequentemente, parecem prescindir daquele que talvez seja o critério mais confiável para dirimir as controvérsias relativas às MMK: i. e., a leitura dos comentários indianos às mesmas MMK. Em contraste com essa tendência, este escrito objetiva investigar as passagens do Prajñāpradīpa e das Prasannapadā onde encontramos explicações, respectivamente de Bhāviveka e de Candrakīrti, que dizem respeito às duas verdades. O resultado dessa pesquisa se apresentará como nitidamente favorável à interpretação da doutrina das duas verdades que, em outros escritos, chamei de 'pedagógica'.

La lettura corrente delle analogie tra Bradley e Śaṃkarasolitamente si concentra sulla seconda parte di "Apparence and Reality", lasciando alla prima il ruolo di lunga premessa, quando invece è nella prima parte – Apparence - che Bradley introduce l’impianto concettuale che permetterà poi di comprendere il suo concetto di Assoluto. In questa prima parte, l’autore inglese opera una destrutturazione delle categorie dell’esperienza comune che ricorda da vicino quella operata da Nāgārjuna nella sue "Strofe sulla via di mezzo". Non è così importante riscontrare una perfetta corrispondenza nei singoli procedimenti logici, quanto mettere in evidenza l’intento de-costruttivo che anima i due autori. Dopo questa prima analisi, che rivelerà come la nozione di apparenza emerga come sintesi tra il momento della “percezione” e quello della “descrizione” del reale, si sarà in grado di comprendere più rettamente l’evoluzione “anomala” di tale nozione, che infatti non si risolve mai completamente in senso idealista e conserva un carattere, problematico e sfuggente, di "indeterminatezza".

In this paper, Nāgārjuna’s philosophical interpretation of the terms kāraṇa and kārya is analysed after having methodologically confined the specific field of interest to the MMK. From the study of all the occurrences of kāraṇa and kārya in the MMK (listed in paragraph 2), it emerges that Nāgārjuna makes use of these two terms to refer to skandhas as causes (kāraṇa) of further skandhas as effects (kārya), hence conveying with this words the idea of, so to speak, subjectivity and (re)birth. From the principal commentaries on the MMK (particularly Bhāviveka and Candrakīrti), we know also that, even when the reference of kāraṇa and kārya to skandhas is not explicitly pointed out by Nāgārjuna, it nonetheless can be inferred from the peculiar context in which these terms are employed.
This conclusion seems to be confirmed also by the crosscheck analysis of the philosophical usage of the parallel terms hetu and phala, which refer either to, so to speak, objective (i.e., concerning general bhāvas) causes and effects, or – in the case of phala as consequence of karman/kriyā – to moral results (these two meanings are confirmed also by some passages from the Vigrahavyāvartanī), but in no cases Nāgārjuna makes recurse to hetu and phala to refer to what could be called a subjective (i.e., concerning skandhas) level of causality.
To the paper an addendum is added, in which the commentaries on MMK 8.4 are taken into consideration. From this excursus it appears that at least Buddhapālita, Bhāviveka and Candrakīrti, by ‘overinterpreting’ the meaning of hetu in the kārika, are consequently forced to readapt the philosophical significance of kāraṇa and kārya in a way that does not fit so much with Nāgārjuna’s original message.

Abstract

The thesis focuses on the relationship between Sanskrit classical grammar, Abhidharma, and the debates between Madhyamaka and Yogācāra. In particular, it shows how the kāraka system, and the idea of lakṣaṇa, influence philosophical argumentation in the context of medieval Indian Buddhist thought. The kāraka system is the way in which classical Sanskrit grammarians discuss syntax, and in particular, actions and agency. Lakṣaṇa means a defining trait, or a definition, at once a scholastic tool and a fundamental way to identify existent entities.

