Is Dharmapāla Criticizing Bhāviveka in the Cheng weishi lun? Silla Yogācāra Master Taehyŏn’s Views on the Dispute between Emptiness and Existence

Is Dharmapāla Criticizing Bhāviveka in the Cheng weishi lun? Silla Yogācāra Master Taehyŏn’s Views on the Dispute between Emptiness and Existence

Sumi Lee

Abstract

The Cheng weishi lun 成唯識論 (*Vijñaptimātratāsiddhi-śāstra), a compiled translation of Indian commentaries on Vasubandhu’s (ca. fourth or fifth centuries CE) Triṃśikā, centering on Dharmapāla’s (ca. sixth century) exegesis, is well known as the foundational text that offers the doctrinal basis of the East Asian Yogācāra school—the Dharma Characteristics school (Ch. Faxiang zong, K. Pŏpsang chong, J. Hossōshū法相宗). In his commentary to the Cheng weishi lun, Kuiji 窺基 (632–682), the de facto founder of the Dharma Characteristics school, considers Dharmapāla’s criticism in the Cheng weishi lun toward those adhering to ‘‘emptiness’’ as aimed at such a Madhyamaka scholiast as Bhāviveka (ca. 500–570).

Kuiji’s interpretation has tended to be generally accepted under the backdrop of the contemporary controversy revolving around the distinct doctrinal views between Dharmapāla and Bhāviveka. However, just as the question of whether Madhyamaka and Yogācāra over the long history of Mahāyāna Buddhism were philosophically consistent with each other cannot find an easy answer, we cannot simply conclude based on the ‘‘orthodox’’ explanation that these two scholiasts were doctrinally antagonistic. Indeed, the Silla Yogācāra monk Taehyŏn 大賢 (ca. eighth century) introduced three distinct interpretations by contemporary scholar monks on this matter. This paper examines East Asian commentators’ interpretations on the relationship between Dharmapāla and Bhāviveka as presented in Taehyŏn’s commentary of the Cheng weishi lun, and further discusses how Taehyŏn explains doctrinal conflict between the seemingly contrasting notions, such as ‘the conditioned’ (saṃskāra) and ‘the unconditioned’ (asaṃskāra), by observing his interpretation of such a concept of ‘the immaculate consciousness’ (amalavijñāna).

Keywords: Madhyamaka, Yogācāra, Cheng weishi lun 成唯識論 (*Vijñaptimātratāsiddhi-śāstra), Dharmapāla, Bhāviveka (ca. 500–570), Taehyŏn 大賢, amalavijñāna

Sumi Lee is a HK Research Professor of the Institute for Buddhist Culture at Dongguk University. Her research interests focus on the relationship between Yogācāra and tathāgatagarbha thought and its soteriological implications, the exegetical interpretations in East Asian Yogācāra tradition. Her publications include, ‘‘The Meaning of ‘Mind-made Body’ (Skt. manomaya-kāya, Ch. yisheng shen 意生身) in Buddhist Cosmological and Soteriological Systems’’ (2014), ‘‘Redefining the ‘Dharma Characteristics School’ and East Asian Yogācāra Buddhism’’ (2015), and ‘‘On the Ālayavijñāna in the Awakening of Faith: Comparing and Contrasting Wŏnhyo and Fazang’s Views on Tathāgatagarbha and Ālayavijñāna’’ (2019).

This article is an edited transcript of a paper I presented at a panel on the Cheng weishi lun at the 2019 Annual Conference of the American Academy of Religion (November 23–26, 2019). I am especially grateful to Prof. Charles Muller, the editor of this special issue on Silla Yogācāra Buddhism, for his consistent help and advice. This work was supported by the Ministry of Education of the Republic of Korea and the National Research Foundation of Korea

Introduction

The Cheng weishi lun 成唯識論 (*Vijñaptimātratāsiddhi-śāstra; hereafter, CWSL), a compiled translation by the famous Chinese translator Xuanzang 玄奘 (602–664) and his disciple Kuiji 窺基 (alt. Ji 基, 632–682) of Indian commentaries on Vasubandhu’s (ca. fourth or fifth century CE) Triṃśikā, centering on Dharmapāla’s (ca. sixth century) commentary, is one of the main doctrinal texts of the East Asian Yogācāra school, namely, the Dharma Characteristics school (Ch. Faxiang zong, K. Pŏpsang chong, J. Hossōshū 法相宗).1 There is no clear historical evidence to disclose the exact reason why Xuanzang revised his original plan to translate all the commentaries and render them into one compilation centering on Dharmapāla’s work. However, when based on the account in Kuiji’s Cheng weishi lun zhangzhong shuyao 成唯識論掌中樞要 (hereafter, Shuyao) that Xuanzang changed his mind following Kuiji’s advice that the interpretive complexity would later cause controversy, the decision seems to have been related to a controversial situation surrounding translation at that time.2 It is in fact well known that as Xuanzang translated canonical texts brought from India, doctrinal disputes emerged over the newly introduced theories. For instance, the theory of ‘‘five lineages’’ (Ch. wu zhongxing 五種性, Skt. pañcagotra) in early Yogācāra texts, such as the Yogācārabhūmiśāstra (Ch. Yuqieshidi lun 瑜伽師地論) and the *Buddhabhūmisūtra-śāstra (Ch. Fodijing lun 佛地經論), was considered to be in conflict with the concept of universal Buddha-nature (Ch. foxing 佛性, Skt. buddha-dhātu). As Xuanzang translated Bhāviveka’s (Ch. Qingbian 清辯/清辨; ca. 500–570) *Mahāyāna-hastaratna-śāstra (Ch. Dasheng zhangzhen lun 大乘掌珍論; hereafter, ZZL) in 649 and in the next year Dharmapāla’s *Catuḥ-śataka-vṛtti (Dasheng guangbailun shilun 大乘廣百論釋論; hereafter, DGS), the commentary on Āryadeva’s Catuḥśatakaśāstra-kārikā (Ch. Guangbai lun 廣百論), the so-called ‘‘emptiness-existence (Ch. kong-you 空有) controversy’’ occurred over the doctrinal distinction between the ZZL and DGS.3 Given this controversial circumstance, the way in which the CWSL was translated focusing on Dharmapāla’s commentary also seems to have reflected the translators’ intention to respond to the ongoing disputes.

In particular, evidence of the controversies around the doctrinal opposition between Bhāviveka and Dharmapāla also appears in Kuiji’s commentaries on the CWSL, such as the Shuyao, in which Kuiji considers Dharmapāla’s criticism as aimed at Bhāviveka. For instance, in commenting on the passage explicating the purpose of the treatise at the beginning of the CWSL,4 Kuiji interprets ‘‘those who consider that internal consciousness does not exist in the same way as external objects do not exist’’ (或執内識如境非有), one of the several types of scholars who misunderstand the meaning of ‘‘consciousnessonly’’ (Ch. weishi 唯識, Skt. vijñaptimātratā), as referring to such masters as Bhāviveka.5 In the CWSL, ‘‘those who have an attachment to the Mahāyāna teaching of emptiness of no characteristics (Ch. xiang 相, Skt. lakṣaṇa) and regard it as the ultimate [[[truth]]]’’ (有執大乘遣相空理為究竟者) are criticized for their negating the ālayavijñāna and all dharmas on the basis of a pseudoinference (Ch. si biliang 似比量, Skt. anumānābhāsa).6 Kuiji asserts that those who hold this view indicate the advocates of ‘‘Mahāyāna of no-characteristics’’ (Ch. wuxiang dasheng 無相大乘), such as Bhāviveka, and criticizes the faultiness of the inference.7 In explaining another passage of the CWSL, in which the doctrine of consciousness-only is described as the middle way (Ch. zhongdao 中道), which is free from ‘‘two extremes of either reifying or denying’’ (Ch. zengjian erbian 增減二邊, Skt. *samāropa-apavāda),8 Kuiji interprets ‘‘the extreme of denying’’ as corresponding to Bhāviveka’s view of emptiness.9 Similarly, Kuiji also contrasts the middle way of Yogācāra teaching with the two extreme views—Bhāviveka’s emptiness and the Hīnayāna doctrine of existence.10 Furthermore, as regards the CWSL’s statement that rejection of both the ultimate truth (Ch. zhendi 眞諦, Skt. paramārtha-satya) and the conventional truth (Ch. sudi 俗諦, Skt. saṃvṛti-satya) corresponds to a wrong attachment to emptiness (Ch. equ kong 惡取空, Skt. durgṛhītā śūnyatā),11 Kuiji comments that this position refers to none other than such scholars as Bhāviveka.12 All this evidence shows that the CWSL’s criticism of mistaken notions of emptiness is interpreted by Kuiji as targeted at Bhāviveka, although Bhāviveka is never mentioned by name in the CWSL.13 The East Asian understanding of the CWSL has been generally based on the commentaries by Kuiji, the alleged de facto founder of the Faxiang school. Kuiji’s view that those who were criticized by Dharmapāla in the CWSL indicated such figures as Bhāviveka has also been broadly accepted. However, despite Kuiji’s acknowledged authority in understanding the CWSL, this matter still requires further consideration, because it is related to a fundamental hermeneutical issue that has repeatedly been raised and discussed over the long history of Mahāyāna Buddhism. The current question of whether these two exegetes were doctrinally opposed to each other is not just confined to the matter of explaining the relationship between these two scholars; more importantly, it is also associated with the broader and crucial question of whether the Yogācāra and Madhyamaka schools are fundamentally commensurable with each other.14

