Pramāṇa · Search · About · API

The Development of Hindu Iconography c10: Chapter III: The Origin and Development of Image-Worship in India

The Development of Hindu Iconography c10

Language
English

Contents

Chapter III: The Origin and Development of Image-Worship in India

c10.2

It has been said in the preceding chapter that the later sections of the Vedic literature distinctly indicate remarkable changes being introduced in the religious outlook of the Indo-Aryans. It is true that they did not relinquish the practices which were performed with so much zest by their forefathers, but there cannot be the least doubt that all these were having more re-orientation due to various factors that were in operation from the very beginning of the period when they first set their feet on Indian soil. The most important among these was undoubtedly the close contact which they had to come in with the previous inhabitants of India. However much they could revile the children of the Indian soil whom they were driving from the more covetable lands into the hills and jungles, with such deprecatory epithets as *dāsas*, *anāsas* (noseless ones), *yātus* or *yātudhānas*, *rākṣasas*, *śiśnadevas*, *mūradevas*, etc., it cannot be denied that some of these latter people possessed a sort of material culture which was much superior to that of their victors. It is a pity that we have not before us any literary record of what these people were like, what they believed and practised, what they thought of their conquerors, presented from their point of view; but the remains that have been unearthed in course of systematic excavations in the Indus Valley have brought to light immense evidence regarding the high and developed state of material civilisation with which their forefathers were endowed. The commingling of cultures of the immigrants and the former inhabitants was greatly responsible for the gradual introduction of various elements which are either not traceable, or traceable only in faint outlines, in the earlier literary works of the Indo-Aryans. The *Ṛgveda*, or the other Vedas and the early Brāhmaṇas, had practically nothing to say on such topics as the law of *Karma*, the transmigration of souls and their necessary concomitant, the somewhat pessimistic view of life; but these were gradually being more and more discussed in the different Upaniṣads. The wholesale pessimism of the Buddhists might not have been the characteristic of the latter, but there is little doubt that the genius of the Upaniṣads wholly differs from that of the *Ṛgveda*, however many ties may bind the two periods. Again, the pantheism of the former can very well be contrasted with the belief in the multifarious nature-gods of the Aryans as portrayed in the latter. All these new elements can be presumed to have grown in the Indian soil, in the inception of which the earlier settlers in India did not play a mean part. Keith has very cautiously presented the problem in this way: "The Upaniṣads, as in some degree all earlier thought in India, represent the outcome of the reflections of a people whose blood was mixed. We may, if we desire, call the Upaniṣads the product of Aryo-Dravidian thought; but if we do so, we must remember that the effect of the intermixture must be regarded in the light of chemical fusion, in which both elements are transformed."¹

c10.3

¹ Keith, *Religion and Philosophy of the Veda and Upaniṣads*, Vol. II, p. 497.

c10.4.1

The one important element, however, which has got special bearing on our subject and the name of which is to be found in at least one of the major Upaniṣads, is *Bhakti*, primarily the loving adoration of some persons by others, but secondarily the deep affectionate and mystic devotion for some personal deity who is the object of worship (in the developed sense of the term, *i.e.*, *pūjā*). If we briefly trace the history of the gradual emergence of *Bhakti* in the religious lives of the Indo-Aryans, we cannot but endorse the view just quoted.

c10.4.2

Among the several constituent factors which make up this element in its secondary aspect, the most important ones are 'belief in one personal god as spiritual being, the faith that his power is sufficient to secure that at the last the good will conquer, and lastly a conception of the nexus that binds together God and his worshippers as mainly moral."¹ In the later stratum of the Ṛgveda, we find the struggling appearance of one supreme entity into which all the separately conceived Vedic divinities are merged.² Some faint traces of the belief in one moral god who looks after the consciences and works of men are certainly present in some of the Ṛgvedic characterisations of Varuṇa to whom prayers for forgiveness are offered by the hymnists.³ Keith has observed, "The thought of India started from a religion which had in Varuṇa a god of decidedly moral character and the simple worship of that deity with its consciousness of sin and trust in the divine forgiveness is doubtless one of the first roots of Bhakti."⁴ But this kind of worship dedicated to such a god was arrested in its growth, and the prominence given to the other gods like Indra, Agni, Soma, and others, intimately associated with sacrifice, adversely affected it. Even then in one of the late hymns of the Ṛgveda (X.

c10.4.3

125), the goddess Vāc is made to say, "I give wealth unto him who gives sacrifice;...I am that through which one eats, breathes, sees and hears;...him that I love, I make strong, to be a priest, a seer, a sage." Eliot remarks about this passage, "This reads like an ancient preliminary study for the Bhagavadgītā. Like Kṛṣṇa, the deity claims to be in all and like him to reward her votaries."⁵ In the Upaniṣads, on the other hand, the mental attitude of the thinkers to the one supreme entity, viz., Brahman-Ātman, gets a character which is, in no very uncertain manner, reminiscent of Bhakti. The growth and development of monotheism, a direct result of the pantheistic conception of the earlier Upaniṣads, was the certain background on which Bhakti was to develop among the intellectual section of the composite population of India. The impersonal-personal Brahman was no doubt ill-suited to play the rôle of the one god of devotion and the strictly monistic character of some of the earlier Upaniṣadic passages was logically inimical to the ideas of loving faith; still there are many passages in some of them, which are significant. We are told in one of them that ‘Ātman cannot be gained by the instruction, nor by understanding nor by much learning; he whom Ātman chooses, by him the Ātman can be gained;

c10.4.4

to him the Ātman reveals its own essence (or person).’ Here, even though the idea of faith or love is not distinctly present, yet the positive assertion that Ātman selects its own and it cannot be gained by proficiency in the Vedic lore and other things does forcibly remind us of the free grace of the personal god.¹ This again seems to be clear in the Kaṭhaka passage (II. 20) which speaks of the ability of a person to see the glory of the Ātman if he is graced by the creator, if the word in this verse is taken to mean ‘by the grace of the creator’ *dhātuḥ prasādāt* and not ‘through the tranquillity of the senses’ (*dhātu-prasādāt*) as Śaṅkara explains it. The Śvetāśvatara Upaniṣad (III. 20) which contains much that is theistic in nature contains the same passage with an alteration which, though slight, is material.² It is in this Upaniṣad among the major ones, that we find for the first time the mention of the word Bhakti which occurs in the last verse of the work.³ From this time onward references to it become clearer and clearer, and Pāṇini in the several sūtras of his Aṣṭādhyāyī lays down rules for various word-formations in which the etymological sense of the word Bhakti, viz., ‘resorting to and then loving the thing resorted to with faith and devotion’ is the central idea.¹

c10.5

¹ N. Macnicol, Indian Theism, p. 7.

