Chapter IV - Mañjarī-jātaka

The text throughout refers to him as “the Kauśikan,” i.e. a member of the Kośika clan, but it simplifies translation to speak of him as “Kośika,” that is, to call him by his clan name. In the Pali Jātaka the corresponding character is Kosiya or Maccharikosiya (“the miserly Kosiya”). Kośika (Kosiya) was the name of a brāhman clan, several members of which are named in the Pali texts (See D.P.N.). There was one actually living in Gotama’s time who from his miserliness was dubbed Maccharikosiya, the exact name of the mythical character in the Sudhābhojana Jātaka.

Vāhitakena mārgena, See p. 27, n. 6.

This is one of the ṛddhis (iddhis) at 5. 5. 282, etc.

Jñātika, Pali ñātaka. But in the Pali version Kosiya is Pañcaśikha’s son and successor. In that version, too, all the other characters are mutually related. Kosiya was the fifth in descent from the wealthy householder who was reborn as Sakka, and whose successive descendants were reborn as Canda, Suriya, Mātali and Pañcaśikha.

Devas of the sun and moon respectively, in Pali Canda and Suriya. See preceding note.

Saṅgrāhaka, Pali saṅgāhaka.

Literally, “exhibited various high and low appearances,” nānāprakārāṇi uccāvacāni varṇāni upadarśeti. As no allusion has so far been made to Pañcaśikha’s transformation, Senart assumes a lacuna in the text. But comparison with J. makes it obvious that this is misplaced rather than incomplete, and should come on p. 52 (text), where Kośika is described as being driven by the various transformations of Pañcaśikha to ask his visitors who they really were.

The baldness of this version as compared with the minute and picturesque detail of the Pali is to be noted.

Reading nālam (na-alam), which is found in one MS, and also in J., for nalam of the text.

The text of this stanza begins natvāham, which, if correct, puts the negative required by the sense in a strange place. The na is required rather with vindate of the next line. Following the analogy of the Pali taṃ taṃ vadāmi, the text has, therefore, been emended into tamtvaham (tam-tu-aham, “this I tell you”) with the negative before ekāṃśam. As the word preceding the latter ends in -nno, the careless omission of na is easily explicable as a piece of haplography.

Reading samīhita, as in J., for samāhita of the text.

The text here is very unsatisfactory. As printed it reads, Sarasvatīṃ so juhoti cāhutāṃ gamaye api. This, as Senart hesitatingly suggests, could only mean something like, “he pours the Sarasvatī in libations and makes it, after being thus poured, flow again”.

The corresponding Pali is:

Sarasañ ca yo juhati bahukāya Gayāya ca
Doṇe Timbarutitthasmiṃ sīghasote mahāvahe
Atra c’assa hutaṃ hoti atra c’assa samīhitam.

That is, as the Commentary suggests, sacrifice to a stream or pool brings its reward only for the man who charitably shares his board. On the analogy of the Pali, cāhutāṃ gamaye has been emended into ca bahukāṃ Gayāṃ. But even with this emendation the Mhvu. falls short of the full meaning of J., for it does not express the idea that the effectiveness of the sacrifice in question is dependent on charitableness (cf. the sentiment of Mātali’s gāthā), but merely the idea that charity is as effective as such sacrifice. (For Gayā as the name of a pond as well as a town, see D.P.N.).

Pāṇisvara—a musical instrument played with the hand.

Reading suptaṃ tarn for suptam na. Cî. J., suttaṃ etam.

I.e. in the music. The allusion here is to the musical abilities of Pañcaśikha, the Gandharva. (See D.P.N. for references.) He pays, that is to say, for any hospitality given him with music. For once, the Mhvu. is fuller and clearer than J.

The latter reads:—

Pāṇissarā mutiṅgā ca murajālambarāni ca
suttaṃ etaṃ pabodhenti, paṭibuddho ca nandati.

Here the tabours, etc., awake Pañcaśikha, which is the exact opposite of what the Mhvu. says. J. introduces the verse as being Sakka’s eulogy of Pañcaśikha, but there does not seem to be much merit in being awakened by music; one need not even be a musician. It would seem, therefore, that the text of J. is here defective.

