Variant Readings in the Subhasutta of the
Dīghanikāya:
Based on Palm-leaf Manuscripts from Five Traditions
C
HAOWARITHREONGLITHBunchird
Currently several printed editions of the Pāli canon are available. The four major editions are European (Ee̶PTS), Burmese (Be̶Chaṭṭhasaṅgīti), Ceylon/Sinhalese (Ce̶Buddha- jayanti), and Siam/Thai editions (Se̶Syāmaraṭṭha). Among them, Ee is the most familiar edition to scholars worldwide.
However, important Pāli texts of Ee were edited long ago when the number of available manuscripts and reference literature of Pāli language were limited. Ee mainly uses Sinha-lese and Burmese manuscripts. They only occasionally referred to readings of Khom man-uscripts, and almost never to Tham and Mon manuscripts. Furthermore, there is a request made by scholars in this field that we need a new critical edition.1)
The Dhammachai Tipitaka Project are now working on producing a new critical edition based on palm-leaf manuscripts of five traditions: Burmese (B)/Mon (M), Sinhalese (C), Khom (K), and Tham (T). In this paper, I will show a nature of various readings and issues we found in the course of editing the text, focusing on the Subhasutta.
1. The Subhasutta
The Subhasutta is the 10th sutta in the Sīlakkhandha-vagga, the first volume of the Dīghanikāya. The story of this sutta is set in the time shortly after the death of the Buddha. Ānanda is the main character giving an explanation about the right conduct (sīla), medita-tion (samādhi), and wisdom (paññā) to the Brahman Subha, who invites him to his place. Several passages have parallels in the Sāmaññaphala-sutta.
2. Materials
manu-scripts of the Sīlakkhandha-vagga2) available to us, five representative manuscripts from each script tradition have been selected for editing where possible.3) Below is a list of the representative manuscripts with their age:
B1̶1679; B2̶1768; B3̶1774; B4̶1792; B5̶1806 C1̶1744; C2̶1783; C3̶1832; C4̶1855; C5̶unknown K1̶1777; K2̶1783–1809; K3̶1824–1851; K4̶1851–1868; K5̶unknown T1̶1598; T2̶1822; T3̶1836; T4̶1836; T5̶unknown M̶unknown
3. Nature of Variant Readings in the Subhasutta
According to a comparative study of our manuscripts, we found the manuscripts agreed for most of the readings. Only 200 problematic words or around 6% were found out of approx-imately 3,200 words of the entire sutta. This implies that the text of the Pāli canon is firmly established and has been well transmitted, and it would not be an impossible task to trace back with the help of all manuscript traditions to the original recension of the Mahāvihāra school.
When encountering problematic words, we will have to determine an original reading from its variant readings by considering the following factors: context, witnesses (includ-ing commentaries), philological analysis, and others such as cultural background etc. Some of the main philological points which emerged in the course of our comparative study are as follows:
3.1. A Majority Reading from Manuscripts Could Be Wrong
Example 1: the Brahman Subha orders his fellow to invite Ānanda to his place.
atha kho Subho māṇavo todeyyaputto aññataraṃ māṇavakaṃ āmantesi: ehi tvaṃ, māṇavaka, yena samaṇo Ānando ten upasaṅkama, upasaṅkamitvā mama vacanena samaṇaṃ Ānandaṃ appābādhaṃ appātaṅkaṃ lahuṭṭhānaṃ balaṃ phāsuvihāraṃ puccha (DN (Be) I 187; (Ee) I 204) ・upasaṅkama (B2.4 K1 M Be Ce Se); upasaṃkama (C1 Ee)
・upasaṅkami (B3.5 K2.3.4.5 T1.2.3.4.5
); upasaṃkami (B1 C2.3.4.5)
The Brahman Subha here is giving an order to his fellow. It is clear from the context that the imperative form is expected to be used in this passage e.g., ehi, upasaṅkama, and puccha. However, the majority of manuscripts read the aorist tense upasaṅkami for upasaṅkama. This is an obvious mistake made by the scribes caused by the general pattern
upasaṅkami, upasaṅkamitvā which occurs thirty-five times whereas the pattern upasaṅkama, upasaṅkamitvā only six times in the Sīlakkhandha-vagga. This sort of scribal error is not uncommon, but it is not difficult to identify a problem and determine a correct reading.