There are five Chapters and two Appendices. Chapter One shows the close links between Sanskrit classical grammar and basic ideas in Buddhist thought, and isolates the kāraka system as being most relevant in this regard. It also show certain structural analogies between the kāraka system and certain important features of Buddhist philosophy. Chapter Two is mostly based on Vasubandhu’s Abhidharmakośabhāṣya and its sub-commentary; it highlights and reconstructs the discussions on lakṣaṇa and agency found therein. Chapter Three shows how Madhyamaka understands similar issues, and what is the proper context and significance of its refutation of lakṣaṇas. Chapter Four shifts the focus upon the Madhyamaka understanding of conventional truths, and the role of the lakṣaṇas of Abhidharma, as well as of the kāraka system, within saṁvti. Chapter Five compares the Yogācāra views on the very same topics.

Appendix 1 is a translation of Prajñākaramati’s commentary to Bodhicaryāvatāra 9.1-34, a work where many of the philosophical debates discussed throughout the thesis are well represented. Appendix 2 is a photographic reproduction of a manuscript, containing an anonymous commentary to Nāgārjuna’s Lokātītastava.

Table of Contents

Abstract 2

Table of Contents 3

Introduction 8

Acknowledgements 12
Chapter 1

Sanskrit Grammar and Buddhism 14


1.1 Language analysis: nirukti and vykaraa 17

1.2 Pini 20

1.3 Buddhist grammarians 21

1.4 Defining ‘grammar’ 24

1.4.1 abdnusana 25
1.4.2 Vykaraa 28
1.4.3 stra 29
1.4.4 Stra, yoga 32
1.4.5 Lak•aa 34
1.4.6 Grammar as Śabdavidyā, in the training
of a Bodhisattva 35


1.5 The Kraka system 39

1.5.1 Vibhakti and kraka 39
1.5.2 What a kāraka is 42
1.5.3 Primary kārakas 45
1.5.4 The autonomous agent 48
1.5.5 What the agent most wishes for: karman 53
1.5.6 Kik, and Jinendrabuddhi: tathyuktam cnpstitam 57
1.6 The primary kārakas and Buddhism 59
1.6.1The kāraka system and basic Buddhist thought:
selflessness 60
1.6.2 The kāraka system and the Mahāyāna:
the purification of the three spheres 62


1.7 Continued agency and dependent arising 63

1.7.1 The Abhidharmakośabhāṣya on pratītyasamutpāda 66
1.7.2 The debate in the prasannapadā 68
1.7.3 Mādhyamika authors: more on the same issue 69


1.8 Conclusion 72


Chapter 2
Abhidharma as a dictionary:
the world in Buddhist terms 74

2.1 Dharma: what upholds its own lak•aa 79
2.1.1 The four lak•aas of all compounded factors 80
2.1.2 The lakṣaṇas and sahabhūhetu 87
2.1.3 Sautrāntika criticism of the four lakṣaṇas 89
2.1.4 On definitions and characteristics:
beyond the four lak•aas 94

2.2 The seer and the knower: identifying the synchronic agent 101
2.2.1 Cak•u payati rpi 102
2.2.2 Kart• and kriy as analytical categories 108
2.2.3 A counter-position, on agency 113
2.2.4 Agents and conventional persons:
a passage from the pudgalavinicaya section 115

2.3 ‘Self’ as indirect reference 119
2.4 Conclusion 120
Chapter 3

Lakṣaṇas and the emptiness of lakṣaṇas 122


3.1 The Lokttastava and the Heart Stra:
emptiness as lack of characteristics 124

3.1.2 From the aggregates to their emptiness 125
3.1.3 The meaning of autonomy and the absence of God 129
3.1.4 Beyond the world, beyond characteristics 132

3.2 Candrakrti and lakṣaṇas 134
3.2.1 The basic status of lakṣaṇa and lakṣya 135
3.2.2 Applying the dialectic to a specific instance:
the four conditions (pratyayas) 138
3.2.3 The example of space 141

