¹ What is in question is probably oversimplification more than misunderstanding. For instance, Zaehner, R. C., who elsewhere in his Hinduism (London, 1962)Google Scholar shows an adequate appreciation of this point, can also say tout court, “in the Rāmāyana we are shown what the Hindus conceive to be a perfect life—the life lived by the God Vishnu in his incarnation as Rāma” (p. 15) and, “Already in the Mahābhārata and the Sanskrit Rāmāyana the gods other than Vishnu and śiva had sunk pretty low” (p. 187).

² Das Râmâyaṇa (Bonn, 1893 [repr. Darmstadt, 1970]), 65.Google Scholar

³ Of the Critical edition, published from Baroda since 1960 under the general editorship successively of G. H. Bhatt and U. P. Shah, there have to my knowledge so far been published the first six books in their entirety and two fascicules of the Uttarakāṇḍa. The present study makes use of this edition, to which all references in numerals only refer, and conclusions reached are based on the text printed in it. References to the Mahābhārata similarly are drawn from the Critical edition published at Poona.

⁴ To be included in these additions are most if not all the verses in longer metres (distinguished by the sign [l.v.] from the śloka passages), both those at the end of sargas and even more those included in the middle of sargas, such as 4.27–30 and 5.45–6. There are also certain śloka passages which I regard for reasons of language and style as later than the core of the epic (some of the arguments for this may be found in my articles, “The verbal system of the Rāmāyaṇa” and “The nominal system of the Rāmāyaṇa”, JOIB, XIX, 1969–1970, 1–34 and 369–415)Google Scholar; they include the following sargas referred to in this article: 2.1–30, 57–8, 61, 65, 85, 94–5, 98, 102, and 106, 3.8–11, 13, 15, and 60, 4.17–18 and 39–42, 5.33–5, 6.47 and 105. Later still probably are the Bāla- and Uttarakāṇḍas.

⁵ See Jacobi, Das Râmâyaṇa, 131. One may note that Molé, Marijan (“Deux notes sur le Rāmāyaṇa”, in Hommages à Georges Dumézil (Bruxelles, 1960), 140–150)Google Scholar has to admit that “La seconde épopée indienne, le Rāmāyaṇa, n'a pas été étudiée de ce point de vue”, that is, the approach of Dumézil and Wikander, and concludes his article by conceding that the reflection of the tripartite structure he sees in it “n'est plus qu'un élément archaīque ne jouant qu'un rôle marginal dans l'épopée de Vālmīki”.

⁶ While all references to Viṣṇu are examined in a separate section below, one may already notice that Viṣṇu is even less prominent than A. Guruge recognized when he said (The society of the Rāmāyaṇa, Maharagama, 1960, 222)Google Scholar: “although the general impression is that the Rāmāyaṇa extols Viṣṇu, the number of similes in which Viṣṇu is upheld as a standard of comparison is about one fourth of the similes glorifying Indra, while Brahmā and Rudra together have less allusions to them than Viṣṇu.” A still less accurate picture is given in a more recent work, Sharma's, RamashrayaA socio-political study of the Vālmiki Rāmāyaṇa (Delhi, 1971)Google Scholar, through reliance on the North-western recension.

⁷ The description of the mourning town of Ayodhyā provided later poetasters with a chance to run riot. Jacobi (Das Râmâyaṇa, 120) long ago commented on the occurrence of 16 similes in as many stanzas and this sarga has by far the greatest number of similes of any in the Ayodhyākāṇḍa, more than double the average proportion of long compounds, and several unusual features of language.

⁸ Elements of poetry in the Mahābhārata (Berkeley, 1964), 29–30.Google Scholar

⁹ Das Râmâyaṇa, 131.

¹⁰ This passage seems to belong to the second stage of growth, since it forms part of Sitā's discourse dissuading Rāma from killing the rākṣasas without cause. In another episode Indra is not involved in the Rāmāyaṇa, though he is in other versions: this is the story of Anasūyā's asceticism at 2.109.12 (cf. Gonda, Jan, “A note on Indra in Purāṇic literature”, Purāṇa, IX, 1967, 222–261).Google Scholar

¹¹ On these phrases see my article “Stereotyped expressions in the Rāmāyaṇa”, JAOS, 90, 1970, 210–27, especially 220. On the subject of man's fate after death see the separate section below.

¹² The frequency of Brahmā takes no account of the common use of brahma- initially in a compound with the general meaning “divine”.

¹³ Aspects of early Viṣṇuism (Utrecht, 1954), 140Google Scholar; cf. Hopkins, E. W., Epic mythology (Strassburg, 1915), 210Google Scholar. One may also refer to Hacker's, Paul publications, summarized by him at WZKSO, V, 1961, 110–112Google Scholar, for “the existence, in the first centuries B.C. and the first centuries A.D., of an influential Hindu sect that adored Brahmā as the highest deity”.

¹⁴ However, the description of Brahmā as four-faced appears only at a later stage of development (cf. 5 App. I. 13.26).

¹⁵ In assessing frequencies, I have disregarded such stock formulae as hanumān mārutātmajaḥ, on which see my article “Stereotyped expressions in the Rāmāyaṇa”.

