Madhyadesika Madhyoddesika And Madhyuddesika

Author(s): J W De Jong
Publisher: J W De Jong
Catalog link: https://jainqq.org/explore/269590/1

Page #1

Madhyadesika, Madhyoddesika and Madhy'uddesika
By J. W. DE JONG*
The Mahāvastu (éd. É. Senart, tome premier, Paris, 1882) begins with a "series of nidānanamaskara-s (pp. 1-2. 12) which Edgerton in his dictionary translated by 'introductory salutations'. This is followed by the following statement: aryamahāsāmghikānām lokottaravādinām madhyadeśikānām pāthena vinayapitakasya mahavastuye ādi. In his introduction Senart rendered this as follows (p. XXI): "Commencement du Mahāvastu, partie du Vinayapitaka, de la recension de la branche des Mahâsâmghikas, dite les Lokottaravadins du Madhyadeça.” J. J. Jones translates: "Here begins the Mahāvastu, which is based on the redaction of the Vinaya Pitaka made by the noble Mahāsānghikās, the Lokottaravādins of the Middle Country" (The Mahāvastu, vol. I, London, 1949, pp. 2–3). In a note Jones remarks that the Middle Country is that part, variously delimited, of central India, which was the birthplace of Buddhism.
In the Prasannapadā (éd. Louis de La Vallée Poussin, Bibliotheca Buddhica IV, St.-Petersburg, 1903-1913) one finds the following reference to the Mahāvastu:
madhyoddeśikās ca mahāvastūpadistabhūmivyavasthayā prathamabhūmisthitam bodhisattvam utpannadarśanamārgam vyācakṣānāḥ samghāntaḥpātinam vyācakşate (p. 489. 1-2).
According to a note the manuscripts have madhyāddesikās. De La Vallée Poussin quotes the Tibetan translation: bar-ma 'don-par byed-pa dag. which he reconstructs into Sanskrit as madhya-pāthaka. Jacques May translates: "Enfin les Madhyadesika, d'après le système des terres enseigné dans le Mahāvastu, affirment que le Bodhisattva de la première terre a produit le chemin de la vision [et] fait partie de la communauté" (Candrakīrti Prasannapadā madhyamakavrtti. Douze chapitres traduits du sanscrit et du tibétain, Paris, 1959, p. 220). Apropos of the reading Madhyadeśikās, May remarks: "Cette leçon paraît préférable à Madhyoddeśikās. Le terme ne semble pas désigner une secte déterminée, mais seulement une localisation géographique. Le tib. a madhyapathaka." In my review of May's work I pointed out that the reading Madhyoddeśika corresponds exactly to Tibetan bar-ma 'don-par byed-pa (IIJ 5, 1961-1962, p. 164). The meaning 'to recite' for uddisati is quite common in Pāli, cf. Critical Pāli Dictionary,
*) This article was first published in the volume in memory of Professor Tucci.