3.3 Candrakrti and the Vaibh•ika-Sautrntika divide:
instants and the emptiness of lak•aas 144

3.3.1 Madhyamaka versus the lakṣaṇas
of all compounded factors 146
3.3.2 Does disappearance exist? 147
3.3.3 Candrakīrti and momentariness 150
3.3.4 Candrakīrti’s conventions: Vaibhāṣika or Sautrāntika? 151



Chapter 4
Conventions, the world, and Madhyamaka 156

4.1 Samv•ti and Paramrtha: Candrakrti and Prajkaramati 158

4.2 Identity and difference between Mdhyamikas and cowherds:
Abhidharma and Vykaraa 164

4.2.1 Candrakrti on the svalak•aa
and Prajkaramati on svasamvedan 168
4.2.2 Are Mdhyamikas ‘satisfied with no analysis’ ? 175
4.2.3 More on Candrakrti and krakas 181

4.3 Identity and difference between Mdhyamikas
and Crvks. Paraloka, karman, and the origin of vijna
The meaning of pratipatt•bheda 185

4.4 A system of graded saṁvṛtisatya 189
4.4.1 Candrakrti and graded conventions 191
4.4.2 Three levels of teachings 191
4.4.3 Analysing space into emptiness: atoms 198
4.4.4 ntideva and Prajkaramati:
conventions and momentariness 199

4.5 Conclusions 210


Chapter 5
Comparisons with the Yogācāra 211

5.1 Twenty Verses: vijñaptimātratā as an Abhidharmic argument 212
5.1.1 The main thesis of Verse 1 212
5.1.2 An alternative translation 217
5.1.3 The proper locus of dependent arising 220
5.1.4 Real referents of a metaphor:
a comparison with the Thirty Verses 223

5.2 The importance of lakṣaṇas 227
5.2.1 The first Chapter of the Madhyāntavibhāga 228
5.2.2 What has lakṣaṇas,
and what has the lakṣaṇa of no-lakṣaṇa 230


5.3 Kārakas and the irriducibility of consciousness 231
5.3.1 Consciousness as an action without agent or object 232
5.3.2 A passage from Asaṅga, on action and agents 235
5.3.3 A note on Śāntarakṣita’s position 237

5.4 The role of momentariness in Yogācāra 239
5.5 Conclusion 242

Conclusion 244

Appendix 1 | Śāntideva’s Bodhicaryāvatāra, 9.1-34
with Prajñākaramati’s Pañjikā: English Translation 251
Note on the Translation 252
The importance of Prajñākaramati’s commentary 256


Appendix 2 | Reproduction of a Devanāgarī manuscript
containing Nāgārjuna’s Lokātītastava,
together with an anonymous commentary 336


Bibliography 341

"Religion 43 (2013), no. 4, pp. 487-498 (+ addendum).

""This article supplements Jens Schlieter’s discussion of the cognitive metaphor of a karmic bank-account,

http://www.tandfonline.com/doi/full/10.1080/0048721X.2013.765630#.Ugy7t5I3Bic

adding selected points on karma monetary/fiscal metaphors as preserved chiefly in Pāli and Sanskrit sources. It explores various strands of the history of South Asian religions where distinct economic metaphors for karma come closer to the late ‘bank-account of karma’: i.e., the Vedic ‘three debts’, a Hindu concept of God as accountant, the varieties of weighing the (mis)deeds, the Buddhist monastic status of debt and fiscal transactions, the equivalence of karma and debt as discussed by Madhyamaka thinkers, and others. While endorsing Schlieter’s point, it also takes into account such modern Western sources as early theosophical discourse and ‘Protestant Buddhism’."""

