¹ xvii, 292 pp. Oxford: The Pali Text Society, 1994.

² xiii, 242 pp. Oxford: The Pali Text Society, 1994.

³ This would be a specific instance of the connexions postulated in Norman's, ‘Theravāda Buddhism and brahmanical Hinduism’ (1991), Coll. papers, IV, 271 ff.Google Scholar

⁴ The collection of translated articles is rather liberally misprinted, the error in some cases going back to the original publication. An instance which causes some perplexity occurs on p. 201, line 1, where anusāvanā stands (clearly unintentionally) for anussāvanā of the original, while p. 203, line 7 has anusāvanaṃ without comment for anu sāvanaṃ of the original (where one might expect emendation to sāvaṇaṃ alone, as cited on p. 201, line 15). An entry *udakamma on p. 147 intends to report an omission from CPD: it misprints labhatu for labbhatu, but itself rests on a misprint, for the reading should be udakakamma and this is duly present in CPD. The PTS style has been adopted whereby new material, occurring especially in footnotes but also in the main text, is somewhat surprisingly not signalized as such. Nevertheless, overall the typography and layout of the articles have been very substantially improved by comparison with the original versions, and their compact presentation will benefit others besides the anglophone community.

⁵ ‘Remarks on the Critical Pali dictionary (I)’ from ZVS, 84, 1970, 177ff.; ‘Remarks on the Critical Pali dictionary (II)’ from ZVS, 94, 1980, 10ff., ‘Pāli ulloka-’ from ZVS, 81, 1967, 247 ff.; ‘Pāli kaṭhati: a contribution to the history of the transmission of the Pāli canon’ from IIJ, 21, 1979, 21 ff.; ‘Pāli samaya and Sanskrit samāja’ from IIJ, 29, 1986, 201 f.

⁶ ‘On the history of the name of the Pāli language’, from Beiträge zur Indienforschung Ernst Waldschmidt zum 80. Geburtstag gewidmet (Berlin, 1977), 237 ffGoogle Scholar. K. R. Norman (Coll. papers, IV, 114 f.) concedes it is unlikely that the error arose independently in the various vernaculars.

⁷ ‘Pāli gotrabhū: the origin and early linguistic history of a philosophical term’ from ZDMG, 128, 1978, 326 ff.

⁸ with Norman, K. R., JPTS, XI, 1987, 38 fGoogle Scholar. (Coll. papers, III, 163 f.).

⁹ ‘Vedic nivāté and Pāli nivātake’ from MSS, 23, 1968, 21 ff.; ‘The Vidyādhara's sword’ from WZKS, 22, 1978, 45 ff.; ‘Three Buddhist legal terms: issaravatā, gīva, and bhaṇḍadeyya’ from Indologica Taurinensia, II, 1979, 275 ff.

¹⁰ ‘Gāthā anacchariyā pubbe assutapubbā’ from ZVS, 84, 1970, 5 ff.

¹¹ ‘Linguistic observations on the structure of the Pāli canon’ from StII, 2, 1976, 27 ff.; ‘The oldest literary language of Buddhism’ from Saeculum, 34, 1983, 1 ff.

¹² ‘The “threefold” effect of karma’ from IIJ, 13, 1971, 241 ff.; ‘Traces of the reduplicated aorist in Pāli’ from MSS, 32, 1974, 65 ff.; ‘The development of the clusters -tm-, -dm- and -sm- in Middle and New Indo-Aryan’ from MSS, 40, 1981, 61 ff.; ‘On the perfect in Pāli’ from ZVS, 96, 1982/83, 30 ff.

¹³ ‘Buddhist law and the phonetics of Pāli: a passage from the Samantapāsādikā on avoiding mispronunciation in kammavācās’ from StII, 13/14, 1987, 101 ff.

¹⁴ Kane (1923) and Keith (1928) rightly saw no reason to believe that Daṇḍin used Bhāmaha, rather the reverse. Bhāmaha simply rejects (1.32 ff.) gauḍīya and vaidarbha and the three items prasanna, ṛju, and komala without explaining what they are, i.e. he presupposes the Daṇḍin material.

¹⁵ Deletion of the unwanted Vedic terms sāmyam and -plutāni from the first RPrāt. anuṣṭubh hemistich substantially leaves an aupacchandasika prior pāda. The term sāmya, for which Müller and Renou had no solution to offer, evidently means ‘prosodically neutral length’ such as is deemed by RPrāt. 17.22 to obtain in all syllables of the sama (pāda) except the penultimate. That Buddhaghosa (or his sources) recognized the connection with RPrāt., and a connection between this definition of prosody and the sakāya niruttiyā passage, seems to emerge from his glosses, discussed below: (i) RPrāt. chandojñāna ∼ Sp kārikā vyañjanabuddhi ∼ Sp gloss vyañjananirutti; (ii) RPrāt. chandobhāṣā ∼ Vin II 139 (sakāya niruttiyā …) chandaso (āropema) ∼ Sp gloss vedaṃ viya sakkatabhāsāya.

¹⁶ RPrāt. 1.12 vyañjanāny eva … pañcavargāh … antaḥsthṣaḥ … ūṣmāṇo … aghoṣāḥ. Here, in the context of Prātiśākhya triṣṭubh verse, antaḥsthā will originally represent a Vedic masculine stem. It would appear that vowels are similarly masculine varṇas to which the masculine term svara and the neuter term akṣara are applied, in order to define their prosodic and syllabic functions: but the text presents the vowels even more laconically than it presents the consonants. In these matters, RPrāt. appears to be more primitive than TPrāt.

¹⁷ Uvaṭa ad RPrāt. 1.12: vyañjayanti prakaṭān kurvanty arthān.