Senart assumes a lacuna before these words, and remarks on the coincidence that both here and previously (p. 49, text) a lacuna occurs precisely where reference would be expected to be made to Pañcaśikha’s transformation into a dog. But as has already been suggested, the latter lacuna is obviated by a transposition of the text, although we would still be left without an explicit statement regarding this transformation. As for the second of Senart’s lacunas, a comparison with J. would seem to suggest that, if it really exists, it does not represent the omission of a reference to Pañcaśikha’s being changed into a dog. Really, however, we have to do not with lacunas, but with the disconcerting brevity of the prose connecting the gāthās, which are left to tell the story themselves. In view of the abruptness with which the gāthās are introduced, the second lacuna need be no more than the omission of Pañcaśikha āha. But even this assumption is unnecessary, and the following verses may equally well be put in the mouth of Śakra, the speaker of the preceding verses. For, according to J., Śakra was as much of a jñātin (“kinsman”) of Kośika as Pañcaśikha was.

Dadyād, 3rd pers. for 1st. Though it would be equally easy to read dadyām here, Senart supports the text reading by citing a similar inflexion in Vol. I, p. 51 (text), adrakṣīt, which may be explained, he says, by Prakrit morphology.

Kṣipihanti, Prakrit fut. act. for passive.

Sugatim. This word is entirely Senart’s conjecture. It does seem rather out of place with the verb pravrajati, which in our text is invariably used in the sense of “to go forth as a recluse,” etc., although the simple verb vrajati is used above in its ordinary sense, “to go,” with sugatim. In J. the corresponding verse has tato aham pabbajissāmi Sakka, and possibly, therefore, the right restoration of our text would be the insertion of Śakra after the reading tato aham pravrajiṣyāmi of two MSS., “hence, Śakra, I’ll go and be a religious.” Cf. note 5.

One of the five mountain ranges that encircle Anotatta. See D.P.N.

Another instance of the abrupt introduction of gāthās without any introductory prose. In J., however, we are told of how Kośika retired to the Himalayas where he lived to old age. In the meantime Sakka’s four daughters, Hope, Faith, Glory and Honour came to Mount Manosilā in Tāvatiṃsa, where Nārada, a brāhman seer, had gone to rest for the day. The nymphs asked him for a branch of the flowering coral-tree which he held in his hand.

Reading upāgato ṛṣivaro sarvapūjito for upāgatā ṛṣivara sarvipūjitum of the text. Cf. J., athāgamā isivaro sabbalokagū.

I.e. the coral-tree of Indra’s heaven, pāricchatta(ka) in Pali. It is called pāripātra at Mhvu. I. p. 32 (text). See Vol. I, p. 27, n. 1.

Literally, “of us,” mo. See Senart’s note Vol. I, p. 601, where instances are given of the use of this form for nom. acc. and gen. pl. But we should, perhaps, here read no, as in J.

Senart’s text here is sharply at variance with that of J., and, to judge from the readings of the two MSS. which he gives in the apparatus, with the MS. tradition of the Mhvu. itself. His reading is śākhām dadensurnama astu māriṣa, “Let them give the branch, we pray; let it be so, sir.” But as J. makes it clear and as the context demands, dadensu (3 pl. aorist in optative sense) should be changed into an imperative or optative singular. The two MSS, referred to read respectively tām hi śākhām dadet sarvaṃgamāsi mā° and tām hi śākhām dada sarvādgamāsi ma.° Sarvam (-ād) gamāsi is obviously reminiscent of the sabbagatī in J., dadāhi no sabbagatī ijjhantu, and seems to be the basis of Senart’s reading °surnama astu. Without a collation of all the MSS, however, it would be unwise to attempt the restoration of the line as a whole, and satisfactory sense may be obtained by merely reading dadet for dadensu. I.e. “Let our worthy friend give and be = You, worthy friend, give, and you will be.”

I.e. the spokeswoman of the four sisters. Senart’s note, in which he claims the sense of “offer” for the middle yācamāna, shows that he misunderstood the whole passage.

