Pramāṇa · Search · About · API

The Development of Hindu Iconography c18: Chapter XI: Cult Icons—Śiva and Śakti

The Development of Hindu Iconography c18

Language
English

Contents

Chapter XI: Cult Icons—Śiva and Śakti

c18.1

CULT ICONS

Śiva

c18.s1.1

Siva is one of the most influential members of the orthodox Brahmanical triad. Though he is specially associated with the act of saṃhāra (destruction) or pralaya (absorption) in the Hindu concept of Trinity, an exclusive worshipper of Siva thinks him no less associated with the other two acts of sṛṣṭi (creation) and sthiti (preservation) generally attributed to Brahmā and Viṣṇu, the other members of the triad. Siva is also endowed with the acts of anugraha or prasāda (‘conferment of grace’) and tirobhāva (‘power of concealment’ or ‘obscuration’). These collectively make up his five-fold activities (pañcakṛtyas), and his numerous manifestations according to his worshippers are connected with the performance of one or other of them for their edification and benefit. Siva is also described in the early and mediaeval Brahmanical texts as the originator and the best exponent of various arts and accomplishments, such as those of deep concentration (yoga), expounding (vyākhyāna) the śastras (the various āgamas were revelations from him), music, dancing, etc. He has also been conceived as the lord of all created beings, and is often described as Paśupati, Bhūtapati and Bhūtanātha. He is the great lord (Maheśvara), the greatest of the gods (Mahādeva), the beloved husband of Umā, the great Mother-goddess (Umāpati—in this aspect he is the primeval Father-god), and is the chief possessor of Māyā (Māyin). Many and various are his aspects and attributes which appear to be delineated in his numerous epithets and names, which are sometimes incorporated in different sections of the epic and Purāṇic literature (cf. Mahābhārata, XIII, 17, which enumerates more than one thousand names of Śiva—Śivasahasranāma). A careful analysis of some of these epithets only will prove that this god-concept in its developed phase arose out of the commingling of many such concepts current among different ethnic units of India or among the different branches of the same ethnic stock.

c18.s1.2

The epic or Purāṇic Śiva undoubtedly had his Vedic and pre-Vedic counterparts or prototypes. Rudra was the Vedic counterpart of Śiva, and the Vedic-Brahmanic characterisation of Rudra enunciates how many of the traits of the god are inherent in the one hundred names of Rudra enumerated in the Śatarudrīya text of the Śukla Yajurveda of the Vājasaneyī school (Vājasaneyī Saṃhitā, Ch. 16). The association of Rudra with the hills (one of his names in this context is Giritra or ‘one who lives in the hills’), his dual nature of a terrific as well as a pacific god, his use of animal-hide as his garment (Kṛttivāsa), the crown of snail-shell like jaṭās on his head (Kapardī), etc., are emphasised by these epithets or names which were the roots of many mythologies elaborately narrated in the epic and Purāṇic literature.¹ The worship of a god similar in some respects to the Purāṇic Śiva seems to have been well in vogue among the early Indus Valley people. The name of this pre-Vedic prototype is not now known to us, but Rudra continued to be one of the principal names of Śiva in the epics and Purāṇas. In the Vedic texts of earlier strata the word śiva occurs as one of the attributive epithets of several Vedic gods, and not of Rudra in particular. In such uses of the word, its original sense of ‘auspicious’ is retained, and it is only in the later Vedic literature that it is used as a proper name. The Śvetāśvatara Upaniṣad, a theistic text extolling the glories of Rudra, uses the word śiva several times as one of the various names of Rudra. The Sāṅkhyāyana, Kauṣitakī and other Brāhmaṇas use such names as Śiva, Rudra-Śiva, Mahādeva, Maheśvara and Īśāna for denoting this great god, as the Śvetāśvatara Upaniṣad also does. Weber suggested long ago that such epithets like Mahādeva, Īśāna, etc., of this god ‘involving quite a special prominence of the deity as compared with the other gods indicated a sectarian worship’.¹ Considering the myths about the origin and growth of the concept of Rudra, Aufrecht also observed that ‘the period of the Brāhmaṇas was one when the old polytheism was in a condition of decline and the new faith which presents itself in Indian religious history as Śaivism was gaining ground.’² The Atharvaveda presupposes the rise of Rudra to the position of the supreme god, for it assigns various such epithets to him as Bhava, Śarva, Paśupati, Ugra, Mahādeva and Īśāna. The Śatapatha and Kauṣītakī Brāhmaṇas add Aśani to this list thus making their number eight (inclusive of Rudra), each four of which typify his two aspects, ghora (terrific) and saumya (peaceful). Thus, Rudra, Śarva (‘arrow-wielder’), Ugra and Aśani characterise his destructive aspect, while Bhava, Paśupati, Mahādeva and Īśāna, his beneficent one. Such well-known names of Purāṇic Śiva as Śambhu and Śaṃkara indicative of his beneficent nature occur as so many epithets of Rudra as early as in the Śatarudrīya which also contains epithets descriptive of the dark and fierce aspects of the god.³