Indeed, in his commentary on the CWSL, the Sŏng yusingnon hakki 成唯識論學記 (Study notes to the CWSL; hereafter, Hakki), the Silla Yogācāra scholiast Taehyŏn 大賢 (ca. eighth century) introduces three distinct views by contemporary Buddhist scholars on the putative tension between Dharmapāla and Bhāviveka, suggesting that this matter was conceived as one of the major doctrinal problems by the ancient East Asian Buddhists as well. Unlike Kuiji who reads Dharmapāla’s statements in the CWSL as adversarial criticism of Bhāviveka, Taehyŏn, somewhat paradoxically, argues that they were in dispute in terms of language while their ultimate aims were the same, as I shall discuss below. A typical interpretation of Kuiji’s Yogācāra school as the ‘‘orthodox’’ in East Asian Buddhist tradition has tended to brand other scholastic views (although they are also classified under the same scholastic designation of the Dharma Characteristics school) as ‘‘non-orthodox’’ and often entailed a failure to give proper attention to such a scholar as Taehyŏn.15 However, it has often been pointed out that historical and theoretical problems are inherent in such a framed understanding of East Asian Yogācāra.16 In this regard, it would also be worthwhile to examine Taehyŏn’s accounts for the three types of view by East Asian commentators on this long-discussed problem of how to understand the disparity between Dharmapāla and Bhāviveka’s doctrinal positions.

Taehyŏn on the Dispute between Dharmapāla and Bhāviveka

Although he is generally regarded as a monk of the Dharma Characteristics school, Taehyŏn is known to have held somewhat differing doctrinal positions from Kuiji.17 In his commentary of the CWSL, Taehyŏn cites previous Faxiang or Pŏpsang scholars, among whom Kuiji is referred to the most. However, Taehyŏn’s accounts at the beginning of the CWSL of the presumed disputes between Dharmapāla and Bhāviveka clearly represent a different view from that of Kuiji. After briefly describing Dharmapāla and Bhāviveka’s doctrinal positions, Taehyŏn introduces three distinct views on whether there was a controversy between them: (1) the view that there was a controversy between them, based on the contrasting explanations in their doctrines; (2) the view that there was no controversy between them, in consideration of the identity of what they ultimately meant; and (3) the view that there was a controversy from the perspective of their different terminologies, but that the ultimate intent beyond their expressions was the same. Among these, Taehyŏn defends the third—the view that assimilates the first and second views. Let us look at the text for the three views in turn, starting with the first:

It is said that there was a debate between these two exegetes. This is because the Fodijing lun (*Buddhabhūmisūtra-śāstra) states, referring to this: ‘‘A thousand years [after the Buddha had taught the Dharma], there was a debate over emptiness and existence (K. kongyu chaengnon 空有諍論) in the Mahāyāna.’’18 What kind of debate? As for the conditioned [[[dharmas]]] (K. yuwi 有爲, Skt. saṃskṛta), the CWSL says: ‘‘Self and dharmas are not existent; emptiness and consciousness are not non-existent; being free from [both of] the existence and non-existence, one can accord with the middle way.’’19 This means that the attachment [of the imaginary nature (K. pyŏn’gye sojip sŏng 遍計所執性, Skt. parikalpita-svabhāva)] is discarded and the two remaining natures [that is, the dependent nature (K. ŭit’aki sŏng 依他起性, Skt. paratantra-svabhāva) and the perfected nature (K. wŏnsŏng silsŏng 圓成實性, Skt. pariniṣpanna-svabhāva)] are preserved. [By contrast,] the ZZL says: ‘‘In order to eliminate the mistake of falling into the extreme view of eternalism, it is said that that [viz., self and dharmas] is non-existent; in order to remove the mistake of falling into the extreme view of nihilism, it is said that this [viz., emptiness and consciousness] is existent. This means that eyes and so forth, which are produced by the power of causes and conditions, belong to the conventional truth and are possessed of their inherent natures. [Thus] they are not the same as sky-flowers, which are entirely non-existent. Only at the level of the true nature (K. chinsŏng 眞性, Skt. tattva), they are established as empty.’’20 This means that the conventional truth is preserved and at the level of the ultimate truth, all are empty.

Also, as for the unconditioned [[[dharmas]]] (K. muwi 無爲, Skt. asaṃskṛta), the two exegetes’ explanations are not the same. Bodhisattva Dharmapāla states in the CWSL as to Bhāviveka’s doctrine that ‘‘two types of emptiness [i.e., emptiness of self (K. agong 我空, Skt. ātma-śūnyatā) and emptiness of dharmas (K. pŏpkong 法空, Skt. dharma-śūnyatā)] directly refer to thusness’’ (K. igong chŭkchin 二空即真): ‘‘‘The nature’ [in the twenty-first verse of the Triṃśikā]21 implies that the two types of emptiness [[[per se]]] do not constitute the perfected nature, because thusness is the nature of transcending both existence and non-existence.’’22 [By contrast,] Bhāviveka Bodhisattva says in the ZZL as to Dharmapāla’s doctrine that ‘‘[[[thusness]] is] what is revealed by the two types of emptiness’’ (K. igong sohyŏn 二空所顯): ‘‘Only with nothing to be attached, thusness is established.’’23 It was not only regarding [the problem of] revealing the essence that the two [[[Wikipedia:Exegesis|exegetes]]’] views were different; also on [the problem of] whether the ultimate truth is existent or non-existent, [their views] were in disagreement and disputes. As the CWSL says: ‘‘If this consciousness were not existent, there would be no conventional truth. Without the conventional truth, there would be no ultimate truth. To abolish and void both truths is a ‘wrong attachment to emptiness’ (K. akch’wi kong 惡取空, Skt. durgṛhītā śūnyatā).’’24 [By contrast,] the ZZL says: ‘‘The Buddha taught at the level of the conventional truth that nirvāṇa is existent. This is just like the Buddha’s teaching that the miraculously transformed (K. hwasaeng 化生, Skt. upapāduka) sentient beings are existent. It is not that we commit a fault of contradicting our tenet just because we admit their existence. Only at the level of the true nature, ‘analytical cessation’ (K. t’aekmyŏl 擇滅, Skt. pratisaṃkhyānirodha; viz. nirvāṇa) is excluded and denied (K. ch’ap’a 遮破).’’25 Taking [all] these as evidence, such scholiasts as Wŏnch’ŭk 圓測 (613–696) say that there really were disputes.26

The view described in this passage takes the position of the Fodijing lun, the position that the debate between Dharmapāla and Bhāviveka actually took place a thousand years after the Buddha had appeared in this world. Dharmapāla’s and Bhāviveka’s positions on the three doctrinal points are also contrasted by comparing the citations from the CWSL and ZZL. Concerning the conditioned dharmas, the CWSL takes the position that among the three natures only the imaginary nature is not existent while the two other natures are preserved; the ZZL asserts that all three natures are empty at the level of the ultimate truth, but accepts the existence of all of them at the level of the conventional truth. In terms of the unconditioned dharmas, the CWSL states that thusness does not refer to emptiness per se, but the nature of emptiness, whereas the ZZL claims that emptiness directly constitutes thusness, saying that it is only when there is nothing to be attached that thusness is established. As for the matter of whether the ultimate truth is existent or non-existent, the CWSL criticizes the view that not only the conventional truth but also the ultimate truth is not existent, whereas the ZZL denies even nirvāṇa at the level of the true nature. In this view, based on all these doctrinal differences, Dharmapāla and Bhāviveka are considered to have been in an actual controversy. Wŏnch’ŭk is mentioned in this passage as one of those who take this position.