¹⁶ Knipe, David M., “The heroic theft”, History of Religions, VI, 1966–1967, 328–360.Google Scholar

¹⁷ The name Śaṅkara is found more in the later stages of the epic, e.g. 2.1062^* 1 and 6.47.122b [l.v.]. The latter reference is cited by Jan Gonda in support of his statement (Viṣṇuism and Śivaism, 90) that “In the Rāmāyaṇa Śiva … is as a rule not regarded as higher than the devas.” However, earlier in this same obviously expanded sarga Rāvaṇa surrounded by the rākṣasas is like Rudra, lord of the immortals, surrounded by ghosts (bhūtair vrto rudra ivāmareśaḥ, 9d [l.v.]; cf. 6.59.12d). Also, at 3.910^* a stanza in longer metre found in several manuscripts of the Northern recension refers to Umā and Jagadīśvara. At 6.105.2a Mahādeva is three-eyed (ṣaḍardhanayanaḥ).

¹⁸ Umā figures under that name in the Bāla- and Uttarakāṇdas (and Śiva is called Umāpati) and in the third stage of growth of the text, when also Śiva is called Nandīśvara (6.1026^* 8). Skanda appears also at, for example, 5.8^* 1 and 6 App. I. 65.15, while an insertion of the Southern recension, 3.203^*, adds several names, including that of Kārtikeya, to those given in 3.11.17–18 (on which see below).

¹⁹ The Lokapālas are indeed found at 2.321^* 5–6 (where they are assigned to the compass points), 563^*, 1465^* 1, 2009^*. 4.252^* 11–14 [l.v.], and 6 App. I. 8.17 and 20. This list of occurrences in late passages is not exhaustive and in fact contains few more than the references taken from the Rāmāyaṇa by Hopkins in his treatment of the World-Protectors (Epic mythology, 149–52; copied with modifications by Gonda, “A note on Indra in Purānic literature”, 244). An even later concept is that of the diśāgajas, found nevertheless in late passages incorporated into the text at 3.13.26c and 5.35.67b.

²⁰ This process of inclusion of Viṣṇu is further illustrated by the number of mentions in ^* passages, of which the more important instances in the Ayodhyākāṇḍa follow. At 2.1.9cd some Northern manuscripts make Rāma an aṃśāvatāra of Viṣṇu, which the Southern manuscripts do in 10^* 1–2; the lack of agreement between the recensions is significant. 581^* 3 refers to Viṣṇu's three strides, 707^* 9 to his heaven, Goloka, and the compound viṣṇupādacyutam is found at 1061^* 11, all insertions of the Southern recension. A comparison of Rāma and his companions with Viṣṇu, Indra, and Śrī is introduced by various Northern manuscripts at 1307^*, while at 1731^* 4 Rāma has the valour of Indra and Upendra. The substitution of Viṣṇu for Brahmā at 102.2–3 and Gonda's comment on it have already been noted.

²¹ In 3.1253^* [l.v.], a tag stanza of the Southern recension, Rāma and Lakṣmaṇa are likened to Viṣṇu and Indra (surendrāv iva viṣṇuvāsavau) but, although the relative importance of Viṣṇu and Indra betrays a later date than the epic proper, the stage of identification has not been reached. At App. I. 10, another Southern insertion, Indra heads another list of gods in which Viṣṇu appears at line 10.

²² The Critical Notes to the Kiṣindhākāṇḍa have a n interesting comment on 16.25: “After 25, D5.6.8–10 S (i.e. all S MSS.) plus D3.11 add the star passage 343^*, which in 1.2 has a ring of treating Rāma as an avatāra. This tendency is not seen anywhere in the constituted text of this Kāṇḍa.” At 17.10d variant readings introduced Upendra, and Nārāyaṇa occurs at 1374^* 6.

²³ 5.93^* and 94^* are insertions of the Northern and Southern recensions respectively, likening Hanumān to Viṣṇu. At 610^* 2–3 Rāma is Puruṣottama and possessed of Viṣṇu's valour, while at 1048^* 11 Hanumān declares him t o be viṣṇutulyaparākrama, calling him lokatrayanāyaka at line 16. App. I. 13 likens the burning Laṅkā to a whole string of gods, among them Viṣṇu at line 29.

²⁴ Important references to Viṣṇu in ^* passages of the Yuddhakāṇḍa include: 1011^* (where Viṣṇu is linked with Garuḍa), 1311^*, 3114^* (a Southern insertion modifying the comparison of Rāma with Indra in the text at 6.99.10 by suggesting that Viṣṇu is a better comparison), 3703^*, and 3709^* (alternative conclusions and phalaśrutis to the kāṇḍa and so to the whole Rāmāyaṇa), App. I. 16.52, App. I. 32. 32–5 and 79–82, and App. I. 65.15.

²⁵ Ananta is also mentioned at 6 App. I. 3.342–3 [l.v.] and as Śeṣa at 3.253^* 1.