c10.6

² Cf., R. V., I. 164. 46: *Indram Mitram Varuṇamagnimāhuratho divyaḥ sa suparṇo garutmān | Ekam sadviprā bahudhā vadantyagnim yamam mātariśvāna-māhuḥ*

c10.7

³ Cf., R. V., I. 25, 1 and 2, and similar other verses.

c10.8

⁴ J.R.A.S., 1915, p. 834.

c10.9

⁵ Eliot, Hinduism and Buddhism, Vol. II, p. 181. He says further, "It is true that the 'Come unto me' (*māmekam śaraṇam vraja*) is not distinctly expressed, but it is surely struggling for expression."

c10.10

¹ Indra says to Pratardana who had asked him for a boon, “Know me only; that is what I deem most beneficial to man, that he should know me. . . . He who meditates on me as life and immortality gains his full life in this world and in heaven immortality.” Eliot remarks about this passage, ‘Though the relation of the devotee to the deity here is purely intellectual and not emotional,’ still the idea, that intellectual devotion directed to a particular deity will be rewarded, is clearly present’: Eliot, op. cit., p. 181. But he forgets that Indra here symbolises the highest principle discussed in the early Upaniṣads.

c10.11

² The last *caraṇa* of this verse, viz., *dhātu-prasādānmahimānamātmanaḥ*, is changed into *dhātuḥ prasādānmahimānamīśam*.

c10.12

³ *Śvet. Up.*, VI, 23—*Yasya deve parā bhaktiryathā deve tathā gurau | Tasyaite kathitā hyarthāḥ prakāśante mahātmanaḥ* ||

c10.13

¹ IV, 95 ff.

c10.14.1

The Upaniṣadic Brahman-Ātman, when conceived in the personal aspect, especially in the theistic Upaniṣads, is usually called not Deva (god), but Īśa, Īśāna, Īśvara, and latterly Parameśvara. But even then, Śvetāśvatara found it necessary to refer to some divine personal entity like Rudra (also mentioned under other names such as Eka deva, Mahān deva, Maheśvara, Māyī and once even Śiva : ‘jñatvā śivam sarvabhūteṣu gūḍham’), who was the recipient of the homage of his devotees. In this work which has not cut itself asunder from the general body of the scheme of the early Upaniṣads (‘beneath the characters of theism are discerned, half obliterated, those of pantheism and under the latter, again, those of idealism,’—Deussen), we are told that the knowledge alone of this one god will break up the fetters of death, and nothing will be gained by him by the learning of the Ṛk verses, who does not know him (Yastanna veda kimṛcā kariṣyati). But evidently such a mental attitude of the thinkers, though no doubt it bespeaks a great deal of progress towards the development of cult religions and sectarianism, was not at all truly sectarian in character. Its natural corollary, however, was the growth of the latter in which the element of Bhakti was the main guiding principle. The gods, round whom these cults developed, were not recruited from the orthodox Vedic Pantheon, but from quite a different source.

c10.14.2

Indra, Prajāpati, Mitra, Varuṇa, Yama, Agni and others could never effectively serve in the rôle of cult-deities, though some attempts were possibly made by those of the Vedic way of thinking to foist one or other of them as rivals to the more important cult-gods. But these, if they were ever seriously made, were destined to failure, and in the developed sectarianism of the Epic and Purāṇic periods we find several of the more important Vedic deities such as Indra, Varuṇa, Agni, Vāyu, Yama and one of the less important ones like Nirṛti relegated to the comparatively insignificant position of the guardians of quarters (Dikpālas), where the highest purpose they could serve was of a mere accessory character. Some of the Vedic gods, again, like Viṣṇu, Rudra and Sūrya came to be merged in the composite cult-deities at a subsequent period, and this merger was so complete and so important for the cults themselves that some of the latter came to be designated, optionally at first, but more constantly at a later period, by the names of the Vedic counterparts of their cult-pictures (cf. the part played by Viṣṇu in the Bhāgavata or Pāñcarātra cult which came to be described as Vaiṣṇava at a later date). But the originals of these gods were actual human heroes like Vāsudeva Kṛṣṇa, the son of Devakī (cf. Kṛṣṇa Devakīputra of the Chāndogya Upaniṣad, III.

c10.14.3

17), Śākyamuni Gotama and Mahāvīra, or mythological beings like Śiva (Rudra-Śiva), the Yakṣas Maṇibhadra, Pūrṇabhadra and others, and the goddess Umā-Durgā-Pārvatī-Vindhyavāsinī.¹ Pāṇini in his sūtra, Vāsudevārjunābhyāṃ vuñ (IV. 3. 98) most probably refers to two sectaries who were the exclusive worshippers of the apotheosised human heroes Vāsudeva and Arjuna, of whom the former was the more honoured and more important. Patañjali’s commentary on this sūtra fully endorses the view; but what is also very interesting is that Patañjali refers to a sect called the Śivabhāgavatas or devotees of Śiva, the Holy One, who carried in their hands an iron lance as an emblem of Śiva whom they worshipped.² The early Buddhist works on many occasions refer to the various kinds of worship that prevailed in India, especially in Central and Eastern India, at a time when Buddha preached his doctrine. R. G.