Senart prints gṛhṇātū (“let her take”) in brackets as it is a substitute for the MŚS. reading of which he cannot make anything. It is probable, however, that pi ta in the MSS. reading so pi ta° is a remnant of a reading which corresponded to piḷayhatthā in J. This latter form is from piḷayhati, shortened from apiḷayhati or apiḷahati, Sk. apinahati, “to tie on,” “wear,” which itself is sometimes found shortened into pinahyate. The original Mhvu. reading may thus have contained a form of this verb with either ḷ or n, and the translation has been made on this supposition.

Reading ko, with J., for sa of the text.

Literally “a sound of anger,” krodhaśabda.

I.e., Indra.

Paramārthadarśino. This epithet sounds out of place here. J. has paramappakopitā “greatly stirred.”

Puraṃdara. It is worth noting that this Vedic epithet of Indra (Śakra, Sakka) as the god who “shattered the fortress” or the cloud, and so released the life-giving rain, has in the Pali texts been distorted into purindada, which was variously interpreted as “a giver of alms in times past” (VvA. 171), “a bestower of the gift of dharma” (MA. 3. 98), or “a giver of gifts from town to town” (S. I.229). For an interesting note on the Pali term and its implication of a belief in the identity of the Buddha and Sakka, see Miss I. B, Horner on M. I.386 in her forthcoming translation of that work.

Reading āyattamanā as in J., for anāttamanā of the text.

Thehapūrvam = (a) tha-iha-pūrvam. But the reading is suspect.

The interlocutors are supplied from J.

Reading prāpaya (as in J.) devasārathi for preśayi devasārathis of the text, which latter would make Mātali himself send the ambrosia. But in J. we read of Sakka deliberating and finally deciding to send Mātali with the ambrosia.

I.e. to Mātali when he arrived with the ambrosia.

I.e. the like of the ambrosia.

Literally, “destroy,” imperative, if, that is, the lacuna in the text .... ] hi is to be supplied by jahi. J. has hanti.

Reading ekāsanam (with bhojanam understood. See references in P.E.D.) for ekasya na of the text, which, as Senart’s translation shows, gives an incongruous sense.

The text has mitraṃ opayikam, which is obviously incorrect. The corresponding passage in J. (5. 397) enumerates four classes of evil doers, to which misers are added as a fifth. That the Mhvu. also intends the same would seem to be indicated by the presence of the numeral adjective pañcama in two MSS. Mitraṃ opayikam, “a suitable friend,” therefore, cannot possibly be the right reading. Senart says that it was suggested to him by one MS. (L); but a better clue to the right conjectures seems furnished by another MS. (c), which has mitra oparipattika. The latter would suggest the Pali paripāṭeti or °paṭeti, causative of paripaṭati, in BSk. paripāṭayati (Divy. 417), “to destroy,” “ruin.” The sense, therefore, would be satisfied by a form of the causative of pat, perhaps opātenta (= avapātentā), “those who ruin or betray friends.” But it is not safe to make a definitive emendation without collating all the MSS. It is to be noted, further, that for the reading of the text ye (a)paharanti artham, “robbers,” two MŚS. have ye svapanti artham, where the verb seems to be an echo of the Pali sapanti in J.’s sapanti subbate “curse holy men.”

Literally “eat,” nāse for nāśe from na-aś, “to eat,” and not as Senart says from naś “to destroy” J. has nāsmiye = na-asmiye, indie, pres. mid. of asati, “to eat,” in sense of future.

The text reads sudhām pi ādāya pratigrahārhām, “(the maidens) bringing the ambrosia that was worthy of acceptance.” But it was not they who brought the ambrosia to Kośika; it had already been brought by Mātali, and they themselves were come to ask for a share of it. The translation, accordingly, follows J., the corresponding verse of which names the four maidens thus Āsā ca Saddhā ca Sirī Hirī tato. The four corresponding Sk. names, though not in the same order, could plausibly be substituted for the text reading. Sudhām and ādāya, at any rate, readily suggest Śraddhā and Āśā, and perhaps the whole verse originally read, Śraddhā ca Āśā ca śiri Hirī tathā.

