Chapter 1: One book, two identities

The Kośa is two books at once, the finest outline of Sarvāstivāda and its most precise critique. The two identities alternate from passage to passage, with no announcement of where the change occurs. This chapter establishes the essay's basic discipline. How can we tell which identity the sentence before us is serving? Only once that rule is in place can the next seven chapters share a premise.

Format and composition. Verse and prose exposition alternate. 頌曰, "The verse states," occurs 527 times, and 論曰, "The exposition states," 522 times. A verse and an explanation form an exceptionally regular format.1 The root verses largely organize Vaibhāṣika doctrine, gathering the contents of two hundred fascicles into a framework of six hundred verses. The verse count for each of the eight chapters is given beneath its title, and the counts total exactly six hundred. The prose explains each verse but repeatedly sets out the author's own views within the explanation. The sequence of the eight chapters has its own design. The chapters on elements and faculties establish the list of dharmas. Those on the world, action, and underlying tendencies explain defiled cause and effect. Those on noble persons, knowledge, and concentration explain pure cause and effect. First the dharmas are listed, then the round of existence is explained, and then its cessation. This sequence is itself Sarvāstivāda. The framework follows the other side entirely. This is the first identity.

The verse at the book's end that states its own position. The book itself admits to the second identity at the end. At the close of fascicle 29, Vasubandhu asks, "This treatise draws together Abhidharma. On what reasoning does it base its explanation of Abhidharma?" The verse answers, "The reasoning discussed in Kashmir is well established. I have largely followed it in explaining Abhidharma. Any slight critical assessment is my fault. Correct judgment of dharmas rests with the Sage." His own prose comment reads, "The reasoning concerning Abhidharma discussed by the Vaibhāṣikas of Kashmir is well established. I have largely followed it in explaining the Abhidharma system. Any slight critical assessment is my fault. Correct judgment of dharmas rests with the Sage."1 The four lines have three layers of meaning. "I have largely followed it" acknowledges that the framework belongs to others. "Any slight critical assessment is my fault" admits that the book contains criticism. His posture is humble, but his pen does not retreat. "Correct judgment of dharmas rests with the Sage" places final judicial authority with the Buddha, not with the evaluators of Kashmir. Later generations summarized this stance in the four characters 理長為宗, "taking the stronger reasoning as one's guiding principle." Neither Puguang's nor Fabao's commentary contains these four characters together. Puguang's Notes on the Kośa has 據理為宗, "taking reasoning as one's guiding principle." Fabao's Commentary on the Kośa has 理勝為宗, "taking superior reasoning as one's guiding principle," and 理長為是, "accepting what has the stronger reasoning." The two characters 理長 first appear in Fabao.2 The summary is apt, but the original verse says something stronger. It amounts to announcing that the school's adjudication is not the final court of appeal.

Three markers for making the distinction. The original text supplies the markers that distinguish the two identities. First is 傳說, "according to tradition." In Xuanzang's translation practice, when Vasubandhu reports Vaibhāṣika doctrine while retaining reservations of his own, these two characters appear. They occur thirty-three times in the work. "It is accepted according to tradition that the account concerns phases," encountered in Chapter 8 of Essay 16, belongs to the same family. Each "according to tradition" is a place where Vasubandhu refuses to sign. Second are attributions such as "The Vaibhāṣikas state as follows." "Vaibhāṣikas" occurs sixty-six times. The reported position has an owner, and responsibility for its words is assigned separately. Third is testimony from the other side. Throughout the Abhidharmanyāyānusāra, Saṃghabhadra calls Vasubandhu Jingzhu (经主) 250 times. The appellation’s meaning remains disputed; the name alone does not establish a sectarian affiliation. The opponent’s specific criticisms offer a further clue to Vasubandhu’s position.3 These markers provide reading clues, but the two identities still need to be distinguished passage by passage. Criticism is densest in the accounts of formations unassociated with mind in the chapter on faculties and in the debate on the three times in the chapter on underlying tendencies. The chapters that follow draw their material from precisely this distribution.

The two Chinese translations. This essay uses Xuanzang's thirty-fascicle translation as its base text, and quotations follow the classical Chinese original. Paramārtha's older Commentary on the Kośa, in twenty-two fascicles, is cited for comparison only where the doctrine differs. It is not a second base text. Reading the two together, their differences are most visible in the verse stating the author's position at the end of the book.4 Differences in format between the translations are explained as they arise, without a separate section.

One Vasubandhu or two? Only the conclusion concerning this scholarly dispute is stated here. The older account identifies the author of the Kośa with the Vasubandhu who turned to Yogācāra in later life. Modern scholarship has raised the question of two Vasubandhus, and the matter remains unsettled. This book makes the smallest commitment. The master who organized the Kośa, was refuted by Saṃghabhadra, and left this verse stating his position is one and the same. His relation to the Yogācāra writings will be addressed in the essay on Yogācāra, as an open scholarly question to be revisited as scholarship develops. In this essay, "Vasubandhu" always means only the author of the Kośa.

On to Chapter 2. The rules are in place. When Vasubandhu criticizes Sarvāstivāda, he draws on another current, called Sautrāntika. We must first see clearly what it holds.