c10.14.4

Bhandarkar quotes a very interesting passage from the Niddesa, which furnishes us with a curious record of the various religious systems and superstitions that prevailed at the period: ‘The deity of the lay followers of the Ājīvakas is the Ājīvakas, of those of the Nighaṇṭhas is the Nighaṇṭhas, of those of the Jaṭilas is the Jaṭilas, of those of the Paribbājakas is the Paribbājakas, of those of the Avaruddhakas is the Avaruddhakas, and the deity of those who are devoted to an elephant, a horse, a cow, a dog, a crow, Vāsudeva, Baladeva, Puṇṇabhadda, Maṇibhadda, Aggi, Nāgas, Supannas, Yakkhas, Asuras, Gandhabbas, Mahārājas, Canda, Suriya, Inda, Brahmā, Deva, Diśā is the elephant, the horse, the cow, the dog, the crow, Vāsudeva, Baladeva, Puṇṇabhadda, Maṇibhadda, etc., respectively.¹ It will be wrong to suppose that this curious assortment of worshippers of particular objects indicates all of them as separate sectaries; what is worth noting, however, is that here is an authentic presentation of a medley in which the sects of Vāsudeva, the Ājīvakas and the Nirgranthas are mixed up with the believers not only in the Vedic gods like Indra, Agni, Candra, Sūrya and others, or with those putting their faith in the efficacy of austerities and asceticism (cf. the Paribbājakas and the Jaṭilas), but also with the superstitious animists.

c10.14.5

The last group, however much they might be deprecated by the cultured intellectuals of the day, played no mean a part in moulding the beliefs and practices of their more advanced contemporaries. Megasthenes, as quoted by Arrian, mentions that Herakles was the special object of worship of the Sourasenoi, an Indian tribe in whose land were the great cities of Methora and Kleisobora (Mathurā and Kṛṣṇapura), and through which flowed the river Iobares (Yamunā); this is a confirmation from a foreign source regarding the existence of at least one sectary among the several named above in the fourth century B.C. in the Yamunā region² We shall see later on that archaeological data from the 2nd century B.C. onwards substantially corroborate the above facts.

c10.15

¹ Elaborate use was made of the descriptions of the Vedic counterparts of the syncretic gods, thus fully substantiating the hypothesis already referred to regarding the composite culture of the post-Vedic period.

c10.16

² Mahābhāṣya, under Pāṇini, V. 2. 76.

c10.17

¹ R. G. Bhandarkar, *Vaiṣṇavism, Śaivism and Minor Religious Systems*, p. 3.

c10.18

² McCrindle, *Ancient India as described by Megasthenes and Arrian*, p. 201. R. G. Bhandarkar was the first to identify the tribe of the Sourasenoi with the Sātvatas, and Herakles with Vāsudeva. The Greek writers appositely designated Vāsudeva Kṛṣṇa as Herakles, for both these deities were very probably apotheosised human beings.

c10.19

A somewhat elaborate discussion about the origin and growth of the idea of *Bhakti* has been found necessary because the solution of the whole problem of the origin of image-worship itself principally depends on it. Some sensible objects were found to be indispensable by the various sectaries who required them as so many visible symbols for the various personal gods to whom they rendered their exclusive homage. The symbols and images in their case analogically did the same sort of service as was done by Fire (Agni) in Vedic ritualism. Fire was specially sacred to the Vedic priests, because it was the carrier of the sacrificers' oblations to their respective gods; in the case of a sectary, the image or icon or any such visible symbol of his deity was the handy medium through which he could transfer his one-souled devotion (*ekātmikā bhakti*) to his god. That was the primary purpose for which they were usually intended, though there is textual evidence regarding their being used secondarily for such purposes as *abhicāra*, etc. (*cf.* the *ābhicārika mūrtis* as described in the *Vaikhānāsāgama*). The rendering of one's homage was done by various acts of *pūjā* in which images were absolutely necessary; these were *abhigamana* or going to the temple of the deity with the speech, the body and the mind centred on him, *upādāna* or collecting the materials of worship such as flowers, incense, sandal paste, offerings (*naivedya*), etc., *ijyā* or the very act of worshipping the *Śrī Vigraha* (the auspicious body of the lord), *svādhyāya* or the muttering of the mantra usual to particular cult-divinities, and lastly, *yoga* or meditation.¹ The last constituent of the act of *pūjā* has got special bearing on the history of the evolution of the icons. One author tells us that the image-maker should fashion images in such a manner that they would conduce to the success of the *dhyāna-yoga*.² Many images are

c10.20

¹ Some of the Mantras special to particular deities are (1) the twelve-syllabled Bhāgavata mantra : *Oṃ namo bhagavate Vāsudevāya*, (2) the five-syllabled Śaiva one : *Namaḥ Śivāya* and (3) the seven-syllabled Śakti mantra : *Parameśvari svāhā*.

c10.21

² *Śukranītisāra*, Ch. IV., Section 4, 147 :

c10.22

*Dhyānayogasya saṃsiddhyai pratimālaKṣaṇaṃ smṛtam |*
*Pratimākārako marttyo yathādhyānarato bhavet ||*

c10.23

known where the deity himself is shown in the pose of a Yogī immersed in deep meditation (*cf.* the images of Jina, Buddha, Yogāsana Viṣṇu, Yogadakṣiṇāmūrti of Śiva and others). A notice of a very interesting passage in the *Mahābhārata* which refers to Nārada’s visit to the Badarikāśrama to see Nara and Nārāyaṇa will not be out of place here. Nārada finds the latter engaged in the act of worshipping; bewildered at this (because Nārāyaṇa was himself an object of worship), Nārada asks him about the latter’s object of devotion. Then the Lord tells him that he is worshipping his original Prakṛti, the source of all that is and that is to be.¹ Here we have a textual evidence in support of deities themselves being conceived in the *dhyāna-yoga*, and their images depicted in this very pose had the practical utility of aiding the devotee to concentrate the mind on his god. The importance of such images as well as their connection with those that were discovered in the Indus Valley has been elaborately discussed by R. P. Chanda in some of his writings.² The true significance and purpose of the image proper of the god must be understood in this light, and this is fully emphasised by the passages appearing in such late works as *Rāmapūrvatāpanīya* and *Jābāladarśana* Upaniṣads and *Mahānirvāṇa Tantra*, even though some of them deprecate the practice of the persons who offer their *bhakti* to their gods through these media.¹ But these works are mainly written from the point of view of those who firmly believed in worshipping the highest principle without the aid of any media (*nirākāropāsanā*), and the attitude of some of them was strictly non-dualist (*Śivamātmani paśyanti*).