In contrast to the first view, the second view argues that there was no debate at all between the two scholars: Some say that there was absolutely no dispute between the two masters. This is because Bhāviveka did not admit the ‘‘non-existence’’ at the level of the ultimate truth. For instance, the ZZL says: ‘‘‘This [viz., ultimate truth] is not existent’ is said just to exclude (Ch. zhe 遮) the nature of existence. The exhaustion of functionality (Ch. gongneng 功能, Skt. samartha) does not directly explains the ‘non-existence.’ This is because, just as common people say, ‘non-white silk’ does not necessarily represent that it is black.’’27 In addition, Bhāviveka resolves a criticism, saying: ‘‘Again, he [viz., a critic] says: ‘If the point of your tenet is that at the level of the true nature all the conditioned dharmas do not exist at all, then it would be nothing but negating all dharmas by following erroneous views.’ [However,] the tenet of this teaching is to discriminate no nature of emptiness from the delusive manifestations, not to disregard all kinds of dharmas and regard them as nothing.’’

Dharmapāla also did not admit the ‘‘existence’’ at the level of ultimate truth. For instance, in the DGS he says: ‘‘Even what exists at present is not existent at the level of ultimate truth, since it is produced by conditions just as are such things as an illusory phenomenon.’’29 He also says: ‘‘‘Emptiness’ is [a word for] excluding/negating (Ch. zhe 遮), not indicating (Ch. biao 表) [something]. [This word] not only empties ‘existence,’ but also empties ‘emptiness,’’’30 and continues the explanation in detail. [Furthermore,] the Yogācāra master (K. sangŭng nonsa 相應論師) refuted in the ZZL is not Dharmapāla. This is because the Yogācāra master whom Bodhisattva Dharmapāla refuted in the DGS also appears the same as that [[[Yogācāra]] master whom Bhāviveka refuted]. Taking [all] these as proof, such masters as Sun’gyŏng 順憬 (alt. 順璟 ca. seventh century) argued that there was no dispute.31 In this second view, it is argued that there was no debate on the grounds that the positions of the two commentators are not opposed to each other at the level of the ultimate truth. According to this view, although Bhāviveka describes the ultimate truth as ‘‘non-existent’’ or ‘‘empty,’’ this does not mean that Bhāviveka admits the ‘‘non-existence’’ of the ultimate truth or thusness in the sense of ‘‘nothing.’’ In other words, the negative expression of the ‘‘nonexistence’’ does not indicate ‘‘nothingness’’ as a particular referent, but is meant just to exclude the delusive nature of the existence. Likewise, even if Dharmapāla expresses the ultimate truth or thusness as ‘‘existent,’’ this does not imply that he accepts the ‘‘existence’’ of the ultimate truth, because ‘‘emptiness’’ at the level of the ultimate truth also empties the emptiness. Moreover, it is argued that the Yogācāra master, who is disputed in the ZZL and has been regarded as such a Yogācāra master as Dharmapāla due to the doctrinal similarity, is not Dharmapāla by pointing out that Dharmapāla too refutes in the DGS a Yogācāra master in the same way. In short, this view claims that the two scholars’ fundamental intent was identical at the level of the ultimate truth, although they used different terminologies at the conventional level.

If the two commentators intended to mean the same thing at the ultimate level, does this mean then that we can say that their doctrinal systems were in complete agreement? Their views, as seen in the first view, are full of the distinct elements and conflicting points and it does not seem that, at least in terms of particular doctrinal points, they were in complete agreement. Further questions may then arise: Is their consensus at the ultimate level justified to nullify all the apparent theoretical dissimilarities and contrasts of their views? In this view, the contrasting aspects in their doctrines are disregarded while focusing on the two scholars’ agreement at the ultimate level. If this means that all the particular distinctions of their doctrines are equalized by the higher end of the ultimate equality, then what would be the point of the practical differences of their doctrinal systems? Furthermore, if the ultimate end subsumes and makes void all particular differences at the conventional level, then what would be the difference between this and the ‘‘monistic one,’’ which encompasses everything under its authoritative power—the very concept Buddhism rejects? It seems that in this regard the fact that all the doctrinal conflicts and tensions are resolved at the supreme level should be considered a separate matter from the idea that all doctrinal variations and disparate elements are dismissed under the supreme aim. Viewed from all these, the second view appears to focus specifically on the two scholars’ agreement at the ultimate level and thereby overlook the differences in their particular doctrines. The third view argues that although Dharmapāla and Bhāviveka were at odds in their expressions of their positions, what they meant was ultimately the same. This view combines the first view that sees the two scholars as doctrinally antagonistic on the grounds of their contrasting doctrines and the second view that regards them as agreeing with each other at the ultimate level. Some say that these two exegetes were in dispute in terms of language but their aims were the same. It is like debating about a stupa, which is rough on the bottom and fine on the top, because one’s position can be established only by acknowledging the position of the other. In Dharmapāla’s teaching, only based on the ‘‘non-existence’’ of what are attached, the tetralemma (K. sagu 四句, Skt. catuṣkoṭi) [viz., the four forms of views on what are attached,] is left behind. This is because such nature as ‘‘emptiness’’ or ‘‘existence’’ are all what are attached [expressed in the form of the tetralemma]. [However,] the two natures [of ‘‘emptiness’’ and ‘‘existence’’] are mysteriously existent (K. myoyu 妙有) because they are not completely non-existent. For this reason, [[[Dharmapāla]]] says that the two types of emptiness [[[per se]]] are not thusness (K. igong pijin 二空非眞).32 The ‘‘Emptiness’’ is one extreme [view] and thus is not empty [in the real sense]. Therefore, when [the linguistic] paths of the ‘‘emptiness’’ or the ‘‘existence’’ are all severed, it is [just] called thusness. [In case of] Bodhisattva Bhāviveka, based on the ‘‘existence’’ of the conventional truth, all kinds of ‘‘non-existence’’ [,which is a form of the tetralemma,] are left behind and [thereby] the true ‘‘non-existence’’ is distinguished from them. This is because the conventional truth is [not only ‘‘existence’’] but also ‘‘non-existence’’ [in the ultimate sense]. [However,] the two natures [of ‘‘existence’’ and ‘‘non-existence’’] are mysteriously nonexistent (K. myomu 妙無) because they are unobtainable (K. musodŭk 無所得). If only the ‘‘existence’’ [of the tetralemma,] is discarded, the ‘‘non-existence’’ [of the tetralemma,] can be directly obtained; and yet the ‘‘non-existence’’ [of the tetralemma] is discarded as well and it is thus said to be ‘‘unobtainable.’’ This ‘‘unobtainability’’ means being free from the tetralemma. Asaṅga (ca. fourth century) states in the Vajracchedikā-prajñāpāramitōpadeśa that the tetralemma is all subsumed under the attachment to dharmas (K. pŏpchip 法執, Skt. dharma-grāha).33 On the grounds of this correct argument [of this third view], such masters as Wŏnhyo 元曉 (617–686) [stated that] while they disputed in terms of their expressions, their aims were the same. This is because of the purpose that students of dull faculties of the latter age can skillfully produce understanding on the basis of these disputes.