Page #2

Madhyadesika, Madhyoddeśika and Madhy'uddesika
139
vol. II, fasc. 9 (Copenhagen, 1975), s.v. uddisati 4.c. esp. to “point out”, i.e. to recite, the pātimokkha (and other texts); uddesaka I. expounder, reciter (of the pātimokkha). The reading madhyoddesikās is confirmed by manuscript R, cf. IIJ 20 (1978), p. 242.
In 1970 Gustav Roth published the Bhikṣuṇī-vinaya of the Ārya-Mahāsāmghika-Lokottaravādin (Patna, 1970). The text begins as follows: om namo Buddhāya / ārya-Mahāsāṁghikānām Lokottaravādinām madhy'-uddesikānām pāthena bhikṣuni-vinayasyadih (p. 1). In his note Roth quotes the colophons of two other vinaya texts of the same school, the Abhisamācārikā Dharmāḥ and the Bhikṣu Prātimokşa-sūtra: Abhisamācārikāḥ samāptāh / arya-Mahāsāmghikānām Lokottara-vādinām madhy'-uddeśa-pāțhakanām pāțheneti; samāptam prātimokşa-sūtram ārya-Mahāsāmghikānām Lokottara-vādinām madhyoddesikānām pātheneti. B. Jinananda, the editor of the first text (Abhisamācārikā, Patna, 1969) changed the reading of the manuscript to madhyadeśapāthakānām (cf. pp. XXIV and 230). W. Pachow and Ramakanta Mishra, the editors of the Prātimokşa-sūtra (The Prātimoksa-sūtra of the Mahāsānghikas, Allahabad, 1956), read mādhyād-desi
kānām, but according to Roth the reading given by him can be established . with certainty. In his recent edition of the Prātimokşasūtra, Nathmal Ta
tia also reads madhyoddesikānām (Prātimokșasūtram, Patna, 1975, p. 38. 1-2).
Apart from the Mahāvastu all the other texts have either madhyoddesika or madhy'uddesika. In his critical apparatus Senart does not indicate any variant for the reading madhyadesikānām. Senart had at his disposal six manuscripts which are probably all copies of one Nepalese manuscript (cf. Introduction, p. IX). An older manuscript (dated Nepāla samvat 777) is preserved in Nepal (cf. Roth, Introduction, p. XV) and it would be interesting to see whether it also has the reading madhyadesikānām. However, it seems likely that the original reading was madhyoddesikānām, which could easily have been corrupted to madhyāddesikāņām. Another copyist probably considered it necessary to correct this reading to madhyadesikānām.
In his History of Buddhism (Chos-'byun), Bu-ston (1290–1364 A.D.) gives an account of the Third Council in which he quotes the opinions of several authorities concerning the languages used by different Buddhist schools, cf. Obermiller's translation: History of Buddhism. II. Part (Heidelberg, 1932), pp. 96-101. Recently Yuyama has edited and translated the text of the relevant sections ('Bu-ston on the Languages Used by Indian Buddhists at the Schismatic Period', in: Die Sprache der ältesten buddhistischen Überlieferung, Göttingen, 1980, pp. 175-181). Yuyama remarks: “It admits of no doubt that Obermiller's English translation has served as the first aid for those who were unable to consult the original Tibetan text because of its inavailability ... In some cases, however, his translation is rather free and misleading, unless the original is referred to." However, Yuyama's translation is not an improvement and does not supersede Obermiller's translation.

Page #3

J. W. de Jong
Bu-ston quotes the following passage from the Prabhāvatī: de-nas rgyal-po dharma aśva-ka si-ba dan / dgra-bcom-pa-rnams-kyis tha-mal-pa dan zur-chag-pa dań bar-mar 'don-pa'i tshig-la mnon-par źen-pa gdul-pa'i dban-gi phyir rim-gyis gźun gźan dan gźan-du sbyar-te rgya-chen-po'i skad-du sbyarba'i mdo-sde-la sogs-pa Ita-bu-ste / Obermiller translates: "Thereafter the king Dharmāçoka died, and the Arhats, in order to put an end to the prac tice of reciting (Scripture) in Prakrit, Apabhramça and in a dialect of intermediate character, gradually rehearsed (the kanonical texts) according to other methods. These new texts were like the sutras which were compiled in Sanskrit." Yuyama translates: "Then King Dharma Aśoka died, and the Arhats, in order to subdue adherence to Prakrit (tha-mal-pa), Apabhramsa (zur-chag-pa) and the language of intermediate speech (bar-mar 'don-pa'i tshig), have gradually compiled the Scripture severally, (which have become) the sütras and the like, composed in the language of great extensiveness (rgya-chen-po'i skad)." The text of the blockprint of the Chos-'byun has rgya-chen-po'i skad. Obermiller notes that the Tanjur has rgyal-po'i skad which he renders as Sanskrit. According to Yuyama rgyalpo'i skad-du seems to be simply a misprint. However, he does not explain why the reading rgya-chen-po'i is preferable. The meaning of rgya-chenpo'i skad is not known. Yuyama states that "it seems to denote not just a single dialect with a small audience. The nucleus of a dialect, absorbing its neighbouring elements as much as possible, would have attracted a more extensive audience. And this appears to have been a specific character of the languages used by Indian Buddhists. It could not be the Sanskrit language (so Obermiller!)." Needless to say Yuyama's remarks are nothing more than pure guesswork. Although Obermiller does not explain why rgyal-po'i skad designates Sanskrit, this is far from being excluded. It is obvious from the text quoted that the new sutras were meant to replace texts composed in Prakrit, Apabhramsa and bar-mar 'don-pa. Obermiller translates the word gźun as 'method', Yuyama translates it by 'Scripture', but gźur often renders Sanskrit grantha and it is probably better to translate the second part of this section as follows: "the Arhats successively composed (the texts) in different scriptures such as the sutras, etc., composed in the royal (or extensive) 1) language."
140
In two other passages we find an expression very similar to bar-mar 'don-pa, i. e. "brin-du 'don-pa. The first is taken by Bu-ston from Padmakaraghosa's Bhiksu-varṣāgrapṛcchä: gnas-brtan-pa / mkhan-po rje rigs... katya-na / skad 'brin-du 'don-pa /. Obermiller translates: "the Sthaviras...
1) Yuyama wrongly translates rgya-chen-po'i skad as 'the language of great extensiveness'. Probably he did not realise that rgya-chen-po'i is a so-called attributive genitive (cf. Michael Hahn, Lehrbuch der klassischen tibetischen Schriftsprache, Hamburg, 1971, p.73). Rgya-chen-po can of course also render Sanskrit udara 'lofty, elevated'. It is not possible to make any suppositions about the nature of the rgya-chen-po language without knowing the original Sanskrit term.