c10.24

¹ *Mahābhārata*, Vaṅgavāsī Edition, Śānti Parva, Nārāyaṇīya Parvādhyāya, ch. 334, verses, 14-45. This passage is a curious amalgam of the Sāṅkhya and the Yoga. The entity who is the object of Nara’s and Nārāyaṇa’s devotion is described thus: *Yattat sūkṣmamavijñeyamavyaktamacalam dhruvam | Indriyairindriyārthaiśca sarvabhūtaiśca varjitam || Sa hyantarātmā bhūtānām kṣetrajñaśceti kathyate | Triguṇavyatirikto vai puruṣaśceti kalpitaḥ || Tasmādavyaktamutpannam triguṇam dvijasattama | Avyaktā vyaktabhāvasthā yā sā prakṛtiravyayā || Tām yonimāvayorviddhi yo’sau sadasadātmakah | Ābābhyāṃ pūjyate so’hi daive pitrye ca kalpate ||* This original Prakṛti, we are told further on, was none other than Hari.

c10.25

² This is ably recounted in one of his latest works, viz., *Mediæval Indian Sculptures in the British Museum*, Ch. I, pp. 6-10. He suggests that the ‘sudden rise of the cult of the images of the Yogī in north-western India (Gandhāra and Mathura) is only a revival of an old cult of the image of the Yogī once prevalent in that region.’

c10.26

¹ *Ciṇmayasyādvitīyasya niṣkalasyāśarīriṇaḥ* | *Upāsakānām kāryārtham brahmaṇo rūpakalpanā* (Rāmapūrvatāpanīya Upaniṣad); *Śivamātmani paśyanti pratimāsu na yoginaḥ* | *Ajñānām bhāvanārthāya pratimā parikalpitā* (Jābāladarśana Upaniṣad); *Evam guṇānusāreṇa rūpāṇi vividhāni ca* | *Kalpitāni hitārthāya bhaktānāmalpamedhasām* (Mahānirvāṇa Tantra). The last-named work thus derides the efforts of those who want to attain salvation through this method: ‘*Manasā kalpitā mūrti nṛṇām cenmokṣasādhanī* | *Svapnatabdhena rājyena rājyena mānavās-tathā* || *Mṛcchilādhātudārvādi-mūrtīśvarabuddhayah* | *Kliśyantastapasā jñānam vinā mokṣam na yānti te* ||

c10.27

It will be profitable to compare this view-point about the usefulness of the images with that presented in the works of the Bhāgavatas or the Pāñcarātras. We have already mentioned the significance of such words as *vigraha*, *bera*, *tanu*, *rūpa*, etc.; these are mostly utilised in such literature replete with sentiments of deep loving faith for the lord Vāsudeva and his principal aspects. The manner of describing euphemistically the images after due consecration as the very bodies or forms of the god is fully emphasised therein by the prescription that the cult-picture of the deity was one of his five-fold forms; they are *Para*, the highest form, *Vyūha*, the emanatory forms, *Vibhava*, the incarnatory forms, *Antaryāmin*, the lord as immanent in the universe and as the inner controller of the individual, and lastly the *Arcā*, the duly consecrated images. This concept of the image is based on its unique sublimation to the very position of the godhead, the object of deep loving adoration to the devotee. The process presupposes a mental preparation, a studied effort on the part of the worshipper, which culminates in the attainment of that frame of mind in which an object fashioned by human hands reaches such an august level. A concept similar to the above is essentially one of the characteristic features of most of the religious cults of India in which the *Bhakti* element was the main guiding principle. The Ālvārs or the Nāyanmārs in the south, the Vaiṣṇava or the Śaiva saints of the north and the Ācāryas of many of the sectarian religious systems of the early and mediæval periods throughout India were highly cultured people, but in their approach to the deity the divine image played a very important part. So, T. A. G. Rao's observation, " the Hindu śāstras prescribe image worship to weak unevolved persons in particular ", will have to be modified. It is true that the root idea of image-worship can be traced to animism, but so also can the idea of the immanence of the godhead be traced, yet in its rationalised and developed form there is very little place for crudity or savagery. It has therefore been truly remarked that " in dealing with savage ideas of the inanimate, it must be kept in mind that non-living things are worshipped or feared not in any symbolical sense, which is altogether foreign to the lower intelligence, but as the supposed home of a spirit, or as in some sense a vehicle of power."' This symbolism is further expressed and emphasised by the usual practices of endowing the mediæval Indian images with many hands, which has been dubbed as a monstrosity by some scholars. Different explanations have been suggested by different scholars with regard to this feature. Macdonell, for example, suggested that it was the direct outcome of the iconographers' necessity to distinguish the image of one deity from the other, when the earlier mode of doing so by the placing of mounts below them was found inadequate due to the gradual increase of the pantheon. He wanted to substantiate his view by referring to one universal feature of the multi-armed images: their natural hands were invariably to be found in such 'action-poses' as *abhaya*, *varada*, etc., whereas the added hands carried different implements which were, according to him, nothing but differentiating marks.² But this statement is difficult to accept. The alternative suggestion, that the hands and the *āyudhas* or implements in them portray attempts to symbolise, however ineffectively, the multifarious activities of the god, is more acceptable. T. A. G. Rao says, ‘The images of the Hindu gods and goddesses are representations of the various conceptions of divine attributes. Sculpturally it may be said, the number of hands in an image represents the number of attributes belonging to the deity, and their nature is denoted by the āyudha held in the hand or by the pose maintained by it.’¹

c10.28

¹ Edward Clodd, *Animism*, p. 78. *Italics are mine.*

c10.29

² J.R.A.S., 1916, pp. 127-8. Refer to A. M. Hocart's article on 'Many-armed Gods' in *Acta Orientalia*, Vol. VII, 1929, pp. 91 ff. Hocart remarks, 'Evidently theological considerations were paramount in deciding the number of arms, and this is far more in accord with what we know of the Indian mind than Prof. Macdonell's theory.'

c10.30

¹ T. A. G. Rao, *op. cit.*, Vol. I, Part I, Introduction, p. 27. The weapons or attributes, in the case of some at least of the Brahmanical images, have also their bases in the anthropomorphic descriptions of their Vedic counterparts.