In this view, Dharmapāla and Bhāviveka are opposed in terms of their vocabularies in expressing their doctrines, and yet, at the ultimate level, they have the same message to deliver. According to Dharmapāla, the four forms of attached views—namely, the tetralemma—can be abandoned only on the basis of understanding that what are attached by us as existent are actually ‘‘non-existent.’’ However, this notion of the ‘‘non-existence’’ of what we attach as existence is, in turn, also a form of the tetralemma and should be abandoned as well. It eventually turns out that the ‘‘non-existence’’ of the ‘‘non-existence of what are attached’’ is not the opposite of the ‘‘existence’’ in the tetralemma, but, Taehyo˘n says, ‘‘what is mysteriously existent.’’ By contrast, Bhāviveka argues that it is by depending on the ‘‘existence’’ of the conventional truth that we can discard all ‘‘non-existence’’ of the tetralemma and finally realize the true ‘‘non-existence’’ at the ultimate level. Generally speaking, the ultimate truth is obtained by transcending the conventional truth. Since, Bhāviveka says, the conventional truth is ‘‘existent,’’ the ultimate truth would be reached by abandoning the ‘‘existence’’ of the conventional truth. The ultimate truth, however, cannot be reached just by discarding the concept of the ‘‘existence’’ of the conventional truth; if only the ‘‘existence’’ of the conventional truth is denied, then the truth obtained as a result would be characterized with the nature of ‘‘non-existence,’’ ending up with a transition from ‘‘existence’’ to ‘‘non-existence,’’ one form to another in the tetralemma. The conventional truth, however, is not just existent; it is non-existent in the ultimate sense, because it is ‘‘conventional’’ truth. Thus, the conventional truth is transcended by abandoning not only the ‘‘existence’’ but also the ‘‘non-existence’’ of it. In this sense, the ultimate truth is ‘‘unobtainable’’ and also described as ‘‘mysteriously non-existent.’’

In this passage, Dharmapāla’s view starts with the ‘‘non-existence’’ (of what are attached) and finally arrives at ‘‘what is mysteriously existent,’’ by transcending both ‘‘existence’’ and ‘‘non-existence.’’ By contrast, Bhāviveka’s position begins with the ‘‘existence’’ (of the conventional truth) and ends up with ‘‘what is mysteriously non-existent’’ by, as in Dharmapāla’s case, transcending both ‘‘existence’’ and ‘‘non-existence.’’ Although the two scholars’ starting points and final destinations are expressed in contrasting vocabularies, it seems clear that they both seek to explain the ultimate truth as transcending the ontologically definitive notions, the ‘‘existence’’ and the ‘‘non-existence.’’ Taehyŏn attributes this third view to Wŏnhyo,35 and he himself agrees with this view.

Then, why should they use differing linguistic expressions, even though what they mean in the ultimate sense was the same? If they were common in considering thusness as transcending both concepts of ‘‘existence’’ and ‘‘nonexistence,’’ then why did they describe thusness in the very different or even contrasting words—one denying that the two types of emptiness directly correspond to thusness, while the other arguing that they refer to none other than thusness? One thing to be considered in resolving these questions is that very often the two scholars’ views are not compared at the same level. In other words, in most cases Bhāviveka’s view at the level of the ultimate truth is compared to Dharmapāla’s view at the level of the conventional truth. For instance, Bhāviveka’s negation of all three natures is often opposed to Dharmapāla’s partial negation—the negation of only the existence of the imaginary nature among the three. What should be noted here is that Bhāviveka negates all the three natures just at the level of ultimate truth;36 at the level of conventional truth he accepts all three natures, somehow representing similarity to Dharmapāla’s view that accepts the dependent nature and the perfect nature. Meanwhile, in the CWSL Dharmapāla, similarly to Bhāviveka, accepts that thoughts and languages are all cut off at the level of true ultimate truth (Ch. zhen shengyi 眞勝義).37 Moreover, it is clearly stated in the CWSL that the whole doctrinal system on the eight consciousnesses is explained from the perspective of the ‘‘conventional truth based on doctrinal teachings’’ (Ch. li shisu 理世俗, Skt. yukti-saṃvṛti).38 Then it seems that although they ultimately have the same intent, they express their identical intent in different doctrinal systems by focusing on different levels of truth. In addition, in the above quoted three views on the relationship between Dharmapāla and Bhāviveka, each of the scholars are described as admitting the other’s position from another perspective than his own. For instance, as seen in the second view, Dharmapāla, although normally arguing for the ‘‘existence,’’ does not admit the ‘‘existence’’ from the perspective of ultimate truth. Likewise, Bhāviveka, who generally discusses the three natures by focusing on the level of the ultimate truth, also considers them at the level of the conventional truth. In this light, it may be seen that their distinct views on the three natures, as Taehyŏn indicates, do not represent a fundamental disparity between their doctrinal positions; they are different expressions of the same intent, based on the different perspectives of truth on which they depend for their doctrines.

Taehyŏn’s view on the doctrinal tension between Dharmapāla and Bhāviveka is twofold, affirming the doctrinal difference or contrast between them on the one hand and denying fundamental inconsistency on the other, unlike Kuiji who considers only the former. Taehyŏn’s view, in this regard, does not conflict with the two previous views, especially the first view that regards the two scholars’ positions as antagonistic. The statement of the third view, ‘‘these two exegetes were in dispute in terms of language,’’ should not be regarded as meant in a rhetorical sense that, say, the ways in which they express their ideas sound like disputes; rather, the opposition in their views should be taken in a semantic sense. Indeed, in the Hakki, Taehyŏn does not completely compromise or identify the two scholiasts’ views; he acknowledges the doctrinal disagreement between them. For instance, in commenting on the passage of the CWSL, in which ‘‘those who have an attachment to the Mahāyāna teaching of emptiness of no characteristics (Ch. xiang 相, Skt. lakṣaṇa) and regard it as the ultimate [[[truth]]]’’ (有執大乘遣相空理為究竟者) are criticized for their denying ālayavijñāna and all dharmas by using pseudo-inference,39 Taehyŏn declares that this is intended for refuting the doctrines of Bhāviveka.40 However, even if Taehyŏn recognizes the doctrinal contradiction between Dharmapāla and Bhāviveka, he does not take the position of supporting one view and rejecting the other. Unlike Kuiji, who condemns Bhāviveka, Taehyŏn shows no explicit criticism of him; Taehyŏn just ignores41 or disagrees42 with Kuiji’s criticism of Bhāviveka. This suggests that Taehyŏn, although perceiving the doctrinal contrast between the two commentators, did not further take this expressive contradiction as a fundamental inconsistency in their message. It seems that, for Taehyŏn, whether their views are expressed to each other in consistent words and whether each of the two views is a valid and coherent teaching that eventually leads sentient beings to the ultimate goal are two disparate matters.

Connecting the Conditioned and the Unconditioned

Taehyŏn’s way of understanding the relationship of Dharmapāla and Bhāviveka— the understanding beyond the literal meanings of their contrasting doctrinal expressions—is also reflected in his interpretation of another dilemmatic problem, the problem of whether, or how, the ‘‘unconditioned’’ (K. muwi 無爲, Skt. asaṃskāra) is connected to the ‘‘conditioned’’ (K. yuwi 有爲, Skt. saṃskāra). In association with the soteriological problem of how the mundane (K. segan 世間, Skt. loka) is connected to the supramundane (K. ch’ul segan 出世間, Skt. lokottara), or how a sentient being in illusion becomes a buddha, the enlightened, this problem has been a controversial issue, on which various interpretations were offered by Buddhist scholiasts according to their distinct scholastic positions. Although this problem does not directly address the current issue of whether Dharmapāla and Bhāviveka’s views were consistent, it would be worth discussing in that, just as in the previous case, Taehyŏn does not take literal meanings of given expressions, but rather seeks to find another way to explain the seemingly contradictory concepts.