1 Lalitavistara 25.4cd. 2 I am not implying that the MĮlamadhyamakakÄrikÄ may be related only to the NidÄnasamyukta as a non-MahÄyÄna source. Compare for example MĮlamadhyamakakÄrikÄ 24.8 with a passage from the EkottarÄgama: dve satye samupÂĜritya buddhÂnÂӅ dharmadeĜan | lokasaӅvӟtisatyaӅ ca satyaӅ ca paramÂrthataҩ || 24.8 ||; dve satye Ĝaikӧasya bahukare bhavataҩ | saӅvӟtisatyaӅ paramÂrthasatyaӅ ca || EkottarÄgama, page 201 in TripÄӷhî's edition. Thai International Journal of Buddhist Studies II (2011): 57-95.

I discuss two critiques of Dignāga’s epistemology, one from Candrakīrti and another from Jayarāśi.  I argue that they are two versions of what I call the core problem: if the content of Dignāga’s epistemology were correct, two fundamental beliefs within this epistemological theory could not be established or known to be true, as Dignāga claims they are.  In response to objections found within the classical Indian tradition as well as several plausible contemporary objections, I then argue that the core problem remains a serious issue with which those sympathetic to Dignāga ought to contend.

The thesis focuses on the relationship between Sanskrit classical grammar, Ahludharma, and the debates between Madhyamaka and Yogacara. In particular, it shows how the karaka system, and the idea of laksana, influence philosophical argumentation in the context of medieval Indian Buddhist thought. The karaka system is the way in which classical Sanskrit grammarians discuss syntax, and in particular, actions and agency. Laksana means a defining trait, or a definition, at once a scholastic tool and a fundamental way to identify existent entities. There are five Chapters and two Appendices. Chapter One shows the close links between Sanskrit classical grammar and basic ideas in Buddhist thought, and isolates the karaka system as being most relevant in this regard. It also shows certain structural analogies between the karaka system and certain important features of Buddhist philosophy. Chapter Two is mostly based on Vasubandhu's Abhidharmakosabhasya and its sub-commentary; it highlig...

Candrakīrti’s (c. 600-650 CE) critique of Bhāviveka (c. 500-570 CE) at his commentary on the Mūlamadhyamakakārikā 1.1 drew scholars’ attention as the locus classicus for dividing the Madhyamaka school into *Svātantrika (Rang rgyud pa) and *Prāsaṅgika (Thal ’gyur ba). This paper discusses the nature of the inferences that Candrakīrti presents at the end of his critique and rejects the tendency of equating them with one of the two purposes of Dignāga’s (c. 480-540 CE) inference, that is, refutation (dūṣaṇa). Candrakīrti discusses the Mādhyamikas’ formulation of inferences in relation to the Madhyamaka identity, and his inferences deliberately violate the conditions that Dignāga assumed to be indispensable for an inference to be valid. Candrakīrti sees the ultimate objective of presenting one’s own inference to others in refuting the opponent and points out that the validity of inferences does not play any role in deciding the victory or defeat of debates. Accordingly, Candrakīrti finally declares Dignāga’s project of defining the characteristics of valid inferences to be useless (niḥprayojana). Candrakīrti’s act of formulating inferences in the form of syllogism itself may be viewed as his compromise with the demands of the times. However, the fact that he presented his own syllogism does not necessarily mean that he considered it to be a purposeful act.

This short paper identifies and diplomatically edits a fragment hitherto unidentified, Bodleian Library Oxford Ms. Sansk. a. 11 (R), a few lines from chapter 24 of Candrakīrti's Prasannapadā. After briefly discussing the extraneous (tantric) material in this bundle, I collate the text with the two available editions and argue that the folio fragment might very likely be a part of the Oxford Manuscript (the so-called Ms. P).

The seventh-century Indian Buddhist master Candrakīrti is well-known for his influence upon Madhyamaka thought and practice in the history of Indian and Tibetan Buddhism, particularly through his Madhyamakāvatāra (Entrance to the Middle Way). This paper outlines significant variant readings of a verse found in the sixth chapter of Candrakīrti's Madhyamakāvatāra (6.79) and discusses the relevance and repercussions of this reading for Indian Buddhist Madhyamaka thought.