c10.31.1

A well-executed image, if it follows the rules of proportions laid down in the Śilpaśāstras and is pleasing to the eye, invites the deity to reside in it and is particularly auspicious to its worshipper.² But deities were not always iconically represented; over and above their concrete representations in anthropomorphic and, rarely, theriomorphic forms, they could also be figured in an aniconic manner. The latter mode is undoubtedly reminiscent of an earlier practice. In India, iconism and aniconism existed side by side from a very early period, and this feature is also present even in modern times. Buddha could be represented by means of such symbols as the Bodhi tree with Vajrāsana beneath it, his foot prints, the stūpa, etc., which were directly associated with him; in the Amaravati and Nagarjunikonda sculptures of the 2nd and 3rd centuries A.D., we find him being depicted iconically and aniconically at the same time, though in the earlier Buddhist art of Central India he used to be represented only in the latter manner. Similarly, Brahmanical cult deities could as well be worshipped in the Śālagrāmas, the Bāṇa-liṅgas and the Yantras, as in images; but here, however, their association with the symbols was not so direct. Attempts were not wanting to account for this connection by the fabrication of mythological stories.¹ The Śālagrāmas, Bāṇa-liṅgas and Yantras are primarily associated with the Vaiṣṇava, Śaiva and Śākta cults respectively.

c10.31.2

Then, there are various sacred stones scattered over different parts of India, which are taken to stand for one or other of the cult divinities. It has been shown that rude stone monuments consisting of menhirs, dolmens, cairns, and cromlechs distributed over parts of Europe, Western Asia and India are essentially sepulchral in character.² The Indian phalli, especially their early specimens, portray this feature to a very great extent. Many instances are known, in India of ancient and modern times, of stones being regarded as aniconic representations of the cult-deities. The well-known Śākta tradition about the severed limbs of Satī falling in different parts of India and about the latter being regarded as so many pīṭhasthānas, particularly sacred to the Śakti-worshippers, should be noted in this connection. In modern times, the most important objects of worship in many of these shrines are usually stone blocks covered over with red cloth, which are described as this or that limb of the goddess. It is interesting to observe here that Hiuen Thsang records in his *Si-yu-ki* some useful details about a great mountain in ancient Gandhāra, ‘which had a likeness (or image) of Maheśvara’s spouse Bhīmā-devī of dark-blue stone. According to local accounts this was a natural image of the goddess; it was a great resort of devotees from all parts of India.

c10.31.3

At the foot of the mountain was a temple of Maheśvara-deva in which the ash-smearing Tīrthikas performed much worship.’ Watters remarks, ‘The image or likeness of Bhīmā-devī here mentioned was apparently a dark-blue rock in the mountain supposed to have resemblance to that Goddess.’³ Watters’ observation about the resemblance is immaterial; but, what is of importance here is that we find in it an authentic reference to a svayambhūmūrti of the goddess in the 7th century A.D. Now, these objects of worship are principally aniconic stones, and numerous textual references to the self-wrought phalli (Svayambhū lingas) have been quoted by Gopinath Rao in his work (section on Liṅgas). It seems that sometimes these aniconic objects were held in more veneration than the images fashioned by human hands, for the list supplied by Rao proves that claims were put forward on behalf of man-made Śiva-liṅgas to be regarded as Svayambhū ones. Then there are sacred trees and other objects which were also held in high respect on account of their association with certain spiritual entities, and in the subsequent religious history of India, these were specially associated with one or other of the cult divinities. Reference may be made to the high esteem in which the bael and tulasi trees were held by the devotees of Śiva and Viṣṇu respectively, and also to the sthala-vṛkṣas associated with particular shrines;

c10.31.4

numismatic data, as I shall show afterwards, seem to prove that more or less similar was the case in much earlier times. The association of the Aśvattha (Ficus religiosa) with Śākyamuni Buddha and that of the various other trees like Puṇḍarīka, Śirīṣa, Pāṭali, Nyagrodha and others with his predecessors were not the special feature of the Buddhist creed alone; these Bodhi trees were the direct descendants of the Caitya Vṛkṣas (rukkhacetiyāni) of more primitive times. The trees and branches appearing so frequently in the numerous seals discovered in the Indus Valley had most probably some cult significance of this character.

c10.32

² *Abhirūpyācca bimbānām devaḥ sännidhyamṛcchati* (Hayaśīrṣa Pañcarātra); *Sukranītisāra*, IV. Sec. 4 : 152-3 :—*Yathoktāvayavaiḥ pūrṇā puṇyadā sumanoharā* | *Anyathāyurdhanaharā nityam duḥkhavivardhinī* ||

c10.33

¹ *Cf.* T. A. G. Rao, *op. cit.*, Vol. I, Part I, Introduction, pp. 8 ff. Rao gives us an elaborate account of these aniconic symbols.

c10.34

² Grant Allen, *The Evolution of the Idea of God*, pp. 68 ff.

c10.35

³ Watters, ‘On Yuan Chwang,’ Vol. I, pp. 221-22. The Bhīmā-devī shrine is evidently identical with the Bhīmasthāna beyond Pañcanada mentioned in the *Mahābhārata*, Vanaparva, Ch. 82, verses 84-85, and probably also with Bhiṣaṇa of the *Mahāmāyūrī* text. According to the *Mahābhārata*, there was a Yoṇitīrtha there, a dip into whose kuṇḍa was regarded as highly auspicious in character. These details are important for the religious history of India; the Yoṇipīṭha is now at Kāmākhyā, near Gauhati, Assam, and the particular Bhairava of the Devī is Umānanda on a rock in the midst of the Brahmaputra near by. In the 7th century A.D. there was a similar shrine in the heart of Gandhāra with the adjacent shrine of the Bhairava (Śiva). For detailed discussions about these, cf. my article in Indian Historical Quarterly, Vol. XIV, 1938, pp. 751-3.

c10.36

It is time now to discuss some of the literary data with regard to the prevalence of images in the post-Vedic period.