3.2. Orthographic Confusion in Traditions
Example 2: a simile of an effect of the first jhāna. A monk s body suffused with rapture and happiness is like a ball of bath-powder with moisture.
seyyathāpi, māṇava, dakkho nhāpako vā nhāpakantevāsī vā kaṃsathāle nhānīyacuṇṇāni ākiritvā udakena paripphosakaṃ paripphosakaṃ sandeyya. sāyaṃ nhānīyapiṇḍi snehānugatā snehaparetā santarabāhirā phuṭā snehena, na ca paggharaṇī. (DN (Be) I 201; [(Ee) I 74])
(Brahman, supposing that a skilled bath attendant or his apprentice pour bath-powder into a bronze dish, sprinkle it with water, and knead it into a ball. At evening, the ball of bath-powder is pervaded by moisture, overwhelmed by moisture, suffused with moisture inside and out, yet would not trickle.)
・sā ssa (C1.2.3.5 [Ee]); sā (B1 [T1.2.3]) ・tassa (B3.4.5)
・sāya(ṃ) (K1.2.3.4 M T5 Be Ce Se); sāha (K5) ・omit (B2 C4 T4)
Note: a source in a square bracket is a reading from a parallel passage in the Sāmaññaphala-sutta.
All Asian printed editions (Be Ce Se), and Khom manuscripts, read sāyaṃ at evening, or a better interpretation sā yaṃ that very (ball). But we prefer the Sinhalese reading
sā ssa which would give smoother translation that (ball of bath-powder) would. This
reading have also a good support from the oldest manuscripts of the Burmese and Tham traditions (B1 and T1). Furthermore, the similes in the Subhasutta shows a tendency to use a finite verb of as in the optative form, assa.4) It is also easy to explain the variation be-tween sā ssa and sāyaṃ / sā yaṃ which is caused by the similarity of the form of the let-ters -ssa and -ya in SEA scripts. The orthographic confusion in traditions can to be found from time to time when working with the manuscript readings.
3.3. Sinhalese vs SEA Readings
The variation between Sinhalese and SEA traditions is one of the most typical variation of manuscript readings.
avocumhā kho mayaṃ bhoto vacanena taṃ bhavantaṃ Ānandaṃ. (DN (Be) I 188; (Ee) I 205) ・bhavantaṃ (B1.2.3.5 K1.2.3.4.5 M T1.2.3.4.5 Be Ce Ee Se)
・bhagavantaṃ (B4 C1.2.3.4.5)
We have here the variation with two groups of the readings̶Sinhalese and SEA traditions. It is obvious that the SEA reading bhavantaṃ would be the appropriate readings here, as in general the word bhagavantaṃ must refer to the Buddha, and cannot be used for Ānanda.
In fact, unlike the above example, in many cases, it is rather difficult to determine an ap-propriate reading between those two traditions. Next is a more difficult example.
Example 4: a simile of an effect of the second jhāna. A monk s body suffused with rapture and happiness is like a deep lake filled by a current of cool water within the lake.
seyyathāpi, māṇava, udakarahado gambhīro ubbhidodako. tassa nevassa puratthimāya disāya udakassa āyamukhaṃ, na dakkhiṇāya disāya udakassa āyamukhaṃ, na pacchimāya disāya udakassa āyamukhaṃ, na uttarāya disāya udakassa āyamukhaṃ, ... (DN (Be) I 195; [(Ee) I 74]) ・puratthimāya, dakkhiṇāya, pacchimāya, uttarāya (B[1].2.3.4.5 K1.2.3.4.5 M T1.2.3.4.[5] Be Ce Se) ・puratthimāya, pacchimāya, uttarāya, dakkhiṇāya (C1.2.3.4.5 [Ee])
The problem here is about the order of the directions. The Sinhalese tradition reads east, west, north, and south, whereas the SEA tradition reads east, south, west, and north, which is a counterclockwise direction. The commentary does not provide any hint to solve the problem. The only way left to us is to find other occurrences of the order of the direc-tions, and look for a clue from them. Fortunately, apart from the above passage, seven other occurrences of the order of the directions were found in the Sīlakkhandha-vagga.5) Not only SEA manuscripts, but Sinhalese manuscripts read east, south, west, and north in all occur-rences as well. Except, only in two occuroccur-rences being mentioned in a simile, the Sinhalese manuscripts still read east, west, north, and south.6) According to Wessels-Mevissen [2001], a typical order of the directions in ancient India seems to be east, south, west, and north, especially in terms of gods of the directions. In sum, in the above passage, the SEA reading should be preferable when considering the Indian cultural background.