Page #4

Madhyadesika, Madhyoddesika and Madhy'uddesika
141
Their chief preceptor was Kātyāyana of the Vaiçya caste ... They spoke an intermediate dialect.” Yuyama translates: "the Sthaviras: The master was Kātyāyana, a Vaisya ... (They) pronounced the language intermediately ...” Yuyama seems to take skad as the object of 'don-pa but the meaning is obviously: "As to their language, it is the 'brin-du 'don-pa.” The second passage quotes the opinion of some scholars (kha-cig): kha-cig / phalchen-po skad 'brin-du 'don-pa'i skad I. Obermiller translates: “According to some the language of the Mahāsamghikas was the intermediate dialect.” Yuyama translates: “Some (say]: - The Mahāsāmghikas spoke the language, the language (which they) pronounced intermediately." Yuyama's strange translation seems to be based upon the fact that he takes the first skad as the object of 'don-pa. The text has to be translated, of course, as in the previous passage: “Some [say]: As to the Mahāsāmghikas, their language is the 'brin-du 'don-pa language.”
Yuyama remarks in a note (17): “it is difficult to know the exact meaning of the Tibetan bar-mar 'don-pa'i tshig ... it is certainly used as synonymous to skad 'brin-du 'don-pa oro-pa’i skad." Undoubtedly, bar-mar 'donpa’i tshig and 'brin-du 'don-pa'i skad (not skad 'brin-du 'don-pa’i skad) designate the same language. However, it is interesting to note that 'bar-mar 'don-pa is the name of a tshig, whereas the text speaks of a rgyal-po'i (or rgya-chen-po'i) skad. In the two other passages 'brin-du 'don-pa is a skad and not a tshig. Probably skad renders Sanskrit bhāṣā "language', but it is impossible to determine the Sanskrit word translated in the Prabhāvatī by tshig.
Although Yuyama remarks in note 17 that it is difficult to know the exact meaning of the Tibetan bar-mar 'don-pa’i tshig, he writes in another note (27): "Tibetan 'brin or bar-ma must be used for something moderate in regard to the quality and quantity-neither too large or loud (chen/che) not too small or low (chun), neither too strong or high (drag) nor too weak or low (zan). In this case it denotes the language of moderate character with moderate hybridity. At the same time it seems to me that the language also refers to the manner of speech, that is to say, a language which is neither too harsh nor too soft, neither too loud nor too low. By using such a language they must have attracted wider audience. The more active they were in propagation, the more their language must absorb the neighbouring dialectal elements ... The most important thing in our text is that the language of moderate character, no matter how one may translate it, refers to a language compared with other dialects synchronically, and never diachronically. That is to say, it denotes by no means the language in the intermediate stage of the Sanskritization process." Yuyama's opinion does not seem to have met with universal approval. G. von Simson, one of the participants in the conference in Göttingen in July 1976, to which Yuyama contributed this paper, was of the opinion that the expressions bar-ma and 'brin-du referred to a language intermediary between Sanskrit and Prakrit/Apabhramsa (cf. p. 187).