The problem emerges when Taehyŏn comments on the passage of the CWSL, in which amalavijñāna, or the ‘‘immaculate consciousness,’’ is presented as one of the various names for the eighth consciousness, i.e., the fundamental consciousness of sentient beings. In general, such an early Yogācāra text as the Yogācārabhūmi or the CWSL are considered to take a position that the unconditioned and the conditioned are two separate abodes. According to the Yogācārabhūmi, thusness, as an unconditioned dharma at the absolute level, cannot be a generative cause of anything. In other words, thusness cannot directly generate anything, including supramundane wisdom (Ch. chu shijian zhi 出世間智, Skt. lokōttara-jñāna), which is required for a sentient being to attain Buddhahood (Yamabe 1997, 198–199).43 In the CWSL as well, the basis of the supramundane achievement is found within the conditioned abode; a sentient being can attain supramundane wisdom on the basis of the ‘‘uncontaminated seeds’’ (Ch. wulou zhongzi 無漏種子, Skt. *anāsrava-bija), which is a kind of conditioned dharma inherent in a sentient being’s mind. In accordance with this position, Kuiji explains amalavijñāna, which is identified with the [[[Wikipedia:perfect|perfect]]] mirror wisdom (Ch. da yuanjing zhi 大圓鏡智, Skt. ādarśa-jñāna),44 as a conditioned dharma, because Kuiji regards this mirror wisdom, the supreme consciousness attained by a sentient being by transforming their fundamental consciousness, ālayavijñāna, as belonging to the conditioned abode, on the grounds that it is originated from the fundamental consciousness of sentient beings of saṃsāra. However, the position that the unconditioned and the conditioned are completely separated contains a theoretical problem. If the basis of the supramundane achievement should be found in the conditioned, or if the mirror wisdom, the supreme consciousness of sentient beings, is regarded as a conditioned dharma, then how can we explain a sentient being’s spiritual transition from the mundane to the supramundane? In addition, the idea that the supramundane elements originate from the conditioned dharmas, which are, in this case, the uncontaminated seeds inherent in a sentient being’s mind, carries the implication that the conditioned dharmas generate the unconditioned dharma—an implication that is contradictory to the premise that the unconditioned and the conditioned are two separate abodes.

Unlike such a typical position presented in the CWSL, Taehyŏn presents a rather paradoxical argument that amalavijñāna may be considered as a conditioned dharma on the one hand and an unconditioned dharma on the other, by attributing amalavijñāna to not only the mirror wisdom but also thusness, as follows: Kuiji says in his commentary: ‘‘The immaculate [[[consciousness]]] (Ch. wugou shi 無垢識, Skt. amalavijñāna) was previously called ‘amoluo consciousness’ or ‘amaluo consciousness.’ In the past it was established as the ninth consciousness, but this was a mistake.’’46 In the Shuyao he [also] says: ‘‘According to the Wuxiang lun 無相論 and the Mahāyānābhisamaya-sūtra (Ch. Dasheng tongxing jing 大乘同性經), the immaculate consciousness is the ‘inherently cognizing mind’ (Ch. zixing shixin 自性識心, Skt. *svabhāvikavijñāna-citta), which parallels the principle of thusness (Ch. zhenru li 眞如理, Skt. tathatā).47 Thus it should be known that the immaculate consciousness may work on both types.’’48 Wŏnch’ŭk says: ‘‘In the Wuxiang lun, Paramārtha made a mistake.’’ Now I respect [the passage cited in] the Shuyao. This [CWSL] states: ‘‘[[[Tathāgata’s]] immaculate consciousness] corresponds to the [[[Wikipedia:perfect|perfect]]] mirror wisdom (Ch. da yuanjing zhi 大圓鏡智, Skt. ādarśajñāna),’’49 and thus we see that it is a conditioned [[[dharma]]]. [On the other hand,] in the sixth fascicle of the Mahāyānasūtrā-laṃkāra it is stated: ‘‘Here we should know that the statement of ‘as for the mind, thusness is named as mind’ means that this mind refers to the ‘innately pure [[[mind]]]’ (Ch. zixing quingjing [xin] 自性淸淨[[[心]]], Skt. prakṛti-prabhāsvara[-citta]). This mind is none other than amalavijñāna.’’50 [Thus] we see that [[[amalavijñāna]]] is an unconditioned dharma [as well].’’

The CWSL, as Taehyŏn mentions, states that amalavijñāna corresponds to the perfect mirror wisdom, implying that it is a conditioned dharma, since the mirror wisdom is generally considered a conditioned dharma since it originates from ālayavijñāna, another conditioned dharma, as the fundamental mind of sentient beings. However, Taehyŏn also introduces the other texts that, unlike the CWSL, address amalavijñāna as thusness, which is an unconditioned dharma. Paramārtha’s Wuxiang lun and the Mahāyānābhisamaya-sūtra, Taehyŏn mentions, identify the immaculate consciousness to the ‘‘inherently cognizing mind’’ (Ch. zixing shixin 自性識心, Skt. *svabhāvika-vijñāna-citta), which parallels the principle of thusness. Likewise, the Mahāyānasūtrālaṃkāra states that amalavijñāna refers to thusness, suggesting that it belongs to the unconditioned. This implies that Taehyŏn considers amalavijñāna as not only a conditioned dharma but also an unconditioned dharma by drawing upon two seemingly contradictory views on amalavijñāna.

Then, the question would be: how can amalavijñāna be considered as a conditioned dharma on the one the hand and an unconditioned dharma on the other? How can the doctrinally conflicting notions of the conditioned and the unconditioned be concurrently involved in the unitary concept of amalavijñāna in Taehyŏn’s understanding? Unfortunately, there is no extant work by Taehyŏn that provides a direct answer to this question.52 However, what may be said at least for now is that Taehyŏn, here again, accepts both of the two seemingly contradictory positions—the position of the CWSL that describes amalavijñāna as a conditioned dharma and the position of the other texts that attribute it to an unconditioned dharma, viz., thusness. Although these two positions look opposed and conflicting, Taehyŏn neither makes an alternative judgment between them, nor interprets them as mutually contradictory, especially in understanding such a concept beyond our normal intelligence as amalavijñāna. Rather, he seems to consider the two distinct positions as constituting parts of a coherent whole of a Buddhist system. If one focuses on the expressive inconsistency between the two positions on amalavijñāna, then s/he should regard it as either a conditioned dharma as presented in the CWSL or an unconditioned dharma as in the Mahāyānasūtrā-laṃkāra or the Wuxiang lun, because these two abodes are ‘‘inconsistent’’ in the form of the tetralemma. However, if one counts on the need to see beyond their expressive contrasts, s/he would not exclude the possibility that the two positions, despite their evident doctrinal or linguistic conflicts, may be understood as compatible from a broader or higher perspective beyond our daily logic.

Conclusion

In East Asia, there were distinct views on whether Dharmapāla and Bhāviveka disputed each other, taking note of Dharmapāla’s arguments as presented in the CWSL. In addition to the easily anticipated views, such as whether or not the two exegetes’ positions are opposed, the Silla Yogācāra scholar Taehyŏn introduces a view that accepts both of the seemingly contradictory positions. In this view, the positions of the two Indian scholars are considered in terms of two levels. At the level of conventional truth, which is the abode of language, the linguistic expressions used to describe their positions are inevitably represented as opposed, while at the level of the ultimate truth transcending linguistic expression, the fundamental meaning that they intend to deliver is considered to be the same. In this view, the theoretical differences between the two scholars are not reduced to a matter of choosing one or the other. The theoretical differences, despite their seeming contradiction, may be understood as distinct expressions of the Buddha’s teaching, each having its own doctrinal validity and heuristic implication. Such a view by Taehyŏn is also reflected in his understanding of the concept of amalavijñāna, the most supreme consciousness that sentient beings can ever have. Unlike the typical understanding that the conditioned and the unconditioned are two separate abodes, Taehyŏn argues that amalavijñāna may be counted as a conditioned dharma on the one hand and an unconditioned dharma on the other. For Taehyŏn, who sees no conflict between the distinct doctrines at the ultimate level, the alternative separation of the conditioned and the unconditioned in understanding amalavijñāna, the supreme consciousness at the edge of the two abodes, no longer seems valid.

Taehyŏn’s approach to compromising the contrasting doctrines should be distinguished from another way of compromise, in which distinct elements of the doctrines are ‘‘equalized’’ in one unitary view. In this approach, the individual particularities are very often disregarded and subsumed within the unitary whole, in the name of ‘‘compromise’’ or ‘‘harmony,’’ and at the end only the one whole remains. In Taehyŏn’s approach, in which the ultimate commonality and the conventional particularities are both appreciated, the distinct elements coexist while each having its own validity, regardless of their seeming conflicts.

Notes

1 It is known that Xuanzang traveled to India and at Nālandā monastic university studied Vasubandhu’s Triṃśikā and its commentaries by ten Indian scholiasts— Dharmapāla, Guṇamati, Sthiramati, Bandhuśrī, Nanda, S´uddhacandra, Citrabhānu, Viśeṣamitra, Jinaputra, and Jñānacandra. Among these scholiasts, Dharmapāla was the teacher of S´īlabhadra (529–645), a Yogācāra master under whom Xuanzang studied at Nālandā.