c18.s1.3

¹ “The epithet Kapardin or ‘the wearer of matted hair,’ is probably due to his being regarded as identical with Agni, the fumes of which look like matted hair”; R. G. Bhandarkar, op. cit., p. 103.

c18.s1.4

¹ Indische Studien, ii, 302. ² A. B. Keith, Ṛgveda Brāhmaṇas, p. 26. ³ ‘He is called the lord of the paths, of the forests and of those who roam in them, of thieves and highway robbers who frequent and move about in lonely places to prevent being detected, and also of outcasts who live away from the usual dwellings of men’; R. G. Bhandarkar, op. cit., p.

c18.s1.5

It is not possible for us to determine the exact date of the complete emergence of the cult centering round Rudra-Śiva. The process of its evolution was undoubtedly gradual, and several literary data of the pre-Christian period seem to indicate the stages of its growth and development. Pāṇini in one of his Sūtras (IV. 1. 112) seems to refer to the followers or worshippers of Śiva. The aphorism (Śivādibhyoṇ) means that ‘the affix aṇ comes in the sense of a descendant, after the names Śiva, etc.¹ The word śaiva, thus formed, may be presumed to denote ‘a worshipper of Śiva’, as the same significance was attached to it in the Mahābhāṣya.¹ Patañjali refers to Śiva as well as Rudra several times. Rudra is twice described as the god to whom animals are sacrificed (paśunā Rudram yajate), and in two other passages the medicinal herbs of Rudra are called auspicious (Śivā Rudrasya bheṣajī).² It has already been noted in Chapter IX of this book that Śiva is mentioned as one of the folk divinities (laukika devatās) in the Mahābhāṣya (supra, p. 338). But it is interesting to note in this connection that one of the earliest clear references to a Śaiva sect is to be found in this text. While commenting on Pāṇini’s Sūtra, V. 2. 36 (ayaḥśūladaṇḍājinābhyāṃ ṭhakṭhañau), Patañjali mentions the Śiva-bhāgavatas who carried an iron lance (āyaḥśūlikaḥ). He further observes that because this type of a sectarian devotee seeks to obtain his desire by violent means, the fulfilment of which should be sought for by mild ways, he is called āyaḥśūlikaḥ (Keilhorn, op. cit., Vol. II, pp. 387-88). Though the commentator does not comment on the term daṇḍājina in the Sūtra, it is apparent that a staff (daṇḍa) and a hide (ajina) were also the characteristic marks of a Śiva-bhāgavata who was an āyaḥśūlika and at the same time a dāṇḍājnika. These characteristic traits of the exclusive worshippers of Śiva should be compared with those of the Sibae (or Siboi), a tribe living in a part of the Panjab during the time of Alexander’s invasion. The Sibae are described by Curtius, Diodorus and other classical writers as dressed in animal skins, having clubs for their weapons, and branding with the mark of a club their oxen and mules. The religious association of this Panjab tribe with Śiva appears to be substantiated by Patañjali who, while commenting on the Vārttikas on Pāṇini’s Sūtra, IV. 2. 104 (avyayāt tyap) refers to the Udīyagrāma (‘northern village’) Śivapura or Śaivapura (Keilhorn, op. cit., Vol. II, pp. 396-97).