Tāravarā oṣadhi. See references in P.E.D. (s.v. osadhi).

Kāñcanavedivigrahe. For vedi J. has velli, and the Commentary explains the compound word as kāñcana-rūpaka-sadisasarīra, “body like a form of gold.” It is interesting to note, however, that velli at J. 6. 456, where it is similarly applied to describe the slender waist of a woman, is explained by the Commentary as vedi (“rail,” “cornice”), the actual form in our text. See J. trans. (5, p. 213, footnote) for references to passages which would seem to make velli = some part of dress.

The last pāda but one of this stanza is wanting. What is printed in its place is the duplicate of the third pāda of the next stanza. The missing pāda is supplied in translation from J., tayā vihīnā na labhanti kiñcanam. The last pāda is misplaced as the concluding pāda of the next stanza, tadidamasādhu yadidaṃ tvayā kṛtam, which corresponds to the Pali tayidaṃ na sādhu yadidaṃ tayā katam, but is here, for the purpose of the translation, restored to its proper place, replacing preśeti dāsaṃ viya bhogavāṃ sukhī which, in turn, is transferred to its proper place as the fourth pāda of the next stanza.

Reading jātimantaṃ pi for jātimantiyā (sic) of the text.

In J., Hope is the next to be addressed.

Manorameśāhvayanāṃ diśam. Senart admits that his restoration of the text is very doubtful here, and he can adduce in support of it, only the expression eśānī diś, a term used for the “north-east.”

See n. 2.

Reading se = asya for me.

Sadṛśā, so translated after the Commentary’s explanation of sadisīsu in J., jātigottasīlādihi sadisīsu.

Reading sukulāṃ dhītaram for svakulaṃ dhītarā of the text.

Reading kumbhadāśiye, “the slave-girl who carried the pot,” for °kāsiye of the text. The latter form naturally puzzles Senart and prompts him to suggest °kāriye, and to imagine a reference to a tale in which a man leaves his high-born wife for the wife of a potter!

Śrutam, unless we read tyāgam (“charity”), corresponding to cāgam in J. The text of the next pāda is very corrupt. The translation of the distich follows J., in which the second pāda is ādāya saddhāya karonti h’ekadā. Corresponding to this our text has śraddhā satī yatra [ ] ekadā. The similarity between the Sk. and the Pali is sufficiently close to justify translating from the latter, but the apparatus contains nothing to help towards a full restoration of the Sk. pāda.

Naiṣa sāvadyena vighātadarśanā. This is rather obscure, but the obscurity is due to the abridged character of the Mhvu. text rather than to defects in it. It is worthy of special note that sāvajja ana anavajja are the adjectives used by the Commentary on the corresponding passage in J. to denote the wrong (faulty) and the right (faultless) type of faith respectively. The right kind of faith has been instanced in the first half of this stanza, but there is no example of its contrary in the second half, only a statement, almost a definition, of its nature, much as though a whole pāda had developed from a gloss (sāvadya).

The text here has jahāti rātrī, but Senart doubts this reading, and we should probably read jaghanyarātrīm, “when night is low (in the west)”; the whole pāda then corresponds to the Pali dighaññarattiṃ aruṇasmi ūhate (J. 5.403), which, however, there introduces the appearance of Honour, not Hope as here.

The text has sara (= śara) bhāya varjitā, which Senart translates “jetée en proie au śarabha.” It is difficult, however, to see the point of throwing a mṛga as prey to a śarabha, practically one deer to another. Seeing that an authority like Senart overlooked the possibility of it, the translator is diffident in making what seems to him the obvious conjecture that sarabhāya hides a form of sara (or śara-), “fear of an arrow,” perhaps the ablative sarabhayā (for sarabhayād). At any rate, this gives a more plausible sense and has been adopted for the translation. It is, moreover, fairly close to the reading of J., sarocāpadhārinā virādhitā “missed by one holding a bow and arrow,” i.e. by a hunter.