2 See Shuyao (T 1831.43.608b29–c14).

3 Bhāviveka’s verse at the beginning of the ZZL was considered by contemporary Buddhist scholiasts as doctrinally contrasting to Dharmapāla’s position in the DGS. The verse of the ZZL reads: ‘‘At the level of true nature (Ch. zhenxing 眞性, Skt. tattva) [viz., the ultimate truth (Ch. shengyi di 勝義諦, Skt. paramārtha-satya)], the conditioned [[[dharmas]]] are empty, since they dependently arise like an illusion. The unconditioned [[[dharmas]]] do not have reality (Ch. shi 實, Skt. bhūta); they do not arise, like illusory flowers in the sky’’ (真性有爲空如幻緣生故無爲無有實不起似空華) (ZZL T 1578.30.268b21–22). This verse is regarded as implying that all the three natures (Ch. sanxing 三性, Skt. tri-svabhāva) are empty, in opposition to Dharmapāla’s view that that only the imaginary nature (Ch. pianji suozhi xing 遍計所執性, Skt. parikalpita-svabhāva) among the three is empty. Furthermore, the *Buddhabhūmisūtra-śāstra (Ch. Fodijing lun 佛地經論), which had been translated by Xuanzang in 649, states that the controversy between the doctrines of ‘‘emptiness’’ and ‘‘existence’’ occurred a thousand years after the Buddha had taught the Dharma

(Fodijing lun T 1530.26.307a5–9). It is not certain, however, that the actual debate between the two Indian exegetes happened, since Xuanzang reports in the Da Tang xiyuji 大唐西域記 that Bhāviveka visited Dharmapāla to have a discussion with him, but Dharmapāla refused to meet him for the reason that he was fully occupied with cultivation (Da Tang xiyuji T 2087.51.930c25–931a8). But it has also been pointed out that even if they did not have an actual argument, there are numerous literary materials that enable us to presume that they still had conflicting views. For this point, see Tillemans (1990, 54n114). For further discussion on the ‘‘emptinessexistence controversy,’’ see Lee (2016).

4 CWSL (T 1585.31.1a15).

5 述曰。此第二計。即學中百清辨等師。依密意教說諸法空。便亦撥心體非實有 (Cheng weishi lun shuji 成唯識論述記 [hereafter, Shuji; T 1830.43.236c3–5]).

6 有執大乘遣相空理為究竟者。依似比量撥無此識及一切法 (CWSL [T 1585.31.16a6–8]). 7 論。有執大乘至及一切法。述曰。第五清辨無相大乘。於俗諦中亦說依他圓成有故。真諦皆空故。今言空者遣遍計所執。彼執此文為正解故。彼依掌珍真性有為空等似比量。撥無此識及一切法皆言無體。言似比量者。謂約我宗真性有為無為非空不空。有法一分非極成過。 (Shuji [T1830.43.359a1–7]).

8 由斯遠離增減二邊。唯識義成契會中道 (CWSL [T 1585.31.39a3–4]).

9 論。由斯遠離至契會中道。述曰。總結。由此二文三師理故遠離二邊。無心外法故除增益邊。有虛妄心等故離損減邊。離損減邊故。除撥無如空花清辨等說。離增益邊故。除心外有法諸小乘執。唯識義成契會中道。無偏執故。言中道者正智也。理順正智名契會中道 (Shuji [T 1830.43.488a8–14]).

10 此中應言三故字。謂有故即妄分別。無故即能所取。及有故即俗空互有。... 是則契中道者。謂非一向空如清辨。非一向有如小乘故。名處中道。謂二諦有不同清辨。二取無不同小部。故處中道 (Shuji [T 1830.43.490b 11–18]).

11 此識若無便無俗諦。俗諦無故真諦亦無。真俗相依而建立故。撥無二諦是惡取空。諸 佛說為不可治者應知諸法有空不空。由此慈尊說前二頌 (CWSL [T 1585.31.39b17–20]).

12 論。撥無二諦至說前二頌。述曰。若撥無識及性。即撥無二諦。佛說為不可治者。沈淪生死病根深故。即清辨等。 (Shuji [T 1830.43.492b25–27]). In commenting on another passage of the CWSL, in which the ‘‘wrong attachment to emptiness’’ is mentioned (CWSL [T 1585.31.39c25–28]), Kuiji also attributes it to Bhāviveka (Shuji [T 1830.43.494b25–26]).

13 In some of his other works as well, Kuiji expresses a strong criticism of Bhāviveka. For instance, the Bian zhongbianlun shuji 辨中邊論述記, the commentary on the Madhyānta-vibhāga, and the Weishi ershilun shuji 唯識二十論述記, the commentary on the Viṃśatikā, also represent his criticism. However, such a harsh criticism by Kuiji toward Bhāviveka appears to have changed to a sort of moderate approval around the time just after Xuanzang translated the Mahāprajñāpāramitā-sūtra. For further discussion of Kuiji’s change of attitude toward Bhāviveka, see Lee (2016).

14 For further discussions on this important issue from various perspectives, see Garfield and Westerhoff (2015). 15 There is a scholarly tendency to divide the Dharma Characteristics school into two strands, the ‘‘orthodox’’ li ne based on Kuiji’s views and the other based on those of Silla expatriate monk Wŏnch’ŭk 圓測 (613–696). These two scholastic lines are called the Ci’en 慈恩 school and Ximing 西明 school, respectively. Taehyŏn has often been considered to have inherited Wŏnch’ŭk’s scholastic position, based on Japanese monk Shō’on’s 照遠 (ca. fourteenth century) record in the Bonmōkyō gekan koshakki jutsu shakushō 梵網經下卷古迹記述迹鈔 that Taehyŏn was a disciple of Tojŭng 道證 (fl. 692), another Korean monk who was one of Wŏnch’ŭk’s disciples on the Chinese mainland (Bonmōkyō gekan koshakki jutsu shakushō, Nihon daizōkyō 日本大藏經, vol. 2, 1973–1978, 233a). However, recently it has also been indicated not only that Wŏnch’ŭk’s views are not necessarily opposed to Kuiji’s, but also that Taehyŏn himself does not always advocate Wŏnch’ŭk’s or Tojŭng’s views. For further discussion on the problem of identifying Taehyŏn’s scholastic position based on his presumed genealogical connection, see Lee (2014, 1–9).

16 For the historical as well as theoretical problems involved in defining the East Asian Yogācāra school under the umbrella term of ‘‘Dharma Characteristics school,’’

17 Taehyŏn, although being generally regarded as a monk of the Dharma Characteristics school, is also known to have defended such a doctrine as the universal ‘‘Buddhanature’’ (Ch. foxing 佛性, Skt. buddha-dhātu), the doctrine that is seen as contradictory to the ‘‘five lineages’’ (Ch. wu zhongxing 五種性, Skt. pañcagotra) theory advocated by the Dharma Characteristics school. Because of such an ambiguity in Taehyŏn’s doctrinal position, there have been controversies in determining his scholastic line, among both traditional Buddhist thinkers and modern scholars. For more information, see Ch’ae (1983, 105–106) and Pang (1995, 27–35).

18 The original text of the Fodijing lun states: 說佛正法但經千載，非佛教法但住千歲。又聲聞藏雖佛去世百年已後即分多部，而菩薩藏千載已前清淨一味無有乖諍，千載已後乃興空有二種異論。 (Fodijing lun [T 1530.26.307a5–9]).

19 我法非有空識非無。離有離無故契中道。 (CWSL [T 1585.31.39b1–2]).

20 如為棄捨墮常邊過說彼為無，亦為棄捨墮斷邊過說此為有，謂因緣力所生眼等，世俗諦攝自性是有，不同空華全無有物。但就真性立之為空 (ZZL [T 1578.30.272b2–5]). 21 The twenty-first verse says: ‘‘The dependent nature consists of discriminations generated by conditions. The perfected nature is eternally free from the previous nature (i.e., the imaginary nature) in this [[[nature]], i.e., the dependent nature]’’; 依他起自性，分別緣所生，圓成實於彼，常遠離前性 (CWSL [T 1585.31.45c10–11]).

