There is one occurrence in the Rākṣasakāvya (of Ravideva) stanza 3cd :
abdaḥ sitābhir asitaḥ śabalaḥ
khagābhir vārjaiḥ sitair iva vibhāti vanaṃ prabuddhaiḥ |
(note that in stanza 14, vārjaḥ is also with the meaning of jalajaḥ but not for the lotus this time)
Dear colleagues,I seem to encounter the word vārja in the meaning 'lotus' in an 8th-century Sanskrit inscription. I don't find any entry vārja in the usual dictionaries. However, I do find it in Narahari's Rājanighaṇṭu (10.173):pāthojaṁ kamalaṁ nabhaṁ ca nalināmbhojāmbujanmāmbujaṁ śrīpadmāmburuhābjapadmajalajāny ambhoruhaṁ sārasam |paṅkejaṁ sarasīruhaṁ ca kuṭapaṃ pāthoruhaṁ puṣkaraṁ vārjaṁ tāmarasaṁ kuśeśayakaje kañjāravinde tathā ||This is from GRETIL: I am unable to check any printed edition. But the meter (Śārdūlavikrīḍita) seems to guarantee that vārjaṁ here is not a typo for vārijaṁ (on which, see Rau, 'Lotusblumen', 1954, p. 510). According to C. Vogel, Indian Lexicography (1979), pp. 376-7, this Rājanighaṇṭu would be no older than 1375.Does anyone know a first-millennium attestation of vārja- 'lotus'?Is it more likely that we confront here a phonetic development from vārija-, or the use of the Vedic stem vār- in a post-Vedic compound vār-ja-?Thank you for any comments and further references.Arlo GriffithsÉcole française d'Extrême-OrientUniversité de Lyon 3 – Jean Moulin
On the pre-1000 AD date of the text, cf.
P. K. Gode, "Date of Rākṣasa Kāvya or Kāvyarākṣasa — Before A.D. 1000", In: Studies in Indian Literary History vol. 1, pp. 195-202;orig. publ. in Journal of Indian History, Madras, vol. 19/3, 1940 pp. 312-319
+ Indology list in last October for more references.
Note that the vār- stem, even if a bit archaic, is nor less classical than the use of nṛ- : both are given as paradigms in traditional manuals like the Siddharūpam (cf. also in Hanxleden's Grammatic Grandonica) and also used in yamaka poems by Vāsudeva (the possible son of Ravideva, both probably poets at the court of the Cera Perumals of Mahodayapuram in the 9th-10th centuries AD).