Senart leaves a lacuna here, reading [ ] pravarāsmi devatā, but as one MS. reads masakkasāra pra° the whole pāda has been restored as masakkasāraprabhavāsmi devatā, which makes it identical with J., masakkasārappabkav’ amhi devatā. Masakkasāra was a name given to Trāyastriṃśa (Tāvatiṃsa). See D.P.N.

Ālambane, literally “on what supports them,” if, that is, this is the right reading, of which Senart is far from assured. Anyhow the translation here given would seem preferable to what Senart proposes for ālambane tatra sīdanti, “ ils sont uniquement appliqués à ce but,” à savoir, gagner de l’argent. J. has te tattha sīdanti atho pi ekadā.

Lacuna in text.

Sīhāṅgadā. Senart takes sīha to be corrupt; but one may compare Pali sīhakuṇḍala, “lion earring” (J. 5. 348). Sīha is used for siṃha above p. 34, n. 5. J. has cittaṅgadā, “spangled bracelets.” Note that this part of the description of Honour is applied in J. to Hope.

Kāñcipramṛṣṭadhāraṇī. J. has kambuvimaṭṭhadhārinī, “wearing bracelets of burnished shells.”

The text has kuśāgraraktā, “coloured like the tip of kuśa grass,” which hardly makes sense here. J. has in the corresponding pāda, kusaggirattaṃ apilayha manjarim, “wearing a sprig the colour of flame fed by kuśa grass.” It would seem easy to emend kuśāgra° into kuśāgni, “the colour of the kuśa flame,” and understand it as being applied to the hair.

The fragrant root of Andropogon Muricatum.

The description now comes back to that of Honour as in J. Our text has for the second pāda alaṃkṛtā lohitamālinā, followed by a lacuna. The first pāda reads gate, yathā prāvṛṣi atra sārade. Some noun is evidently wanted here to complete the simile implied by yathā. J. has for the first pāda kālā nidāghe-r-iva aggijāt’ iva which J. trans., following the Commentary, takes to refer to a plant growing on burnt ground and identified in a footnote as ipomoea. The second pāda in J. reads anileritā lohitapattamālinī, and the corresponding one in our text has been emended and completed accordingly as anileritā lohitāpatramālinī. The latter compound is taken to qualify the implied word for “plant” required to complete the simile. Neither J. nor Mhvu. text seems above suspicion, but the Pali has in its first pāda a more definite allusion to a particular plant.

Reading sudheṣinî for sukheṣinī of the text.

In spite of lakṣate on p. 58 (text), which Senart takes to be for passive lakśyate, “est vu,” it is better to take lakṣase here either as a mistake for lapsyase (=labh) or as a Sanskritisation of Pali lacchasi in the corresponding pāda at J. 5. 404. Otherwise the sense is not clear.

Dharmeṇa.

Diospyros embryopteris.

Hyperanthica moringa.

Also rodhra, Pali lodda. Symplocos racemosa.

Lacuna in text, but no more than a conjunction seems lacking.

Pāṭalā. Bignonia suaveolens.

Barringtonia acutaṅgula.

“A species of tree with beautiful flowers, so-called either because it is similar in some way to the sesamum plant, or because it is used as an ornament on the forehead”.(M.W.)

Nauclea cor difolia.

Michelia champaka.

The text has prasātikā śyāmaka tatra taṇḍulā, literally “(a kind of) small-grained rice, millet (Panicum frumentaceum), and (rice) grain (or, “a vermifuge plant”). The allusion is obviously to certain plants, not to grain, but there is no means of ascertaining what particular kinds of rice-plant are meant. Hence the paraphrase above.

This pāda is too corrupt for solution, and even as restored by Senart—āśītakurvī harate niṣaṇṇo—does not admit of a coherent translation. The translation given above is of part of the corresponding pāda in J. (p. 407), Hirim etad abravi. The karate of our text would seem to suggest that originally the pāda contained the word Hirīm, if not also etad. In any case, the assumed translation fits in with the next pāda, which is practically identical with the corresponding one in J.

Reading jaṭājinaṃdhārin for jaṭāyantaṃ dhanena.