22 性顯二空、非圓成實、眞如離有離無性故 (CWSL [T 1585.31.46b17–18]).
23 於唯無有一切所執立為真如 (ZZL [T 1578.30.274b9–10]).
24 此識若無便無俗諦，俗諦無故真諦亦無，真俗相依而建立故。撥無二諦是惡取空
(CWSL [T 1585.31.39b17–18]).
25 就世俗說有擇滅,出離涅槃寂靜微妙,如佛說有化生有情,說有無爲涅槃亦爾,許此有故無違宗過,但就眞性遮破擇滅 (ZZL [T 1578.30.274a22–24]).
26 有說。此二實有諍論。如佛地論。千年已後。大乘之中空有諍論。即此謂故。其諍云何。且有為中。唯識云。我法非有。空識非無。離有離無。契於中道。此遣所執。存餘二性。掌珍論云。如為棄捨墮常邊過。說彼為無。亦為棄捨墮斷邊過。說此為有。
謂因緣力所生眼等。世俗諦攝。自性是有。不同空華全無有物。但就真性。立之為空。此存世俗。勝義皆空。
又無為中二說不同。護法菩薩對清辨宗二空即真。唯識論云。性顯二空非圓成實。真如離有離無性故。清辨菩薩對護法宗二空所顯。掌珍論云。於唯無有一切所執立為真如。
非但出體二說不同。勝義有無。亦為乖諍。如唯識云。此識若無便無俗諦。俗諦無故。真諦亦無。撥無二諦。是惡取空。掌珍論云。佛就世俗說有涅槃。如佛說有化生有情。許此有故。無違宗失。但就真性遮破擇滅。以此為證。測等傳說實有諍論
(Hakki [X 818.50.027a13–b5]). For the translation of this part of Taehyŏn’s Hakki,
I have consulted Muller’s translation (2012, 134–143; sometimes modified).
27 The original passage in the ZZL is as follows: 此非有言、唯遮有性。功能斯盡、無有勢力。更詮餘義、如世間說、非白絹言不可卽執, 此言詮黑 (ZZL [T 1578.30.270c11–
13]).
28 又彼所言，若就眞性,一切有爲都無所有.是立宗義,卽謗一切皆無所有.如是所立, 墮邪見者. 此中宗義, 如前廣說, 謂空無性. 虛妄顯現門之差別, 非一切種皆謗爲無.
故汝不應作如是難 (ZZL [T 1578.30.270c1–5]).
29 DGS (T 1571.30.206c12–13).
30 又此空言,是遮非表,非唯空有,亦復空空 (DGS [T 1571.30.219b27–28]).
31 有說。二師都無諍論。清辨不許勝義無故。如掌珍云。此非有言。唯遮有性。功能斯盡。更不詮無。如世間說。非自絹言。未必彼言即詮黑故。又通難言。又彼所言。若就真性。一切有為都無所有。是立宗義。即謗一切隨邪見者。此中宗義。謂空無性虗妄顯現門之差別。非一切種皆謗為無。
護法勝義亦不許有。如廣百云。現在亦非勝義諦有。從緣生故。如幻事等。又說空言。是遮非表。非唯空有。亦復空空。乃至廣說。掌珍所破相應論師。非為護法。護法菩薩。廣百釋中破相應師。亦同彼故。為以此證。順憬師等。傳無諍論
(Hakki [X 818.50.027b5–15]).
32 Compare this to Bhāviveka’s view, as discussed in the first view above, that the two types of emptiness refer directly to thusness (K. igong chŭkchin 二空卽眞).
33 Not identified.
34 有說。此二語諍意同。如諍浮圖下麤上細。必由許他。自始成故。護法宗必舉所執無表離四句。空有等性。皆所執故。二性妙有。不全無故。由此說言。二空非真。空謂一邊。亦不空故。空有路絕。名真如故。
清辨菩薩。舉世俗有。離諸無。簡諸真無。俗亦無故。二性妙無。無所得故。若唯遣有。便可得無。亦遣無故。言無所得。無所得者。離四句義。無著般若論云。四句皆是法執攝故。由此正理。元曉師等。語諍意同。為末代鈍根之徒。依此諍論。巧生解故 (Hakki [X 818.50.027b15–24]).
35 Indeed, Wŏnhyo is well known for his paradoxical way of approach to Buddhist doctrines regarded as contradictory to each other, both affirming and negating them. For a full discussion of Wŏnhyo’s method of argumentation, see Muller
(2015).
36 At the level of the conventional truth, Bhāviveka accepts the conditioned and the unconditioned dharmas. See note 26 above.
37 如前所說識差別相，依理世俗，非真勝義。真勝義中心言絕故
(CWSL [T 1585.31.38c8–9]).
38 See note 37 above. In the early Yogācāra texts the problem of which perspective or level should be applied in composing or reading a canonical text does not seem to have been much considered. However, as the theoretical differences between Yogācāra and Madhyamaka became more prominent, the perspective from which a canonical text was described began to emerge as a matter to be considered when a text was composed, and the CWSL seem to be one of them. See Katsumata (1961, 195–196).
39 有執大乘遣相空理為究竟者。依似比量撥無此識及一切法 (CWSL [T 1585.31.16a6–8]). 40 述曰。第五破清辨宗遣所執相大乘為其了義撥一切法。 (Hakki [X 818.50.68c4–5]).
See footnote 7 above.
41 As discussed above, Kuiji relates the CWSL’s statement that rejection of both the ultimate truth and the conventional truth corresponds to a wrong attachment to emptiness (Ch. equ kong 惡取空) to Bhāviveka (See note 12 above), but in commenting the same passages, Taehyŏn does not mention Bhāviveka.
42 For the passage, in which Kuiji criticizes the pseudo-inference of Bhāviveka (See note 6 above), Taehyŏn disagrees with Kuiji by saying ‘‘The [explanation] as such is not reasonable’’ (此等非理);基云。彼掌珍頌。依似比量。撥無此識及一切法(不引中觀十八論)。似比量者。有法一分不極成過。空有真性。互不許故 (此等非理如本母釋) (Hakki [X 818.50.68c18–20]).
43 In the case, Yamabe says, where thusness is termed the cause of supramundane wisdom, this just means that thusness works as its cognitive object (Ch. suoyuanyuan 所緣緣, Skt. ālambana-pratyaya). However, he also indicates that from the perspective of Yogācāra history, we see many exceptions wherein thusness is identified with the unconditioned. See Yamabe (1997, 199n32).
44 The CWSL cites a verse from the Rulai kung te chuang yen ching 如來功德莊嚴經, which says that amalavijñāna corresponds to the [[[Wikipedia:perfect|perfect]]] mirror wisdom: 如來無垢識,是淨無漏界,解脫一切障,圓鏡智相應 (CWSL [T 1585.31.13c23-24]). The Rulai kung te chuang yen ching is not extant. Also, see the quoted passage from the Hakki below.
45 This conundrum is also represented in relation with the paradoxical problem implicit in the Mahāyāna concept of ‘‘nirvāṇa of no-abiding’’ (Ch. wuzhu[chu] niepan 無住[[[處]]]涅槃, Skt. apratiṣṭhita-nirvāṇa)—‘‘the problem of how a Buddha can be personally free from the cognitive conditions that imprison beings in samsara, yet actively participate within their [sic] cognitively conditioned realms of suffering until all are freed’’ (Makransky [1997, 214]). For further discussion on this issue, see especially chapters four and five of Makransky (1997).
46 論。或名無垢識至所依止故。述曰。唯無漏依。體性無垢。先名阿末羅識。或名阿摩羅識。古師立為第九識者。非也 (Shuji [T 1830.43.344c9–11]). Kuiji goes on to say: ‘‘This ‘immaculate consciousness’ is the name for the perfect mirror wisdom (Ch. da yuanjing zhi 大圓鏡智, Skt. ādarśa-jñāna), which is attained by transforming the mind essence of the eighth consciousness’’; 此無垢識。是圓鏡智相應識名。轉因第八心體得之 (Shuji [T 1830.43.344c12–13]).
47 This quotation cannot be identified in the extant portions of the Wuxiang lun or the Mahāyānābhisamaya-sūtra (Ch. Dasheng tongxing jing 大乘同性經). The Wuxiang lun is known as consisting of three texts: the San wuxing lun 三無性論, Zhuanzhi lun 轉識論, and Xianshi lun 顯識論.
48 依無相論同性經。無垢識是自性識心。則真如理。故知無垢通二種也 (Shuyao [T
1831.43.634c8–9]). The implication here is that the immaculate consciousness has twofold aspects of the consciousness (識) and the principle (理), or the conditioned and the unconditioned. In the Dasheng fayuan yilin zhang 大乘法苑義林章, Kuiji also cites the Wuxiang lun and the Mahāyānābhisamaya-sūtra with the similar implication: ‘‘The Wuxiang lun 無相論 and the Mahāyānābhisamaya-sūtra say that if thusness is regarded as the ninth [[[consciousness]]], this is because the ultimate and conventional (i.e., the unconditioned and the conditioned) are synthetically discussed. Now [the correct interpretation is that] the pure level of the eighth consciousness is considered as the ninth [[[consciousness]]] because the impure and pure states of the base consciousness are separately discussed’’; 無相論同性經中。若取真如為第九者。真俗合說故。今取淨位第八本識以為第九。染淨本識各別說故
(Dasheng fayuan yilin zhang [T 1861.45.261b17–19]).
49 See note 44 above.
50 已說心性淨而為客塵染不離心真如別有心性淨釋曰。譬如水性自清而為客垢所濁。如是心性自淨而為客塵所染。此義已成。由是義故。不離心之真如別有異心。謂依他相說為自性清淨。此中應知。說心真如名之為心。即說此心為自性清淨。此心即是阿摩羅識 (Dasheng zhuangyanjing lun [T 1604.31.623a3–9]).
51 基疏云。無垢先名阿末羅識。或阿摩羅識。古而立為第九識者。非也。樞要云。依無相論同性經。無垢識是自性識心。即真如理。故知無垢通二種也。測云。其無相論真諦謬 (今存樞要。此既云鏡智相應。知通有為。大莊嚴論第六云。此中應知。說心真如名之為心。即說此心為自性清淨。此心是阿摩羅識知通無為) (Hakki [X 818.50.64c1–6]).
52 Although we cannot find Taehyŏn’s direct explanation on this problem, the concept of amalavijñāna that belongs to not only the conditioned but also the unconditioned may be related to the notion of tathāgatagarbha in East Asian Buddhist tradition. The idea of connecting the conditioned and the unconditioned in the concept of amalavijñāna has much in common with the notion of tathāgatagarbha, thusness that is inherent in sentient beings. For instance, in the Awakening of Faith, one of the representative ‘tathāgatagarbha texts,’ it is stated that tathāgatagarbha (the neither-arising-nor-ceasing [[[mind]]]) is unified with sentient beings’ arising-and-ceasing [[[mind]]] in a neither-identical-nor-different condition, and this is called ālayavijñāna
(Dasheng qixin lun [T 1666.32.576b8–9]: 所謂不生不滅與生滅和合，非一非異, 名為阿梨耶識). From the way in which the tathāgatagarbha (viz., thusness; an unconditioned dharma) and the sentient beings’ mind (a conditioned dharma) are unified, that is, ‘in a neither-identical-nor-different condition,’, we may glimpse a clue to the problem of how to explain the connection of the conditioned and the unconditioned. Taehyŏn composed a commentary on the Awakening of Faith, the Taesŭng kisillon naeŭi yakt’amgi 大乘起信論內義略探記, but there is no extant work that addresses this particular problem from the standpoint of Yogācāra. For detailed discussion on two distinct positions on the concepts of ālayavijñāna and tathāgatagarbha in the Awakening of Faith, see Lee (2019).