c18.s1.6

¹⁰³ Rudra is described as Eka-Vrātya, ‘the Vrātya par excellence’, and this epithet characterises in a very interesting manner the association of elements in his composite character, not derived from the orthodox Vedic order.

c18.s1.7

¹ Pāṇini also mentions several names of Rudra such as Bhava, Śarva and Mīḍha in one of his Sūtras (IV. 1. 49).

c18.s1.8

² Keilhorn, Mahābhāṣya, Vol. I, pp. 331, 424; Vol. III, pp. 14, 403.

c18.s1.9

Mahāmāyūrī, one of the five great mantra-formulas of northern Buddhism, composed sometime before the 4th century A.D. (it was translated into Chinese four times between the 4th and the 8th centuries A.D.), refers to the tutelary deity of the āhāra (district town) of Sivapura as Śiva. Sylvain Levi long ago connected Sivapura of this Buddhist text with Udīcyagrāma of Patañjali (*Journal Asiatique*, 1915, pp. 37, 70). It can legitimately be inferred from all these data that there lived in parts of Northern India exclusive worshippers of Śiva even long before the time of Patañjali.

c18.s1.10

The Śiva-bhāgavatas of Patañjali seem to have had some association with the Pāśupatas referred to in later texts and inscriptions. The Nārāyaṇīya section of the Mahābhārata enumerates five systems such as Sāṃkhya, Yoga, Pāñcarātra, Veda and the Pāśupata. It describes the teacher of the last as Śiva-Śrīkaṇṭha, the lord of Umā, the master of spirits, and the son of Brahmadeva (Śāntiparvan, Chs. 64-7). This may or may not imply the existence in history of a possible human founder of the religious system developing round Śiva. But Purāṇic and inscriptional data undoubtedly refer to a human being, Lakulīśa by name, an inhabitant of ancient Kāyārohaṇa (modern Karvan in the Kathiawar Peninsula), who became intimately associated with the growth and development of the Pāśupata cult. The Mathura stone inscription of the time of Chandragupta II (G.E. 61 = 381 A.D.) helped D. R. Bhandarkar to assign to Lakulīśa (described in the Purāṇas as the 28th and the last incarnation of Śiva) a date in the first half of the 2nd century A.D. Such a date would not justify us in describing him as the originator or founder of the Pāśupata school, if the Pāśupatas are to be identified with the Śiva-bhāgavatas of Patañjali.¹ If the identification of the two orders be correct, and there are good grounds for this assumption, then Lakulīśa must have to be regarded as the organiser or systematiser of the system, and not as its actual founder.' The great drive he gave to the pre-existing worship of Śiva-Paśupati according to the already accepted tenets was continued afterwards; his four immediate disciples, Kuśika, Mitra, Garga and Kauruṣya, are traditionally regarded as the expounders of four sub-sects branching out of the Pāśupata system. That Kuśika and others were not legendary figures seems to be proved by the fact that Ārya Uditācārya, the donor in the Mathura inscription noted above, describes himself as tenth in apostolic succession from Kuśika (Kuśikād-daśamaḥ). Lakulīśa composed a work called Pañcārthavidyā, a passage from which is quoted by Mādhavācārya in his Sarvadarśanasaṃgraha. It lays down that 'a Pāśupata should bathe thrice a day, should lie upon dust (or ashes), and oblation is an observance divided into six members'. In the Atharvaśiras Upaniṣad, a sectarian work devoted to the exaltation of Rudra (Śiva)'s glory, the Pāśupatavrata is described as consisting of besmearing the initiate's body with ashes in conjunction with the muttering of a mantra. By performing this vow (vrata), the worshipper of Rudra-Śiva has his bonds loosened, and he is delivered from the fetters of his individual existence. This is Paśupāśavimokṣaṇa, and it is one of the principal characteristics of the Pāśupata school. Varāhamihira refers to the Pāśupatas as sabhasma-dvijas (cf. supra, p. 230) and Hiuen Tsang describes the same sect as 'the ash-besmeared tīrthikas'. The Pāśupata Sūtras and the commentary on it by Kauṇḍinya (works earlier than the Bṛhatsaṃhitā) not only mention this vow about the ashes, but also incorporate other practices of the Pāśupata ritualists, which are of an outlandish character.² The Kāpālikas, the Kālāmukhas and similar other Śaiva sects of a ghora type were developed at a fairly early date from the Pāśupata sect as organised by Lakulīśa, and the peculiar and apparently unsocial and outlandish practices of these extreme forms of Śaiva sectaries appear to be reminiscent of the terrific aspect of Rudra-Śiva. Literary and inscriptional data of the Gupta and post-Gupta periods give us a glimpse into these rituals which substantiate to a great extent the association of the ghora forms of the god with these extreme schools.¹