Reading yad icchamānāya for jayettamānaye.

Literally “in haste,” tvarito. See next note.

Our text here reads sā adhyabhāṣi tvaritā mahāmunim, but this is obviously incorrect. For the reply of Honour is given a little further on, when she is said to be actually holding the gift in her hands. It is a simple matter to amend this pāda into sudhāmdbhydharṣīt tvarito mahāmuni, on the analogy of J.’s sudh’ābhihāsi turito mahāmuni.

Name of Śakra (Indra). Several explanations are given of the title. See D.P.N.

Abhilabdhe,? aorist. If this form is inadmissible, Senart would suggest the past part, abhildbdhā in an active sense.

? (a)vatārād, i.e. where the charioteers “alighted” and the chariots were kept.

Literally “went,” gatās. The participle is taken to imply that the pictures or images were realistic and the animals depicted instinct with movement. Cf. J. mig'ettha... yuddhāyutā, “Here antelopes and deer are seen as if prepared to fight.”

Upāgatā, see preceding note.

Two inexplicable words. J. affords no clue to their interpretation.

Suvarṇacandra. J. has suvaṇṇacanda, which the translation (5. p. 217) renders “peacocks of gold,” although the Commentary explains by suvaṇṇamayā candakā ettha rathe. The verse passages in J. trans. are too often poor guides for construing the text, accuracy being sacrificed for the sake of poetic style.

Lacuna.

See p. 52, n. 1.

Lacuna in text. As, however, the epithets applied to the others correspond in both J. and our text, it is justifiable to supply here the Sk. equivalent of the addhā of J., viz. ardrā, “moist, soft.” The expression “partial jade” is borrowed from J. trans.

A paraphrase selected to suit the context, rather than an exact translation of pravarjitā, a conjecture of Senart’s, which he admits to be very doubtful. For it demands a strained application of the technical sense of this word, “to place in or on the fire” (in a sacrifice), from which Senart would deduce the meaning of “mis au feu,” “éprouvé.” The MSS. have pravrajitā, which would suit the context, but is unmetrical. Could not the right reaḍiṅg, however, be the causative form pravrājita “driven away” (from the fight), i.e. fleeing?

Reading vipadyamānā for vipaśyamānā, which is Senart’s conjecture for vipuny° of the MŚS., but which he admits is an expression “d’excessive platitude.”

Reading śarair upadrutā for surabhi upadrutā. Senart is at pains to prove that surabhī can mean “humble” or “resigned.”

Mahallike. According to P.E.D., mahallika is “a distorted mahariyaka > ayyaka > allaka.” Both mahallaka and mahallika ate found in the Mhvu., e.g. 1. 262, 302; 3. 206, 265. At 3. 265 mahallikā is a feminine form.

Reading dṛṣṭim for dṛsṭam. Senart leaves a lacuna for the verb, but this may be supplied in translation from odahi (< o = ava + dhā) of the corresponding pāda in J., which is otherwise identical with our text.

A name for Brahmā or Mahābrahmā deva. (See DPN.)

See p. 49, n. 4. Kauśika is a well-known name of Indra.

Tasyaiva bhāvasahavratānuja (voc.), but the last compound is suspect. The apparatus criticus, however, affords no help to bring the pāda nearer the corresponding one in J., which reads ajjeva tvam Indasahavyatāṃ vajā ti “to-day enter into fellowship with Indra.” Perhaps we should read accordingly °vratāṃ vraja. Sahavratā is, of course, the BSk. equivalent of Pali sahavyatā.

Ucchraya. Cf. samucchraya, Vol. I, p. 134, n. 1.

Asadhuryabhūta. So interpreted by Senart.

This implies that the brāhman seer who has throughout been referred to by his clan name Kośika (Kauśika) had as his personal name Nārada. But in J. the latter is the name of the seer who first caused the dispute between the four nymphs by offering a branch of the Pāricchattaka tree to the best of them. He is identified with Sāriputta, not with the Buddha as here, while Kosiya (Kośika) is identified with an unnamed dānapati bhikkhu, “a monk of lordly generosity.”