Introduction

K.R. Norman

I. The Dhammapada Genre

§1. The Pāli Dhammapada is perhaps the best known of all Buddhist texts. We know, however, of other examples of this type of collection or anthology, which we might describe as the Dharmapada genre of literature, belonging to other schools of Buddhism, and it is likely that a text of this type was included in the canons of all the Hīnayāna schools of Buddhism.

§2. Winternitz suggested that, since more than a half of all the verses of the Dhammapada have also been traced in other texts of the Pāli canon, there was scarcely any doubt that, generally speaking, the compiler of the anthology took them from the setting in which we still find them today. He did, however, note that the collection has come to include some sayings which were originally not Buddhist at all, but rather drawn from that inexhaustible source of Indian gnomic wisdom, from which they also found their way into Manu’s law book, into the Mahābhārata, the texts of the Jains, and into narrative works such as the Pañcatantra. Since Winternitz wrote, investigations have suggested that in many cases the Dhammapada did not borrow from elsewhere in the canon, but that in the canon as a whole borrowing took place from a store of verses which in all probability pre-dates the canon in its present form. Although we talk about Dharmapada literature, the Pāli parallels of verses in the other Dharmapada literature are sometimes to be found, not in the Dhammapada, but in the Suttanipāta, or the Saṃyutta-nikāya, or the Jātaka, or occasionally in other canonical texts.

§3. We can be certain that all versions we have of the Dharmapada/Dhammapada are translations from earlier versions, all going back ultimately to the basic store of Dharmapada verses, which Brough called “a body of floating verses”. Even if we could date the versions which we have, we should be dating only the translation from an earlier version. Many of the verses in the store had no specific Buddhist flavour, which explains why parallel versions are found in both Jain and Brahmanical texts. The Buddhists began to add to and draw upon this store at the time of the Buddha, although it is very likely that additions from other sources were also made to the corpus after that time. If we look at any one of this group of texts we will find that each one of them has some features which might reasonably be surmised to be old, and yet at the same time each one has features which are manifestly incorrect or late.

§4. The relationship between the various Buddhist Dharmapada texts is very complicated, with patterns of equivalence between them varying from verse to verse, and sometimes even from pāda to pāda. The fact that any two or more of them agree in some feature tells us only that in some way, in the history of the texts, they were dependent upon a common source for that particular feature. The number of verses each redactor selected, the number of vargas into which they were sorted and the way in which the verses were apportioned to each varga, give us no information whatsoever about the date at which each selection was made.

II. Parallel Versions

§5. Beside the Theravādin version in Pāli (OvH/KRN 1994), we have a version in the Gāndhārī Prakrit (Brough 1962), which has on good grounds been assigned to the Dharmaguptakas. There is a version in a highly Sanskritised Prakrit, the so-called Patna Dharmapda (Shukla 1979, Roth 1980, Cone 1989), which may belong to the Mahāsāṅghikas, and there are several closely related versions of the Udāna-varga in Buddhist Hybrid Sanskrit, probably belonging to the Sarvāstivādins or Mūlasarvāstivādins. Bernhard (1965) sometimes prints two versions side by side when they are irreconcilable. If we compare Nakatani’s edition (1987) with Bernhard’s we sometimes find evidence of yet another tradition. Brough noted several places where the Tibetan translation of the Udāna-varga seems to follow yet another tradition.

§6. There are also portions of Dharmapada material in the Mahāvastu, which belongs to the Lokottaravādin Mahāsāṅghikas. The Sahasra-varga is quoted by name (bhagavāṃ Dharmapadeṣu Sahasravargaṃ bhāṣati) at the beginning of the verses at Mvu III 434-6, and is probably complete. The verses at Mvu III 421-3 are not specifically ascribed to the Dharmapada but from their content they are almost certainly a Bhikṣu-varga, which is possibly not complete.

III. The Arrangement of the Verses

§7. In all the versions of the Dharmapada genre, we find the same pattern of named vargas consisting, for the most part, of verses appropriate to the title of each varga. Although they clearly go back to a common body of verses, each text contains verses not found in the other versions. These may be taken from the common fund, and their absence may be due simply to editorial choice, or they may be imported from a source known only to the individual editor or his successor, or they may be the invention of the editor or of someone later in the train of transmission. It has been suggested, for example, that 109 was inserted in its position merely to provide an introduction to 110-5. It is found in that position only in Dhammapada.

§8. Despite the resemblances, it is easy to show that not one of these anthologies is directly dependent upon any other. Some verses in the Patna Dharmapada are so similar in form to verses in the Dhammapada that it is tempting to think that the former is simply a more Sanskritised version of the latter, but the nature of the language, and the number of the vargas and the distribution of the verses in the vargas shows clearly that this is not so. When we compare the Dharmapadas belonging to the various traditions, we find that no two versions agree for more than a few verses, at most, at a time, although clearly all versions had access to the same basic store of material.

§9. There is no total agreement about varga names. There appears to be no consistency in the ordering of the vargas, or of the verses in the vargas. Brough suggested that there was some significance in the fact that the Brāhmaṇa-varga is the first in the Gāndhārī Dharmapada, but the last (of 26 vaggas) in the Dhammapada and (of 33) in the Udāna-varga. He made his suggestion before the publication of the Patna Dharmapada, where it is third (of 22), which would seem to indicate that the section had no particular significance for the composer of that collection.

§10. There are 423 verses in the Dhammapada, arranged in 26 vaggas, according to their subject matter, although a verse may sometimes seem appropriate to more than one varga, so that it appears twice. So 80 appears in Paṇḍita-vagga, but reappears as 145 in the Daṇḍa-vagga. In the Puppha-vagga (44-59) each verse contains either the word puppha or the name of a specific flower, except for 50 and 57, as Rau points out. Sometimes it is possible to see why such anomalous insertions have occured. We can surmise that 57 was inserted because sampannasīlānaṃ seemed appropriate in view of sīlavataṃ gandho in 56, in just the same way as 58 and 59 make a pair, with yathā saṅkāradhānasmiṃ in 58 parallel to evaṃ saṅkārabhūtesu in 59.