¹⁸ cf. Norman, K. R., ‘The nasalisation of vowels in Middle Indo-Aryan’ in Philosophy, grammar and Indology (Delhi, 1992), 331 ffGoogle Scholar. (Coll. papers, V, 107 ff.).

¹⁹ In Bechert, H. (ed.), Die Sprache der ältesten buddhistischen Überlieferung (Göttingen, 1980, 62 fGoogle Scholar. (Coll. papers, II, 129 f.).

²⁰ JPTS, XVI, 1992, 83 fGoogle Scholar. (Coll. papers, V, 77 f.).

²¹ JOIB, XX, 1971, 329 ffGoogle Scholar. (Coll. papers, I, 122 ff.).

²² The coincidence seems to imply that a fanciful rendering of the two elements chando- and -bhāṣām, as vedam viya and sakkata-bhāsāya respectively, is superimposed on chandaso: In the RPrāt. context, ‘metrical text’ must be meant: ‘Vedic language’, as chandobhāṣā is usually rendered, can hardly be right.

²³ Norman, K. R., JPTS, XVII, 1992, 217 (Coll. papers, V, 82)Google Scholar.

²⁴ MW offers ‘water’ for himāriśatru in BCar., perhaps by confusing the two passages. I thank Dr. R. Söhnen-Thieme for drawing my attention to Johnston's suggestion of a connection between Daṇḍin and BCar.

²⁵ ‘The man who is beset by the rays of (the father of the enemy of the son of the loser to the bird =) the sun welcomes the sky filled with (the holders of [water] which is hostile to [fire] the banisher of cold =) clouds.’ For the key to the long compound, see below.

²⁶ Johnston's emendation -ghātane (‘probably’ consistent with the Tibetan) is borne out by the rūpaka thus yielded. A contrast antare versus āntare is to be expected in such a yamaka rhyme.

²⁷ Johnston emends the Chinese by replacing ‘bird’ with an improbably Ṛgvedic sense ‘fire’: but his translation to the effect ‘just as fire goes out in the rain’ can surely be ruled out. The alleged occurrence of ‘fire’ in both pādas (himāri, dvija) would destroy the symmetry of the simile, and the desiderative participle (yāti vimokṣayan) has to imply intent. Daṇḍin's verse alludes to the quenching of fire by water, but this replaces himāriśatru, not dvija.

²⁸ thus different from the numerical uses of RV ékasmin ‘in only one’.

²⁹ whereby presumably the passive base was associated with kaṭṭha and the eventual active base with kas-. The postulation of a denominative from the participle is thus unnecessary.

³⁰ as distinct from the forms involving assimilation and double aspiration: aḍḍhuḍḍha (ardhacaturtha) and pavaḍḍhanti abhivaḍḍhaṃ (pravardhanti abhivṛṣṭam).

³¹ On the strength of the Dhātup. gloss alone, kaḍḍ- is assigned to the kaṭhina group at DEDR 1148.

³² the more so, since the previous 14 items (e.g. bhayakathā) are definitely ‘tales of …’. As the final items are also certainly ‘tales of …’, the compromise ‘tales of goings-on at …’ seems to be wanted for visikhākathā, kumbhaṭṭhānakathā, rather than ‘street-corner gossip, well gossip’ as in Rhys Davids (D I 7) or ‘description of streets’, etc. as in the commentary.

³³ Belvalkar, I, [33] protests that prekṣārtha and śravya in v. 39 are two subdivisions of drama, not of literature, but this is no solution. Daṇḍin has kept this twofold classification of genre for the closing verse of the section that deals with genres, preferring to work hitherto using the threefold classification of v. 11; and his treatment of drama explicitly begins and ends in v. 31. There is no problem if we assume that his prekṣārtha genre corresponds to the Pali heading -dassana, a representational genre in which drama probably does not figure.

³⁴ 72 sómo sutá ṛjṣeṇājahān mṛtyúm,

ṛténa satyám indriy´m vip naṃ śukrám ándhasa

índrasyendriyám idám páyo ՚ mádhu.

73 adbhyáḥ kṣīráṃ vy àpibat krúṅṅ āṅgirasó dhiy ,

ṛténa satyám, etc.

74 sómam adbhyó vy àpibac chándasā haṃs´ḥ śuciṣát,

ṛténa satyám, etc.

75 ánnāt parisrúto rásaṃ bráhmaṇā vy àpibat kṣatráṃ páyaḥ sómaṃ praj patiḥ,

ṛténa satyám, etc.

King Soma escaped death with the last remnants, this is the ‘drinking out’ of the juice, the milk, the ambrosia. The crane of the Angirases with its thought ‘drank out’ the milk from the waters. The goose in heaven with its prosody ‘drank out’ the soma from the waters. Prajāpati with his word ‘drank out’ (sap from the liquid food) kingship, milk, soma.

³⁵ Hence perhaps the later use of nāda to designate primordial sound, and certainly the identification of madhyamasvara as the note of the crane. The VS passage appears to associate the crane's call with pure dh. The Pali context may suggest that, based on 72 ajahān mṛtyúm, there has subsequently been an association of ideas with the cranes' alarm call: ‘their high-pitched kurr, kurr, kurr uttered in varying keys has been aptly likened to the distant roaring of the sea’ (Sálim Ali).

³⁶ RV 1.71.2 vīdú cid dṛḍh pitáro na ukthaír ádriṃ rujann áṅgiraso ráveṇa: the Angirases broke the rock with their song, so creating a path to high heaven (2c) and facilitating the flow of dhītí (3a), the provision of soma-milk as práyas (3d), and the passage of Bhṛgavāṇa Agni (4b).

³⁷ The commentator's ‘ignorant’ (so CPD, etc.) for añña seems implausible and irrelevant.