c18.s1.11

¹ R. G. Bhandarkar has identified the Śiva-bhāgavatas with the Pāśupatas, and has placed Lakulīśa a little earlier than or contemporaneous with Patañjali; *op. cit.*, pp. 116-17.

c18.s1.12

¹ For this view, cf. the present writer's article on 'Lakulīśa, the Founder or Systematiser of the Pāśupata System' in Proceedings of the Jaipur Session of the Indian History Congress, pp. 32 ff.

c18.s1.13

² Some of these are 'doing a thing condemned by all' (avitatkaraṇa), 'speaking nonsensical and absurd things' (avitadohāsaṇa), 'showing oneself to be in love by means of amorous gestures on seeing a beautiful woman' (śṛṅgaraṇa), etc. These constitute the Vātulatantra mentioned by Utpala, the commentator of the Bṛhatsaṃhitā (cf. supra, p. 230).

c18.s1.14

¹ Bhavabhūti refers to the practices of the Kāpālikas in his Mālatīmādhava; the Kāpālikas are described there and in some other texts as Mahāvratadharas. The Nirmaṇḍ copper-plate inscription of the time of the Maukhari chief Sarvavarman mentions the installation of the image of Kapāleśvara (C.I.I., III, pp. 286-91; Proceedings of the Bombay Session of the All-India Oriental Conference, 1949, pp. 298-300). R. G. Bhandarkar quotes a few other instances of the allied Kālamukha sect., op. cit., p. 120.

c18.s1.15.1

A brief reference may now be made to the Śaiva schools of the other group whose tenets are of a moderate character. As a class this group is of a later date, and the tenets of most of them are more philosophical than ritualistic in character. One of the most important and perhaps the earliest in this group is the Śaiva system, another later and developed variety of which was the school of the Śuddha-Śaivas or the Āgamānta Śaivas. These Śaivas are to be distinguished from the Vedānta Śaivas who made full use of the teachings of the Vedas and Upaniṣads. The Śaivas or the Śuddha-Śaivas, on the other hand, based their tenets principally on the Āgamas which were, according to them, direct revelations from Śiva, and thus were of far greater importance than the Vedas regarded by them as only ‘his breath’ (Yasya niśvasitaṃ Vedaḥ). The Śaiva Āgamas (28 in number) enjoin that there are three principles,—the Lord (Pati, i.e., Śiva), the individual soul (Paśu), and the fetters (Pāśa),—which bind the individual to his mundane existence. Like the Pāñcarātra saṃhitās, the Śaiva Āgamas discuss the four parts or stages, Caryā, Kriyā, Yoga and Jñāna, of a well-regulated religious life. Caryā is the stage of practical piety and performance of the prescribed duties and rites; Kriyā consists of the confirmatory sacrament (dīkṣā) and the five purifications associated with true worship;