References


Abbreviations


CWSL Cheng weishi lun 成唯識論
DGS Dasheng guangbailun shilun 大乘廣百論釋論
Hakki Sŏng yusingnon hakki 成唯識論學記
Shuji Cheng weishi lun shuji 成唯識論述記
Shuyao Cheng weishi lun zhangzhong shuyao 成唯識論掌中樞要
T Takakusu Junjirō 高楠順次郞 and Watanabe Kaikyoku 渡邊海旭, eds. Taishō shinshū daizōkyō 大正新修大藏經 [Revised Buddhist canon compiled during the Taishō reign-period]. 100 vols. Tokyo: Daizōkyōkai, 1924–1935.
X Wanzi xuzang jing 卍字續藏經
ZZL Dasheng zhangzhen lun 大乘掌珍論


Primary sources


Bonmōkyō gekan koshakki jutsu shakushō 梵網經下卷古迹記述迹鈔, Nihon daizōkyō 日本大藏經, 100 vols. Vol. 2. Tōkyō: Kōdansha, 1973–1978.
Cheng weishi lun 成唯識論 (*Vijñaptimātratāsiddhi-śāstra). T 1585.
Cheng weishi lun shuji 成唯識論述記. T 1830.
Cheng weishi lun zhangzhong shuyao 成唯識論掌中樞要. T 1831.
Da Tang xiyuji 大唐西域記. T 2087.
Dasheng fayuan yilin zhang 大乘法苑義林章. T 1861.
Dasheng guangbailun shilun 大乘廣百論釋論 (*Catuḥ-śataka-vṛtti). T 1571.
Dasheng qixin lun 大乘起信論. T 1666.
Dasheng tongxing jing 大乘同性經 (Mahāyānābhisamaya-sūtra). T 673.
Dasheng zhangzhen lun 大乘掌珍論 (*Mahāyāna-hastaratna-śāstra). T 1578.
Dasheng zhuangyanjing lun 大乘莊嚴經論 (Mahāyānasūtrālaṃkāra). T 1604.
Fodijing lun 佛地經論 (*Buddhabhūmisūtra-śāstra). T 1530.
Sŏng yusik non hakki 成唯識論學記. X 818.


Secondary sources


Ch’ae Inhwan. 1983. ‘‘Silla Taehyŏn pŏpsa yŏn’gu (I): haengjŏk kwa chŏjak’’ 新羅大賢法師硏究 (I): 行蹟과著作. Pulgyo hakpo 20: 85–107.
Garfield, Jay L., and Jan Westerhoff, eds. 2015. Madhyamaka and Yogācāra: Allies or Rivals? Oxford; New York: Oxford University Press.
Katsumata Shunkyō 勝又俊敎. 1961. Bukkyō ni okeru shinshiki setsu no kenkyū 仏敎における心識說の研究. Tōkyō: Sankibō busshorin.
Lee, Sumi. 2014. ‘‘Toward a New Paradigm of East Asian Yogācāra Buddhism: Taehyŏn (ca. 8th century CE), a Korean Yogācāra monk, and His Predecessors.’’ PhD diss, UCLA.

——. 2015. ‘‘Redefining the ‘Dharma Characteristics School’ and East Asian Yogācāra Buddhism.’’ The Eastern Buddhist 46(2): 41–60
——. 2016. ‘‘From Criticism to Approval: A Reconsideration of Ji’s Yogācāra Position on Madhyamaka.’’ Asian Philosophy 26(4): 329–353.
——. 2019. ‘‘On the Ālayavijñāna in the Awakening of Faith: Comparing and Contrasting Wŏnhyo and Fazang’s Views on Tathāgatagarbha and Ālayavijñāna.’’ Religions 10(9): 536.
Makransky, John J. 1997. Buddhahood Embodied. New York: SUNY press.
Muller, A. Charles, ed. 2012. Doctrinal Treatises: Selected Works. Collected Works of Korean Buddhism 6. Seoul: Jogye Order of Korean Buddhism.
——. 2015. ‘‘Wŏnhyo’s Approach to Harmonization of the Mahāyāna Doctrines (Hwajaeng).’’ Acta Koreana 18(1): 9–44.
Pang In. 1995. ‘‘T’aehyŏn ŭi Yusik ch’ŏrhak yŏn’gu’’ 太賢의唯識哲學硏究. PhD diss., Seoul National University.
Tillemans, Tom J.F. 1990. Materials for the study of Āryadeva, Dharmapāla, and Candrakīrti. Vienna: Arbeitskreis für tibetische und buddhistische Studien, Universität Wien.
Yamabe, Nobuyoshi. 1997. ‘‘The Idea of Dhātu-vāda.’’ In Pruning the Bodhi Tree: The Storm over Critical Buddhism, edited by Jamie Hubbard and Paul L. Swanson, 193– 204. Honolulu: University of Hawai‘i Press.