§11. Similarly, every verse in the Brāhmaṇa-vagga (383-423) contains the word brāhmaṇa (very frequently in the refrain tam ahaṃ brūmi brāhmaṇaṃ), except for 394. This undoubtedly was inserted after 393 because, like 393, it contained the word jaṭāhi and in its rejection of non-Buddhist ascetic practices it seemed to echo the condemnation of the idea that a brāhmaṇa was such because of the caste he was born into.

§12. In the same way, in the Nāga-vagga (320-33), we find not necessarily the word nāga, but nevertheless some other word for “elephant” in every verse, except 321 323 325 328 331-3. We can see that sometimes a verse is included because it makes a pair with another verse which is appropriate, e.g. 321 and 323 are included because of the word danta which they have in common with 322. There seems no reason for 325, as Rau points out, except the fact that it is concerned with a large animal. It is interesting to note the relationship between 328-33, and the train of associations which explains to some extent why they are in the Nāga-vagga. There is no mention of nāga in 328, but it forms a pair with 329, where nāga is mentioned. Pāda d of 330 matches pāda d of 329. In 328-30 there are references to sahāya(tā), which leads on to 331, which includes sahāya, but not nāga. It also includes a reference to sukha, which leads to 332 and 333, neither of which refer to nāga.

IV. The Commentary

§13. There is a commentary, which is said to have been written at the request of a thera named Kumārakassapa, and to be based upon commentarial material in the language of the island. At the end of the work it is said that it was composed by Buddhaghosa while residing in a residence built by king Sirikūṭa. Doubt has been cast upon this ascription because the style of the commentary is so different from Buddhaghosa’a other cties, but it is possible that Buddhaghosa collected together a number of traditional stories and his role was rather that of an editor.

§14. Each verse, or group of verses, is ascribed to the Buddha, and the commentary includes a narrative portion explaining the circumstances in which the verses were uttered. The information given in the story is frequently helpful in understanding the verses. Sometimes the story is based upon an event which is mentioned elsewhere in the canon, e.g. the story which accompanies 266, although some additional details are given. The source of many of the stories is unknown, and we cannot tell if they represent the genuine background to the utterance, or whether they have been manufactured at a later date to suit the circumstances. All the stories end with the Buddha uttering the verse or verses and, since the authorship of those verses which are also found in non-Buddhist sources must be uncertain, the circumstances narrated in the stories must also be uncertain. As in the case of the Jātaka stories, it would appear that occasionally a verse has been married to a story because of a supposed connection which may not be there. In other cases the verse is of such general application that it could have been uttered in any circumstances.

§15. There is also a grammatical section explaining the meaning of the words of the verse(s). I quote extensively from this portion of the commentary, to explain why I translate in the way I do, although it is clear that the commentary or the tradition it was following did not understand the construction of some verses, e.g. 42, or some grammatical forms, e.g. pahātave 34, or the meaning of some words, e.g. loka-vaḍḍhana 167, mattā 290, veyyaggha 295, paribbāja 313.

§16. Some readers of my earlier translations have regretted the fact that I have also quoted copiously from the commentaries there, without giving any translation. The need to do so in the case of the commentarial portions of the Dhammapada-aṭṭhakathā has now been met in an exemplary manner by C&P, whose work complements the translation of the stories by Burlingame.

V. The Translation

§17. This translation is based upon OvH/KRN, except where otherwise indicated, even where a comparison with other versions of the text in Prakrit and Sanskrit suggests that the Pāli tradition has misunderstood or changed the text in some places, e.g. 26d 369b. In such cases I have discussed the alternative versions in the notes, and have tried to explain why the Pāli tradition shows variations. Very occasionally, however, where I think the Pāli tradition is incorrect, I have translated what I think is the correct reading, e.g. 259d. I have discussed such points in the notes.

§18. The notes and the frequent quotations from the commentary they contain are intended to be a guide to the way in which I arrived at my translation. Where I differ from the commentary’s explanation, I give my reasons for doing so. I have made frequent references to parallel versions where those texts help to throw light upon the meaning of Dhammapada. I have not tried to give a reconstruction of the “original” form of the Dhammapada—fascinating though that would be.

VI. The Title of the Translation

§19. I think it is important, in order to avoid confusing potential readers, to have different titles for a text and for its translation. The word dhammapada, like Skt dharmapada (see BHSD s.v. dharmapada) is used in two ways: as the title of a text and in the sense of “religious doctrine”. I assume that the title has the same meaning as the occurrences of the word in the text itself, dhammapadaṃ sudesitaṃ 44-5 and ekaṃ dhammapadaṃ seyyo 102. Kalupahana entitles his version “A Path of Righteousness”, doubtless relying on 44-5 where he translates dhammapadaṃ sudesitaṃ as “the well-taught path of righteousness”, presumably taking pada in the sense of patha, although in the notes he includes a reference to “the well-taught verses of the doctrine”, and he translate ekaṃ dhammapadaṃ in 102 as “one statement of the doctrine”.

§20. If we accepted the meaning “path” for pada, then we could translate “the path of the doctrine”. If we prefer “place” for pada, then we can translate “the place of the doctrine”. I prefer the equation with 44-5, and follow the translation I give there and in 102. I have therefore entitled the translation “The Word of the Doctrine”.

VII. The Metres of Dhammapada

(omitted)

VIII. Metrical Licence

(omitted)

IX. Orthography and Phonology

(omitted)