c18.s1.15.2

Yoga denotes the stage at which the individual soul attains communion with Śiva as a result of the eight usual observances of the Yogin; and lastly Jñāna, the highest stage preparatory to final liberation, is that of true knowledge which makes the individual soul pre-eminently fit for intimate union with god.¹ Śambhūdeva, one of the exponents of the school, calls it the Siddhāntaśāstra, or the true Śāstra based upon the Mantras revealed by Śiva himself. The Sāṃkhya and the Yoga systems of philosophy no doubt influenced the tenets of the Āgamāntins, but in the popular presentation of the latter women, Śūdras and Prātilomas were allowed to participate. The holiest Pañcākṣara Mantra (Namaḥ Śivāya) of the Śaivas could be muttered and meditated on by every Śivabhakta, and an initiated Śūdra could take part in the ritual worship of the Śaiva images. Some of the Śaiva images again, as will be shown later, illustrated in an esoteric manner different sections of the philosophy of this school. The other moderate and highly philosophical school of Śaivism is that which was given shape and developed in Kashmir, as the Āgamānta Śaiva school was given final shape in the south, at first by Vasugupta and then developed by his two pupils Kallaṭa and Somānanda in the 9th century A.D. This school, known as the ‘ Trika ’ system, for it also dealt with the three categories, Pati, Paśu and Pāśa, had two branches, Spanda and Pratyabhijñā, associated respectively with Kallaṭa and Somānanda.

c18.s1.15.3

These two branches did not lay any great stress on disciplinary ritualism; the first, i.e., Spandaśāstra emphasised intense contemplation for the individual aspirant after the highest realisation of god within himself, the second on the spiritual aid of the preceptor in helping his disciple to recognise truly the identity of his own self with god. But the followers of this monistic form of Śaivism were not much concerned with the worship of Śaivite icons which appear to have been made much use of in their disciplinary *sādhanās* by the Pāśupatas and the Śaivas. The sect of the Vīraśaivas or Liṅgāyats, which was systematised by Vasava as late as the second half of the 12th century A.D. in Southern India, did not go in also for image worship; but the worship of the ‘Iṣṭa-Liṅga’ (the aniconic emblem of the god of their choice,—their *iṣṭadevatā* Śiva), which they used to carry on their body, comprised the chief act of divine worship by the followers of this sect. The special class of mediaeval Tamil Śivabhaktas, known generally as the Nāyanārs (or Nāyanmārs), on the other hand, appear to have been mostly emotional worshippers of the god, to whom temple-ritual was necessary for the outward manifestation of their one-souled devotion to their deity. The first seven collections of the numerous songs in Tamil composed by them comprised the most sacred *Devāram* hymns sung ceremonially in the Śaiva shrines of the south.

c18.s1.15.4

The best honoured among these saints was Tiruñānasambandha who appears to have flourished in the seventh century A.D.; he composed as many as three hundred and eighty-four hymns known as *Paḍigam*. There are references in these songs to many south Indian Śaiva shrines and the particular aspects of the god associated with them; the devotional Śaiva hymns composed by Appar, Sundaramūrti, Māṇikkavasahar and other saints of the south also contain such references. The songs characterise the pacific aspect of the god in a loving manner, though in some of them veiled references to his destructive or *ugra* aspects are made.

c18.s1.16

¹ Five purifications are the removal of ‘ evil desires, anger, and I-ness from one's heart, and replacing them by God, and feeling his presence there at all times’ (St. Vallalar); the eight observances or elements of a Yogin are ‘ Good qualities (yama), good acts (niyama), right posture, breath control, removal of mental distraction, concentration of the mind on God, spiritual contact (yoga), and spiritual experience ”; S. Shivapadasundaram, The Śaiva School of Hinduism, pp. 177-78.