References

References

  1. Abhidharmakośa Śāstra, thirty fascicles, composed by Vasubandhu, translated by Xuanzang, Taishō vol. 29 (T1558). The counts of 頌曰 ("The verse states"), 527; 論曰 ("The exposition states"), 522; 傳說 ("according to tradition"), 33; and 毘婆沙師 (Vaibhāṣikas), 66, all come from a complete search of the classical Chinese text in the CBETA edition. The passage stating the author's position at the end reads: "This treatise draws together Abhidharma. On what reasoning does it base its explanation of Abhidharma? The verse states: The reasoning discussed in Kashmir is well established. I have largely followed it in explaining Abhidharma. Any slight critical assessment is my fault. Correct judgment of dharmas rests with the Sage. The exposition states: The reasoning concerning Abhidharma discussed by the Vaibhāṣikas of Kashmir is well established. I have largely followed it in explaining the Abhidharma system. Any slight critical assessment is my fault. Correct judgment of dharmas rests with the Sage." The exact number of verses follows the counts beneath the chapter titles in Xuanzang's separate one-fascicle translation of the Root Verses of the Abhidharmakośa (T1560): elements, forty-four; faculties, seventy-four; world, ninety-nine; action, 131; underlying tendencies, sixty-nine; noble persons, eighty-three; knowledge, sixty-one; concentration, thirty-nine. The total is six hundred. Counting line by line gives the same result. The eight chapters contain 2,404 lines of verse. All verses have four lines except two in the chapter on the world, which have six each. Fascicle 1 of Puguang's Notes on the Kośa says that Vasubandhu "composed six hundred verses of the Kośa," while the prose exposition amounted to "eight thousand verses." Fabao's Commentary on the Kośa, fascicle 1, gives a finer count: "This treatise has 607 verses in all, broadly divided into three parts. The first three are the introduction to the exposition of the verses. The next six hundred are the root verses and constitute the main teaching. The final four verses and the chapter refuting self form the concluding section of the exposition of the verses." Thus the opening three and closing four are counted separately from the six hundred. The biographical tradition associated with Paramārtha's translation also gives six hundred. The Biography of Master Vasubandhu records that he "in this sequence composed more than six hundred verses to gather the meaning of the Vibhāṣā." The modern Chinese translation appears in this site's Kośa section, translated from the CBETA edition. ↩↩

  2. Puguang, Notes on the Kośa, thirty fascicles, Taishō vol. 41 (T1821); Fabao, Commentary on the Kośa, thirty fascicles, in the same volume (T1822). Complete searches of the two commentaries give zero occurrences of the four characters 理長為宗 together. Puguang has "as one's guiding principle" but not "stronger reasoning." In discussing dependent origination in fascicle 1, he writes, "The master takes reasoning as his guiding principle, not partisan attachment." In fascicles 8 and 10 he writes, "The author takes reasoning as his guiding principle, not the evaluators of the Vibhāṣā as his authority." There are three occurrences in all. Fabao has both halves. Fascicle 1 reads, "In the prose exposition he takes only superior reasoning as his guiding principle, without favoring one school, though within it he more often takes Sautrāntika as the correct position." It also reads, "Thus we know that here he accepts what has the stronger reasoning, rather than adhering to one fixed school." Fascicle 10 reads, "The author does not hold rigidly to one school; wherever the reasoning is stronger, he adopts it." Puguang studied under Xuanzang, and his notes came first; Fabao's commentary came later. Yet the two characters 理長, "stronger reasoning," first appear in Fabao. The fixed four-character formula does not come from either commentary. It has another source, which this book has not yet established. ↩

  3. Abhidharmanyāyānusāra Śāstra, eighty fascicles, composed by Saṃghabhadra, translated by Xuanzang, Taishō vol. 29 (T1562). The designation 經主, Jingzhu, occurs 250 times across the work, according to a complete search of the classical Chinese text in the CBETA edition. For the disputed appellation, see Collett Cox, Disputed Dharmas: Early Buddhist Theories on Existence (Tokyo: International Institute for Buddhist Studies, 1995), p. 56 and p. 61, notes 15–17; and Robert Kritzer, Vasubandhu and the Yogācārabhūmi: Yogācāra Elements in the Abhidharmakośabhāṣya (Tokyo: International Institute for Buddhist Studies, 2005), pp. xxxi–xxxii. Sūtrakāra, understood as the author of the root verses among other interpretations, is an evidenced but disputed reading, not a settled basis for assigning sectarian affiliation. Kritzer also cautions that positions criticized by Saṃghabhadra need not all be Sautrāntika or Yogācāra positions. The figure 250 counts the string 經主; it does not represent a fresh semantic examination of every occurrence. ↩

  4. Abhidharmakośa-bhāṣya, twenty-two fascicles, composed by Vasubandhu, translated by Paramārtha, Taishō vol. 29 (T1559). The fascicle count follows Huikai's own statement in the opening preface: "The treatise has twenty-two fascicles, the verses one fascicle, and the explanatory commentary fifty-three fascicles." The one-fascicle verse text has not survived, so the older translation has no independent verse text to count. Translation began in the first month of Tianjia 4 under the Chen dynasty, and retranslation and revision were completed in the twelfth month of Guangda 1. Of the passages cited in this essay, the denial of acquisition's real existence appears in fascicle 3 of the older translation, the denial of the unconditioned's real existence in fascicle 5, and the opening of the refutation of self in fascicle 22. Their doctrine agrees with Xuanzang's translation, though fascicle numbers and terminology differ. The substantive difference lies in the verse stating the author's position at the end. In the older translation it appears at the close of fascicle 21: "The Vaibhāṣika reasoning of Kashmir is well established. I have largely followed it in setting out this treatise. Partisan attachment concerning the true Dharma is my fault. In judging dharmas correctly, the Buddha is the authority." The third line speaks of partisan attachment, where Xuanzang speaks of critical assessment. The older prose explanation says, "Only the Buddha, the Blessed One, is the highest authority ... If noble disciples of the Buddha judge the true Dharma apart from the Āgamas and reasoning, they are not even a middling authority." Xuanzang has, "Correct judgment of dharmas rests only with the Blessed One and the great noble disciples of the tathāgatas." The two translations are opposed on whether noble disciples have authority. The older one restricts judicial authority more tightly. ↩