3.4. The Commentaries Are a Great Help
The most useful part of the commentaries for the work of textual editing is the canonical citations provided by Buddhaghosa or the commentators. The complete agreement of the citations from different traditions of the commentaries would give us an opportunity as if we were using the same manuscripts with Buddhaghosa which were much older than any
extant manuscripts available to us.7)
In some particular cases, Buddhaghosa s silence could be probative, I would call it a neg-ative citation. In the Subhasutta, a list of twenty-six low arts (tiracchāna-vijja) are listed.8) Buddhaghosa cites and gives his explanation to every single low arts, except the third one.9) Besides, an extensive variation is found here: the SEA manuscripts read bhūtikamma/bhūta-°/puri-°/bhūmi-° whereas all Sinhalese manuscripts omit the reading. Considering the negative citation of the commentary and the extensive variation of the reading, the omission in the Sinhalese tradition is preferable, and the SEA readings must be an interpolation.10)
The commentarial citation and explanation can be useful in more difficult cases.11) In the simile of the divine eye found in the Subhasutta which Ee omits and refers back to the Sāmaññaphala-sutta, I found the inconsistency of the canonical reading vīthiṃ sañcarante (DN (Ee) I 83.6), against the citation in the commentary vītisañcarante (Sv (Ee) I 224.1). The citation here gives an interesting reading. A further study of the commentary is needed to solve this particular problem.
Notes
1) See Norman (1997: 2). For an overview of the need of re-editing the PTS edition, see Balbir (2009).
2) Twenty-six Sinhalese, thirty-two Burmese, fifteen Khom, nine Tham, and one Mon script manu-scripts.
3) For the method of selecting manuscripts, see Wynne (2013: 141–142), Somaratne (2015: 217–226). 4) DN (Ee) I 71.33, 72.5, 72.16, 72.23, 74.21, 75.19, etc.
5) One occurrence is found in the Kūṭadanta-sutta (DN (Ee) I 142.16ff.) mentioning the directions in which the donation took place; three in the Mahāli-sutta (DN (Ee) 153.1ff., 154.1ff., 154.36ff.) mention-ing the directions where the divine body and voice came from; two in the Poṭṭhapāda-sutta (DN (Ee) I 194.18ff., 198.21ff.) mentioning a simile of the whereabouts of a castle, and one in the Kevaṭṭa-sutta (DN (Ee) I 22.23ff.) mentioning a simile of the directions in which the bird is flying.
6) In fact, the order of east, west, north, and south also exists in the SEA tradition, but in a different context. It is used when measuring a huge object such as a city, castle, etc. See DN (Ee) II 170.2ff. This might be a cause of the confusion in the Sinhalese tradition.
7) See Somaratne (2015: 213).
8) DN (Be) I 191, (Ee) I 206 which refers a parallel passage in the Sāmaññaphala-sutta (DN I 69.21ff.). 9) Sv (Be) I 91ff., (Ee) I 97ff.
10) Wynne (2013: 159–160) has already pointed out the same issue in the different list of low arts. 11) For more uses of the commentarial source in textual editing, see Wynne (2013: 159–164).
Abbreviations DN Dīghanikāya
Sv Sumaṅgalavilāsinī̶the commentary on the Dīghanikāya Bibliography
Balbir, Nalini. 2009. Thoughts about European Editions of Pāli Texts. Thai International Journal for Buddhist Studies 1: 1–19.
Norman, K. R. 1997. A Philological Approach to Buddhism. The Buddhist Forum, vol. 5. London: School of Oriental and African Studies.
Somaratne, G. A. 2015. Middle Way Eclecticism: The Text-Critical Method of the Dhammachai Tipiṭaka Project. Journal of Buddhist Studies 12: 207–239.
Wessels-Mevissen, C. 2001. The Gods of the Directions in Ancient India: Origin and Early Development in Art and Literature (until c. 1000 A.D.). Berlin: Dietrich Reimer Verlag.
Wynne, Alexander. 2013. A Preliminary Report on the Critical Edition of the Pāli Canon Being Pre-pared at Wat Phra Dhammakāya. Thai International Journal for Buddhist Studies 4: 135–170. Key words Pāli canon, Critical edition, Palm-leaf manuscript, Subhasutta, Dīghanikāya
(Researcher, Dhammachai Tipitaka Project, Thailand)
An Early Text on the History of Rwa sgreng Monastery
The Rgyal ba’i dben gnas rwa sgreng gi bshad pa nyi ma’i ’od zer
of ’Brom Shes rab me lce
Harvard Oriental Series 82
New Publication
Maho Iuchi
7×10 inches, x, 210pp., $40 Harvard University Press, 2016