c18.s1.17

It has just been mentioned that the Vīraśaivas of the south mainly worshipped the emblem of Śiva, and were not much in favour of worshipping him in other types of images. But from a fairly early period the Śaivas in general used to place this emblem (Śivaliṅga) as the principal object of their worship in the main sanctum of the shrines of their god, and the various types of his anthropomorphic figures were carved in the different parts of the temples more or less as accessory figures. The Kailāsa temple of Ellora, like many other Saiva temples of India, housed the emblem in the main sanctum, and the numerous anthropomorphic forms of Siva, mostly illustrative of stories associated with the god, were arranged in rows in the covered verandas of the quadrangle, as a veritable sculpture gallery. It thus behoves us first to say something about the real nature of this emblem in its earlier phase, and note some details about its developed varieties of a comparatively late date. It may be observed at the very outset that the principal idea underlying it in its most primitive aspect is phallic in character. Gopinath Rao demonstrated it clearly by many quotations from several Purāṇas and other early and late texts, as well as by his elaborate study of one of the earliest such emblems with the figure of Siva carved on it discovered by him at Guḍimallam (near Renigunta in Andhra State).¹ It has been shown in the fourth and fifth chapters of this book how the numismatic and glyptic representations of the emblem from the 3rd-2nd century B.C. to the 4th-5th century A.D. also support the hypothesis of Gopinath Rao. Incidental references to phallicism prevailing among the early Indus Valley people and to the Vedic deprecation of this practice have also been made in the 2nd and 3rd chapters of this book. One of the reasons for the hatred of the orthodox Vedists towards the Saivas might be traced to this practice, as has been suggested by Rao. There is very little doubt that the orthodox section of the Hindus at first showed tardiness in giving sanction to this practice, and this is proved by the fact that no clear reference to the phallic emblem forming a part of the ritualism of some of their people is found in the older sections of their literature. It is only in the epic literature that we find for the first time unmistakable evidence of the worship of Śiva in his phallic form, and that too in sections adjudged as late ones by Indologists.¹ The reason for the earlier non-acceptance of the emblem by a section of the Hindus is undoubtedly to be found in its stark realism in the early period. But the old ideology about it as a symbol of the virile father-god was too strong among a large section of the people to be brushed aside and ignored by the less numerous but more intellectual section. The latter, however, appears to have made a compromise. They accepted the emblem as the holiest one of Śiva, but on one condition, that of conventionalising it in such a way that its original realism was thoroughly subdued by stages. The gradual change started from the Gupta age, the age of cultural renaissance in India. So great was this change in the manner of its ultimate representation, that some modern scholars thought that it was derived from the Buddhist stūpa model.² It is also not quite correct to suppose that it symbolised the union of the male and female principle even from its early phase. True it is that the pīṭha part of the comparatively late Śiva-liṅgas of the conventionalised shape is often described as the arghya or yoni (the female principle) in many late texts; but Tāntricism had developed to a great extent when they were written, and the projecting portion of the pīṭha really served the very useful purpose of draining off the water profusely poured on its top to some distance from its base.

c18.s1.18

¹ Rao, op. cit., Vol. II, Pt. I, pp. 55-63, 65-71. His identification of the inscribed sculpture found at Bhita (near Allahabad—it is now in the Lucknow Museum) as a Pañcamukha Śivaliṅga of the pre-Christian period (pp. 63-5) does not bear scrutiny; cf. the present writer's observations on it in J.I.S.O.A., Vol. III, 1935, pp. 40 ff.

c18.s1.19

¹ Mahābhārata, XII. 14. 231-33 : Upamanyu tells Kṛṣṇa that ‘Mahādeva is the only deity whose organ of generation was worshipped in former times or is now worshipped by the gods such as Brahmā, Viṣṇu and others, and Śiva and Umā were the real creators of animals, because the latter carry on their bodies the marks of these two and not the marks of other divinities’.

c18.s1.20

² E. B. Havell was once the exponent of this view; The Ancient and Mediaeval Architecture of India (1915), pp. 106-07.