Page #5

J. W. de Jong
Yuyama's gratuitous speculations seem to be entirely based on the meaning of the Tibetan words bar-mar and 'brin-du which both mean 'in the middle'. All the other terms used by Bu-ston, such as tha-mal-pa (Prākṛta), zur-chag (Apabhramśa), legs-par sbyar-ba (Samskṛta), and sa-za (Paiśācī), represent Tibetan renderings of Sanskrit names of languages. It therefore seems reasonable to assume that this is also the case with regard to bar-mar 'don-pa or 'brin-du 'don-pa. According to the texts quoted by Bu-ston it is a language different from Prakrit and Apabhramsa and used by the Sthaviras or the Mahāsāmghikas. We have seen above that 'bar-ma 'don-par byed-pa renders Sanskrit madhyoddeśika, a term used in association with the Mahāsāmghikas. It is therefore very likely that both bar-mar 'don-pa and 'brin-du 'don-pa render Sanskrit madhyoddeśika or Buddhist Hybrid Sanskrit madhy'-uddesika. In the texts quoted by Bu-ston Madhyoddesika is the name of a language but it is by no means sure that this is the original meaning of the Sanskrit term. We do not know the linguistic form of this term in the Prakrit underlying the language of the Vinaya texts of the Mahāsāmghika-Lokottaravādins. The literal meaning of madhyoddeśa would seem to be something like 'intermediate recitation'. It is to be hoped that other texts will throw light on the original meaning of madhyoddeśika and its later development.
142
*
My deepest thanks are due to Dr. Gustav Roth for having been so kind as to send me photocopies of two manuscripts of the Mahāvastu which have been microfilmed on behalf of the Nepal-German Manuscript Preservation Project. In his letter of the 2nd May 1981 Dr. Roth writes: "One of them, the most important one, is registered in our Seminar library [Seminar für Indologie und Buddhismuskunde, Göttingen] under the number Xc 51. It is a palmleaf manuscript of 428 leaves, complete, in a script which points towards the twelfth or thirteenth century. It may be the manuscript to which Senart referred. On fol. 2 a, line 2, the following reading is perfectly clear: arya-mahäsämghikānām lokottara-vādinām madhy'-uddesikänäm pathena vinayapitakasya mahāvastuye ädi /. The other complete Mahāvastu manuscript is written on 446 paper leaves dated: likhito 'yam 815 (1694) A.D.). One finds here the note: PSB 30, Mahävastu, Privatbesitz 456 Bl., 41.5 x 16 cm, Papier. The copy of it in our Seminar is registered under number Xc 50. Here we read on fol. 2 a, lines 2-3: äryya-sarva-sämghikanām lokottara-vādinām madhyumdeśikānām pāṭhena vinaya-piṭakasya mahāvastuye adi /. However, at the end of the MS, fol. 446 a, line 4, the text has: arya-mahäsämghikänäm lokottara-vādinām pāṭhena /. Dr. Roth adds: "It is clear that madhyumdesikänäm is nearer to madhyuddesikānām than to madhyadesikanām. Thus we can accept the readings madhy'-uddeśikānām and madhyoddelikānam as the correct readings for the Mahāvastu and the

Page #6
143 Madhyadesika, Madhyoddeoika and Madhy'uddesika 143 Bhiksu Pratimoksa-sutra, and as characteristic for the terminology of the Lokottaravadins in general." The first manuscript referred to by Dr. Roth is of the greatest importance for the study of the Mahavastu and it is very much to be hoped that it will be edited as soon as possible and/or made available in a facsimile edition. The reading madhy'uddesikanam is most probably only one example of many in which this manuscript must have preserved better readings than those found in the manuscripts used by Senart for his edition of the Mahavastu. * Additional note: Only recently I discovered that in his review of Gustav Roth's edition of the Bhiksuni-vinaya (Patna, 1970), John Brough suggested the meaning 'who preaches the Middle (Way)' for madhyoddesika- (BSOAS 36, 1973, pp. 675-677). Brough does not quote other instances to support this hypothesis; the passage in which he discussed the meaning of madhyoddesika- is the following: It has generally been accepted that the Mahavastu belonged to the 'Lokottaravadins of Madhyadesa'. The form madhyuddesikanam here (madhyoddesikanam in their Pratimoksasutra) may raise doubts. On reflection, it is somewhat curious that a Buddhist sect should specify that it belonged to Madhyadesa; and madhyoddesika- would readily have been corrupted to madhyadesika-, while the converse is extremely improbable. Although uddesa- can mean 'place' as well as 'instruction', we may suggest as a hypothesis for future consideration that the sect designated itself as 'Lokottaravadins who preach the Middle (Way)'. In view of their transcendental doctrines (cf. A. Bareau, Les sectes bouddhiques du Petit Vehicule, Saigon, 1955, 77), this would probably mean a doctrine holding to a middle path between the Sravakayana and the more extreme forms of the nascent Ma hayana.