Article analyzing the exegetical parallels found between the Sanskrit commentaries on Nagarjuna's Mūlamadhyamakārikā. The paper also provides an extensive methodological discussion of how to study classical Indian exegetical traditions on the basis of textual borrowings and parallels.

In this paper, a spintronic neuromorphic reconfigurable Array (SNRA) is developed to fuse together power-efficient probabilistic and in-field programmable deterministic computing during both training and evaluation phases of restricted Boltzmann machines (RBMs). First, probabilistic spin logic devices are used to develop an RBM realization which is adapted to construct deep belief networks (DBNs) having one to three hidden layers of size 10 to 800 neurons each. Second, we design a hardware implementation for the contrastive divergence (CD) algorithm using a four-state finite state machine capable of unsupervised training in N+3 clocks where N denotes the number of neurons in each RBM. The functionality of our proposed CD hardware implementation is validated using ModelSim simulations. We synthesize the developed Verilog HDL implementation of our proposed test/train control circuitry for various DBN topologies where the maximal RBM dimensions yield resource utilization ranging from 51 to 2,421 lookup tables (LUTs). Next, we leverage spin Hall effect (SHE)-magnetic tunnel junction (MTJ) based non-volatile LUTs circuits as an alternative for static random access memory (SRAM)-based LUTs storing the deterministic logic configuration to form a reconfigurable fabric. Finally, we compare the performance of our proposed SNRA with SRAMbased configurable fabrics focusing on the area and power consumption induced by the LUTs used to implement both CD and evaluation modes. The results obtained indicate more than 80% reduction in combined dynamic and static power dissipation, while achieving at least 50% reduction in device count.

Abstract The article discusses the relevance of the syntactical implications of the absolutive ending (lyabanta) in interpreting the Madhyamaka term upādāyaprajñapti, and hence Mūlamadhyamakakārikā 18.24. The views of both Sanskrit and Pāli classical ...

Although Atiśa is famous for his journey to Tibet and his teaching there, his teachings of Madhyamaka are not extensively commented upon in the works of known and extant indigenous Tibetan scholars. Atiśa’s Madhyamaka thought, if even discussed, is minimally acknowledged in recent modern scholarly overviews or sourcebooks on Indian Buddhist thought. The following annotated translation provides a late eleventh century Indo-Tibetan Madhyamaka teaching on the two realities (satyadvaya) attributed to Atiśa Dīpaṃkaraśrījñāna (982–1054 c.e.) entitled A General Explanation of, and Framework for Understanding, the Two Realities (bden gnyis spyi bshad dang/ bden gnyis ’jog tshul). The text furnishes an exposition of the Middle Way (madhyamaka) thought of Nāgārjuna based on an exegesis of conventional reality and ultimate reality within the framework of Mahāyāna path structures found in texts attributed to Maitreyanātha. The General Explanation fills an important gap in the historical knowledge of Madhyamaka teachings in eleventh century India and Tibet. The text presents a Madhyamaka teaching brought to Tibet by Atiśa and provides previously unknown evidence for the type of pure Madhyamaka teachings that circulated among the communities of early followers of Atiśa. These teachings were disseminated before the rise of the early Bka’-gdams-pa monastery of Gsang-phu ne’u-thog and its debating traditions that, particularly beginning in the twelfth century, placed emphasis on the merger of Madhyamaka and Epistemology (pramāṇa).

Candrakīrti's Prasannapadā(PsP) is the only complete Sanskrit commentary on Nāgārjuna's Mūlamadhyamakakārikā(MMK). As a commentary on MMK, PsP cites the verses of MMK sequentially. This paper calls this kind of MMK's verses as the “basic verses”. Besides them, in order to explain a certain thought or idea, Candrakīrti also freely cites MMK's verses as his āgama. Among them, some MMK's verses contain different readings from the basic ones. This paper names them as “variant verses”. This paper will collect all the variant verses cited in PsP, compare them with six Sanskrit manuscripts, and discuss the possible causes of them.