c18.s1.21

Archaeological data also conclusively support these suggestions. One of the earliest extant realistic emblems of Śiva is the Guḍimallam Liṅga just mentioned. Carved out of hard igneous stone of a dark brown colour with some polish, it abruptly stands five feet above the floor level (there is no pedestal) and is almost a foot in thickness. The nut is differentiated from the shaft by a deep slanting groove cut near the top, and the front part of the shaft bears on it the two-armed figure of Śiva in high relief standing on the shoulders of the crouching figure of Apasmāra-puruṣa. It may be incidentally noted that the urdhvaliṅga sign usually found on Śiva figures of the post-Christian period is not emphasised (the ‘organ’ shown downwards is, however, clearly discernible through the somewhat diaphanous drapery), and the Śivaliṅga is fixed in a hole cut into the floor. Śiva holds in his right hand a ram, and in his left a small water vessel with a battle-axe resting on his left shoulder (Pl. XXXI, Fig. 3). The exhibit No. H. 1 in the collection of the Lucknow Museum described as a ‘Liṅga with a broadened top’ is a realistic type of this emblem from the north, but without the figure of Śiva on its shaft. This red sandstone sculpture comes from the Mathura region. The nut is clearly demarcated from the shaft by the broadening of the top from a slanting groove cut round the junction of the shaft and the nut, and there is a decorative band in the middle of the latter; it is of the Kushan period. Another huge stone Śivaliṅga in the Mathura Museum collection (measuring as much as 200 cm.) is divided into three sections: (1) a roughly square undressed lower portion, (2) the round middle section with its girth shortening upwards, and (3) the tapering round nut broader than the top portion of the middle section. The nut is distinctly marked on one side of its lower end by a dentate groove helping to suggest some realism. The suggestiveness is, however, somewhat subdued here, and it is evidently of a little later date. With the passing of time, the process of conventionalisation made much advance, and the stages of its rapid progress can be seen in any assortment of the phalli representative of different periods. The Karamdaṇḍā inscribed Śivaliṅga of the time of Kumāragupta I (Gupta year 117) illustrates an early stage in this process.

c18.s1.22.1

Gopinath Rao gives an exhaustive account of the various types of Śivaliṅgas and their classifications on the basis of various Śaiva āgamas and other texts.¹ I shall only make some observations on a few of the types, for it will be impossible for me to incorporate much about the individual groups of Śaiva emblems and images in the first part of Chapter XI of this book. The *Makuṭāgama* classifies the Sthiraliṅgas (*acala* or *dhruva*, i.e., immovable or permanently fixed) into four classes, the Daivika, the Ārṣaka, the Gāṇapa and the Mānuṣa, whereas the *Kāmikāgama* groups them under six heads,—the Svāyambhuva, the Daivika, the Ārṣaka, the Gāṇapatya, the Mānuṣa and the Bāṇaliṅgas. Students of iconography are not much concerned with most of these classes, the last two only of the second list being necessary for study. The Bāṇaliṅgas are natural objects like the Śālāgrāmas, and are also fished out of a particular river-bed like the latter. They are mostly picked up from the river Revā or Narmadā, one of the seven most sacred rivers of the Hindus, as the Śālāgrāmas (specially associated with Vaiṣṇavism) are procured from the bed of the Gaṇḍakī, a tributary of the Ganges in the north. These are the types of emblems which many Śaivas carry on their body and offer daily worship to them.

c18.s1.22.2

The Mānuṣa or manmade liṅgas from the largest group of the Sthiraliṅgas, and are made up of three parts known as Brahmabhāga, Viṣṇubhāga and Rudrabhāga.¹ The first is the square lowest section, the second, the octagonal middle one, while the third or the topmost one is generally cylindrical. Rudrabhāga is known also by the name of Pūjābhāga, for the offerings of water, flower and other objects are usually put on its top; the two other sections are inserted inside the pedestal (*pīṭhikā*) and the ground. The *Bṛhatsaṃhitā*, though it does not give us the names of these sections, supplies us the same information, and adds one more to the effect that the periphery of the piṇḍikā or pīṭhikā would be the same as the height of the cylindrical portion.¹ It should be noted that the author does not refer to the original phallic character of the emblem, and it is presumable that either its full conventionalisation had taken place by his time or more probably he did not want to say anything about it. The carving of certain lines technically known as *brahmasūtras* on the Rudrabhāga of the emblem is enjoined in some later iconographic texts; this is nothing but a device to demarcate the nut from the shaft. It is interesting to note that according to *Makuṭāgama*, the *brahmasūtra* design is not to be shown in the Daivika and Ārṣaka Liṅgas, *i.e.*, those associated with the Devas and the divine Ṛṣis;