c10.37.1

Incidental reference has already been made to one or two among them in the first few pages of the second chapter of this work. But, a collected presentation of some of those as well as several other data will be necessary for the better understanding of the topic under discussion here. Pāṇini's sūtra, *jīvikārthe cāpaṇye* (V. 3. 99) as explained by the later commentators is interesting; it gives us positive information about the concrete representations of deities in the 5th century B.C.¹ But from this cryptic sūtra, we can have no idea about the kind of deities whose *pratikṛtis* were made means of livelihood by a certain class of people. It can justifiably be presumed, however, that these were not the orthodox Vedic gods, but were popular objects of worship like the Yakṣas and the Nāgas; they could also be even of Vāsudeva, Arjuna and the Mahārājas (Kubera, Dhṛtarāṣṭra, Viḍūdhaka and Virūpākṣa, the guardian deities of the Northern, Eastern, Southern and Western quarters respectively—this is A. K. Coomaraswamy's interpretation, and it seems to be the correct one), because Pāṇini under IV. 3. 95 (*yeṣāṃ bhaktir yap*) lays down rules for the formation of words denoting the *bhaktas* or the worshippers of Vāsudeva, Arjuna and the Mahārājas (IV. 3. 98: *Vāsudevārjunābhyāṃ vūñ* and IV. 3. 99: *Mahārājātthañ*). But Patañjali is much more informative on this matter in his comment on Pāṇini's sūtra (V. 3. 99). His *bhāṣya* reads:—*apaṇya ityucyate tatredaṃ na sidhyati |

c10.37.2

Śivaḥ Skandaḥ Viśākha iti | kiṃ kāraṇam | Mauryairhiraṇyārthibhirarcāḥ prakalpitāḥ | bhavet tāsu na syāt | yāstvetāḥ sampratipūjārthāstāsu bhaviṣyati*. This passage is highly important, because it throws a flood of light on our problem. He mentions a few of the gods, viz., Śiva, Skanda, Viśākha whose images were being made for worship at his time (*sampratipūjārthā*); again, his assertion that the Mauryas devised the expedient of replenishing their royal coffer by the selling of images shows that images were in great demand among their subjects; lastly, it is significant that none of the three gods mentioned above can be described as Vedic in character. Such texts as the Arthaśāstra and the Manusaṃhitā also supply us some valuable data about the subject. Kauṭilya, in his chapter on Dūrganiveśa (Buildings within the Fort) says, ‘In the centre of the city, the apartments of gods, such as Aparājita, Apratihata, Jayanta, Vaijayanta, Śiva, Vaiśravaṇa, Aśvina, and the abode of Goddess Madirā shall be situated. In the corners, the guardian deities of the ground shall be appropriately set up.’¹ These are evident allusions to the shrines of the above-named gods, and it is presumable that the images of the latter were enshrined in them; an analysis of the names shows that only one among them (or possibly two, if Vaijayanta be taken to be a synonym of Indra), viz., Aśvi (the twin gods Aśvins) is distinctively Vedic in character.

c10.37.3

Vaiśravaṇa is the same as Kubera, the lord of the Yakṣas, Jayanta is most probably the same as mentioned in the Āpastamba Gṛhyasūtra (VII. 20. 3, Jayanta in this passage should not have been translated as ‘the conqueror’ as it has been done by Max Müller in the S. B. E. series) already noted in the second chapter; the image of Śiva is referred to in the Mahābhāṣya, here, and probably also in the Āpastamba Gṛhyasūtra noted above (Īśāna’s image is mentioned there and Īśāna is the name of one of the aspects of Śiva); the goddess Madirā may be the same as Mīḍhuṣī mentioned in the latter work in the same context and translated by Max Müller as ‘the bountiful one.’² Kauṭilya also refers to the figures of the “goddesses and altars which were to be carved on wooden door frames of the royal underground chamber” (*Vāsagṛhaṃ bhūmigṛhaṃ vā āsannakāṣṭhacaityadevatāvidhānadvāraṃ*, etc. : Kauṭilya on *Niśāntapraṇidhiḥ*); these figures had most probably protective utility and acted as sorts of charms, and were not meant for worship. In the chapter on *Apasarpapraṇidhiḥ* Kauṭilya refers to the images and flags of the gods (*devadhvajapratimābhirvā*) in the guise of which weapons will be supplied by the spies outside to the spies inside the enemy’s fort; in the same section we are told about the procession of gods (*i.e.*, the images of them—*daivatapretakāryotsavasamājeṣu*), etc., which would be taken advantage of by the spies in harming the enemy.

c10.37.4

Very great importance is assigned to the images of the gods in *Manusmṛti* and there are various passages in the work which lay down that *daivatam* (images of gods) are to be circumambulated (IV, 39), that one should not voluntarily step over the shadow of the gods (IV, 130), that at the *parvans* one should go to the images for protection (IV, 153), and that he who destroys a bridge, the flag of a temple (really the votive column in front of it), a pole (really a pillar) or images (*saṃkramadhvajayaṣṭīnāṃ pratimānāṃ ca bhedakaḥ*) shall repair the whole (damages) and pay 500 (*paṇas*) as fine (IX, 285). Manu gives us another interesting information that though images were highly venerated by the people in general, temple-priests, whose duty was to minister to these icons, were greatly deprecated, and they were placed in the same class with the Brahmans who earned their livelihood by medical practice, selling of meat and trading (*Cikitsakān devalakān māṃsa-vikrayiṇastathā | Vipaṇena ca jīvanto varjyāḥ syurhavya-kavyayoḥ*; III, 152). The same social stigma attaches to the temple-priests in modern times also; it can be explained by the suggestion that it was so because

c10.38

¹ Cf. H. C. Raychaudhuri's discussion about Pāṇini's date in his *Materials for the Study of the Early History of the Vaiṣṇava Sect*, Second Edition, pp. 24-30.

c10.39

¹ Arthaśāstra, translation by R. Shamasastry, 2nd Edition, p. 59. The translator notes that ‘the worship of the Aśvins and Vaiśravaṇa seems to have been prevalent at the time of this work.’ The original text reads :—Aparājita-pratihatajayantavaijayantakoṣṭhakān Śivavaiśravaṇāśvi śrimadirāgṛhaṃ ca puramadhye kārayet | Koṣṭhakālayeṣu yathoddeśaṃ vāstudevatāḥ sthāpayet ||

c10.40

² In the Ap. Gr. S., these three deities, viz., Īśāna, Mīḍhuṣī and Jayanta are mentioned together; Haradatta explained them as images of the three gods. Hiraṇyakeśin, Gr. Sūtra (II. 3. 8), in connection with the Śūlagava sacrifice meant for Rudra for averting cattle diseases, furnishes us with the interesting fact that the cow (the consort) of the spit-ox (i.e., the ox to be symbolically sacrificed) and her calf are euphemistically described as Mīḍhuṣī and Jayanta respectively; the

c10.41

sacrificer then prays to the three gods, viz., Iśāna, Miḍhuṣī and Jayanta, to touch the three beasts. If Miḍhuṣī and Madirā be identical, then both of them are to be regarded as the consorts of Śiva; one of the names of Śiva is Miḍhuṣa and Madirā is one of the synonyms of Durgā-Ambikā.