²⁶ Several avatāras are mentioned in 3.1191(A)^*, inserted by one manuscript D3, namely the boar, Nṛhari (= Narasiṃha), Rāma (= Paraśurāma), and Kalki. The term narasiṃha does occur in the text but simply as a n epithet for kṣatriyas, a variant for the commoner naravyāghra, puruṣavyāghra, and so forth.

²⁷ Such identification is of course made in later passages, cf. nn. 20–4.

²⁸ These two stages are strikingly similar to, though much briefer than, the successive stages of Kṛṣṇa's self-revelation in ch. 9–10 and 11 of the Bhagavadgitā. No doubt the same factors were at work in each case.

²⁹ Several Gandharvas and Apsarases are individually named at 2.85.14–15 and 41–4, which is one more indication of the lateness of that particular sarga; of the names there occurring few are found elsewhere, only Tumburu at 3.3.18c (also Mbh. 3.44.28), Viśvāvasu at 5.1.16cd, Ghṛtācī at 4.34.7a, and Hemā (named at 2.2013^* 1) at 4.50.14–17. Another list of Gandharvas is found at 4.40.41, again a later passage, as is 4.41, 17b, where Gandharvas are termed “terrible” (tarasvin). Other individual Apsarases named are Urvaśī, along with Purūravas, at 3.46.18d, Rambhā at 3.3.2Od, and Puñjikasthalā, born as Añjanā, at 4.65.8b.

³⁰ The divine physician Dhanvantari finds mention at 2.1083^* 6.

³¹ Several others are mentioned only in the Bālakāṇḍa or in ^* passages: Ahalyā, Ṛṣyaśrṅga, Jaimini, Vālmīki, and Śatānanda, for instance.

³² Specific allusion is made to this elsewhere, most fully in 6.103.14: nirjitā jivalokasya tapasā bhāvitātmanā / agastyena durārdharṣā muninā dakṣiṇeva dik.

³³ His other half, Ketu, occurs at 3.338^* 1, but apparently not elsewhere.

³⁴ The majority of these names do not occur elsewhere in the Rāmāyaṇa apart from this passage, 3.13, but it may be noted that 7–9 is similar to Mbh. 12.321.33–5 (21 Prajāpatis) and 17cd-32 to Mbh. 1.60.54–67.

³⁵ It is of interest to note that two of Krodhavaśā's daughters appear in the first sarga of the Sundarakāṇḍa, but, whereas Surasā is named at 3.13.22d, 28c, and 31d, Siṃhikā is named only at 3.246^* 1.

³⁶ cf. Kane, P. V., History of Dharmaśāstra II (Poona, 1941), 561Google Scholar. This proverb is followed immediately by another obviously late one: eṣṭavyā bahavaḥ putrā … api kaścid gayāṃ vrajet (2.99.13a+d, cf. Mbh. 3.82.85ah and 13.88.14ab).

³⁷ Much more elaborate in both syntax and detail is 3.1017^* 3–4: dīpanirvāṇagandhaṃ ca suhṛdvākyam arundhatīm / na jighrati mumūrṣur yo na śṛṇoti na paśyati.

³⁸ See Epic mythology, 67. In fact this sarga probably belongs to the second stage, since it contains also the catalogue of how Rāma spent an otherwise lost ten years and the episode of Vātāpi and Ilvala.

³⁹ The opening of the Ayodhyākāṇḍa has clearly been extensively worked over and approximately the first 30 sargas have been so altered or inflated as to bear no real relationship stylistically to the rest of the book. Both sarga 65, which describes Bharata's return to the melancholy Ayodhyā, and sarga 66, in which Kaikeyī eventually breaks the news of Daśaratha's death and Rāma's banishment, have clearly been greatly expanded.

⁴⁰ Greater detail is found in ^* passages: some Northern manuscripts bring in a comparison of a śūdra laying hands on the Vedic scriptures at 3.926^*; the Southern recension adds the gāyatri to a list at 3.203^*; one manuscript even includes the Śatarudríya at 2 App. I. 31.21; and further examples could be quoted.

⁴¹ This is no doubt part of the reason why both recensions make insertions at this point, among which may be noted the reference at 2.1210^* 4 (a Southern addition) to a bali to the Viśvedevas, Rudra, and Viṣṇu—an interesting grouping. Cf. also 3.14.21–3.

⁴² This point might seem obvious but Hopkins (Epic mythology, § 29, pp. 70–2) draws a distinction between the Mahābhārata and the Rāmāyaṇa on this point, suggesting that the latter does mention temples and other shrines. But the majority of the instances he cites have been relegated to ^* passages in the Critical edition. One exception is 2.6.4, where Rāma and Sītā sleep in Viṣṇu's holy āyatana. For a more recent discussion see Viennot, Odette, “Le culte de l'arbre dans l'Inde ancienne”, Annales du Musée Guimet, LIX, 1954, 88–89.Google Scholar

⁴⁸ Attempts to elucidate the significance of the last word have been made by Ingalls, D. H. H. (Mélanges d'Indianisme à la mémoire de Louis Renou, Paris, 1968, 393)Google Scholar and Kuiper, F. B. J. (Indo-Iranian Journal, IV, 1960, 259–261).Google Scholar

⁴⁴ Epic mythology, 71.