c10.42

these people prostituted their bhakti by making it a means of their livelihood. This is supported by a verse in the Nārada Pañcarātra (Bhāradvāja Saṃhitā, IV, 29) which says that one should never make the images of gods the means of their livelihood (na cāpyarcopajīvikāḥ). The whole verse reads :—Na ca mantropajīvā syānnacāpyarcopajīvikah | Nāniveditabhogaśca na ca nindyanivedakaḥ. The two texts, viz., Arthaśāstra and Manusmṛti, thus furnish us with some important data regarding the prevalence of image-worship in India of the 1st and 2nd centuries A.D., if not of an earlier period. The Mahābhārata, in like manner, refers often to the images of the gods, especially in connection with various Tīrthas (sacred places). There was an image of Viṣṇu named Śālagrāma in the Puṇḍarīka tīrtha (Śālagrāma iti khyāto Viṣṇuradbhutakarmakaḥ ; III, 84, 124) ; in the Jyeṣṭhila tīrtha were the images of Viśveśvara and his consort (Tatra Viśveśvaraṃ dṛṣṭvā devyā saha mahādyutim | Mitrāvaruṇayorlokānāpnoti puruṣarṣabha ; III, 84, 135) ; these, however, might have been aniconic—the former, a Śālagrāma, and the latter, the phallic emblem of Śiva, in which Śiva and Umā are symbolically represented. Reference to the image of Nandīśvara is to be found in XIII, 25, 21 (Nandīśvarasya mūrtiṃ tu dṛṣṭvā mucyeta kilviṣaiḥ) ; in the Mataṅgāśrama near Dharmaprastha was an image of Dharma, touching whom one would attain spiritual rewards, equivalent to those of an aśvamedha-sacrifice (Dharmam tatrābhisamspṛśya vājīmedhamavāpnuyāt ; III, 84, 102) ; an image of Brahmā is probably referred to in III, 84, 103-4 (Tato gaccheta rājendra Brahmasthānamanuttamam | Tatrābhigamya rājendra Brahmāṇaṃ puruṣarṣabha | Rājasūyāśvamedhābhyāṃ phalaṃ vindati mānavaḥ). Numerous such instances can be quoted from other sections also of the epic literature, but what is of special significance, in this connection, is that the results to be attained by a pious person visiting these tīrthas or worshipping the images therein are often estimated in terms of the fruits attainable by the performance of such Vedic sacrifices as Agniṣṭoma, Jyotiṣṭoma, Aśvamedha, Rājasūya, etc. A careful search among the early literature of the Buddhists and Jainas,—sectaries heterodox from the point of view of a Vedic initiate, throws much light on the form of worship prevalent in this period—in which both iconic as well as aniconic symbols played a great part.

c10.43

It is interesting to note that Quintus Curtius records that an image of ‘Hercules’ was carried in front of the army of Porus as he advanced against Alexander. Coomaraswamy thinks that this may have been an image of Śiva or of a Yakṣa.¹ The Greek author Stobaeus, flourishing circa 500 A.D., quotes a passage from Bardasanes who reports the visit of an Indian to Syria in the time of Antoninus of Emesa (218-222 A.D.). It contains a striking reference to an image of Arddhanārīśvara (the androgynous composite image of Śiva and Durgā; Fergusson, H.I.E.A., p. 54). Hiuen Thsang frequently refers to Brahmanical shrines and sometimes also to the images worshipped there by the sectaries in his Si-yu-ki.

c10.44

¹ A. K. Coomaraswamy, op. cit., p. 42, fn. 5. But ‘Hercules’ in this passage may also have meant Kṛṣṇa; we have seen above that Heracles’ name is mentioned in connection with the Saurasenas and Mathura by Megasthenes. Dionysios is the Greek counterpart of Śiva.

c10.45

It has been shown above how some of the post-Vedic literature of India furnish us with valuable data regarding the prevalence of concrete representations of gods as the objects of worship in India during a few centuries before and after the Christian era. It may be argued that all of these passages do not definitely prove that actual images were being worshipped, and some of them may only refer to the aniconic symbols that might have served the same purpose as well. Archæological data will now help us to throw fresh light on this question, and a careful study of these, divisible into three groups, viz., epigraphic, monumental and numismatic, will show that in India of the pre-Christian and early post-Christian period, worship was being conducted by the various sectaries among her people through media, both iconic and aniconic in character. In some cases the data supply us with direct evidence while in others with indirect. Before a reference is made to a few inscriptions associated with one particular sect, it will be of interest to refer to the interpretation of one or two passages of Aśoka's edicts, which have been taken by some scholars as alluding to the representations of divine figures. In the first part of the Fourth Rock Edict of Aśoka occurs a passage which has been translated by Hultzsch as follows:—' showing the people representations of aerial chariots, representations of elephants, masses of fire and other divine figures' (*Vimāna-darsaṇā ca hastida(r)saṇā ca agikhamdhāni ca añāni ca divyāni rūpāni dasayitpā janam*). He suggests that the figures of elephants stood for the celestial elephants, the usual vehicles of the four Mahārājas or Lokapālas, mentioned above; *agikhamdhāni*, according to him, may be taken in the sense of radiant beings of another world, and *divyāni rūpāni* (identical in sense with *deva* in the Rupnath edict, E) means the gods in effigy (*i.e.*, the images of the gods). It may be observed in this connection that the same sense is indicated by the word *daivata*, used in the Gṛhyasūtras and some Smṛti works, as is done by the words *divyāni rūpāni* and *deva* found in Aśoka's edicts. By exhibition of these objects in large gatherings of his subjects (these *samājas* were considered meritorious by Aśoka), Aśoka desired to remind them of the gods whose abodes they would be able to reach by the zealous practice of *dhamma*.¹ These divine images and other representations had merely edificatory value and were not objects of regular worship in shrines.

c10.46

¹ Hultzsch, CII, Vol. I, Girnar Rock Edict, p. 7, fn, 7.

c10.47

Several pre-Christian epigraphic records, however, like the Ghosuṇḍi and the Besnagar ones, refer to Bhāgavata shrines. The former, discovered on the wall of a *bāolī* (deep masonry well) in the village of Ghosuṇḍi, originally hailed from Nāgarī, 4 miles to the south-west of it, in the old Udaypur State in Rajputana; Nāgarī has been correctly identified with ancient Madhyamikā on the basis of numismatic evidence. Further discoveries of two other copies of the same record (the last made in 1934-35 by the Government Epigraphist) have enabled D. R. Bhandarkar to present to us a complete reading of the three-line inscription which runs thus:—

c10.48

(1) Kāritoyaṃ rājñā Bhāgavatena Gājāyanena Pārāśarī putreṇa Sa-

c10.49

(2) rvatātena Aśvamedha-yājinā bhagava(d)bhyāṃ Saṃkarṣaṇa-Vāsudevābhyāṃ

c10.50.1

(3) anihatābhyāṃ sarveśvarābhyāṃ pūjāśilāprākāro Nārāyaṇavāṭikā. It has been translated by him as follows:— ‘(This) enclosing wall round the stone (object) of worship, called Nārāyaṇa-vāṭikā (Compound) for the divinities Saṃkarṣaṇa-Vāsudeva who are unconquered and are lords of all (has been caused to be made) by (the king) Sarvatāta, a Gājāyana and son of (a lady) of the Parāśara-gotra, who is a devotee of Bhagavat (Viṣṇu) and has performed an Aśvamedha sacrifice.’ Here is an undoubted reference to a shrine of the two gods round which a stone enclosure was built in the 1st century B.C. (that is the date assigned by Bhandarkar to the records, though previous opinion was to place them somewhat earlier); but we are not certain about the nature of the objects which were enshrined there. J. C. Ghosh suggested that these were two śalagrāma stones (pūjā-śilā) corresponding to the varieties of Saṃkarṣaṇa and Vāsudeva as laid down in the Agnipurāṇa. Bhandarkar is justified in criticising this view and in his interpretation of pūjā-śilā-prākāra; but his own suggestion that the objects enshrined were ‘the footprints of the two brother gods carved in stone’, on the basis of his discovery in the western part of the wall at Hāthībāḍā (Nāgarī) of the inscription Śrī-Viṣṇu-pādābhyāṃ in characters of the 7th century A.D., is also not convincing.

c10.50.2

In course of his excavations at Nāgarī, he found in the western half of the Hāthībāḍā enclosure the remains of a brick platform which ran from east to west; he says there is no evidence of any superstructure on it, which fact also led him to arrive at the above conclusion. The superstructure may have been a wooden one, as he himself suggests, or even made of brick, all traces of which may have disappeared in course of time. A shrine was thus most presumably on the spot, and it is extremely probable that the objects of worship there were the two images of the gods. We shall presently see that there are other archaeological data which conclusively prove the existence of figure sculptures of the gods in this period. The above inscription also incidentally shows the composite character of the religious practice of the higher section of the Indians; king Sarvatāta, belonging to the Bhāgavata creed and erecting the enclosure round the shrine of his chosen gods, had already performed the Vedic Aśvamedha sacrifice (cf. the practice of the imperial Guptas). The well-known Khambaba pillar inscription at Besnagar of the 2nd century B.C.

c10.50.3

records the erection of a Garuḍa-dhvaja in honour of devadeva Vāsudeva by ‘Bhāgavata Heliodora (Heliodorus), son of Diya (Dion) and an inhabitant of Taxila, who came as an ambassador from the Greek king Antialkidas to king Kāsīputra Bhāgabhadra of Vidiśā.’ It can very well be presumed that this Garuḍa column was erected in front of the shrine of Vāsudeva who was, to this Greek convert to Bhāgavatism, the God of the gods, the chosen one; the name Garuḍa also shows that, by this time, the association of the sectarian god Vāsudeva with the Vedic Viṣṇu (cf. my previous observations about Sun conceived as the bird Garutman and Viṣṇu as one of the Ādityas) had already been established. That there were other shrines of Vāsudeva at Besnagar is proved by a fragmentary inscription on the shaft of another octagonal Garuḍa column found in a narrow street of Bhilsa, evidently hailing from Besnagar; it records that ‘this Garuḍa column of the excellent temple of the Bhagavat was erected by Gautamīputra..., a Bhāgavata, in the 12th year after the installation of Mahārāja Bhāgavata’ (Gotamiputena bhāgavatena... Bhagavato prāsādottamasa Garuḍadhvaja kārito dvādaśavasābhiṣite....Bhāgavate ma...). So, there cannot be any doubt with regard to the existence of the shrines of Bhagavat before which these votive columns were erected (this was also a common custom in the mediaeval period and is still pursued).

c10.50.4

In these excellent temples (uttama prāsāda) must have been enshrined objects of worship which were most presumably images. A few of the seven Brāhmī inscriptions from Mathura and its vicinity, edited by H. Lüders in the *Epigraphia Indica*, Vol. XXIV, have special bearing on the subject at issue (some of these inscriptions were previously edited, but Lüders suggested improved readings for them). The Mora Well Inscription of the time of Mahākṣatrapa Rajuvula’s son Svāmī (Mahākṣatrapa Soḍāsa) records the establishment of the images of the worshipful pañcavīras of the Vṛṣṇis in the stone shrine...; these images are called ‘five objects of adoration made of stone radiant, as it were, with highest beauty...’ The part of the original, translated here, reads—‘(i) Mahākṣatrapasa Rājūvulasa putrasa svāmi...(ii) bhagavatām vṛṣṇīnām paṃcavīrānām pratimāḥ śailadevagri...(iv) ārcādeśam śailam paṃca jvalata iva paramavapuṣā . . . . . ’.

First 65 of 111 passages. Read on

Linked texts (25)