c18.s1.22.3

but it must be present on Mānuṣa Liṅgas which are made by human hands according to canons laid down in Śaiva Śāstras (*Śivaśāstroktamārgeṇa śirovartanayā yutam| Brahmasūtra samāyuktametamānuṣamucyate*). In some Sivaliṅgas of the early mediaeval period the *brahmasūtra* lines are incised, but in many later specimens they are seldom emphasised. The Mānuṣa liṅgas are grouped under various heads on the basis of different criteria. One method is based on the measurements of the three sections of the emblem, the names varying as their proportions differ; some of these names are Sarvasama, Vardhamāna, Svastika, Sārvadeśika, Trairāśika, etc. Several classes of Sivaliṅgas are distinguished by such names as Dhārāliṅga, Aṣṭottaraśataliṅga or Sahasraliṅga, chiefly on the basis of the different ways of modelling their Rudrabhāga. If the cylindrical shaft of a Mānuṣaliṅga has fluted facets, it is called Dhārāliṅga, the number of the vertical flutes according to the *Suprabhedāgama* being five, seven, nine, twelve, sixteen, twenty-four and twenty-eight (*Pañcadhārāssaptadhārā navadvādaśadhārakah| Dhārāṣṣoḍaśavimśacca triraṣṭāvāṣṭāvimśatih*). The flutes on such emblems served the purpose of draining the water poured on their top, which then was ejected through the *nāla* projection of the *pīṭhikā*.

c18.s1.22.4

The Aṣṭottaraśata- and the Sahasra-liṅgas, as their names indicate, are those on the Rudrabhāga of which are carved 108 or 1000 emblems by the incision of parallel horizontal and vertical lines, and by chamfering the edges of the resultant criss-cross sections. Multiple representations of the emblem are regarded as very pious by a Śaiva devotee, and sometimes 108 Śivaliṅgas designated by 108 of the numerous names of Mahādeva are installed in concentric rows of separate shrines. Mukhaliṅgas constitute another class of Mānuṣaliṅgas, conventional in character, and as a class much later than the types of the realistic liṅgas like Guḍimallam (the Dhārā-, Aṣṭottaraśata- and Sahasra-liṅgas are also of later origin). They denote those emblems which show on their Rudra- or Pūjā-bhāgas one or more human faces (the faces standing for one or more aspects of Śiva). The *Kāraṇāgama* lays down that ‘the face should be 13½ *aṅgulas* in length, and the number of the faces should be 5, 4, 3 or 1; Mukhaliṅga with four faces should have them facing four quarters, a three-faced one should not have a face on its back, and a one-faced one should have the face placed a little high up’. There is also some textual injunction about the number of faces being determined by the number of doors of the central shrine. But it is doubtful whether it was always followed. The *Rūpamaṇḍana* writes about three-, one- and four-faced Mukhaliṅgas;

c18.s1.22.5

‘in a one-faced type the face should be shown in front, and in a three-faced one, the back face is absent. In a four-faced variety, the western face is white, the northern red, the southern face black and terrific, while the eastern face is of the colour of a well-kindled fire. The (five) faces stand for the five aspects of Śiva—Sadyojāta, Vāmadeva, Aghora, Tatpuruṣa and the fifth Īśāna which is beyond the comprehension of even the Yogīs’.¹ But it should be noted that though the Kāraṇāgama refers to a five-faced Śivaliṅga by the epithet śarānanam (śara means arrow, and the number of Cupid's arrows is five), it does not refer to the position of the fifth face. The Rūpamaṇḍana does not mention such a type, for the fifth, i.e., the Īśāna face, is beyond the ken even of the Yogīs (pañcamam ca tatheśānam yogīnāmapyagocaram). The extant specimens of Mukhaliṅgas usually show one, three and four faces carved on the Pūjābhāga or Rudrabhāga; they hail from all parts of India, and the earliest of them seldom goes prior to the early Gupta period. Exhibit No. 42 in the Lucknow Museum is an interesting specimen of the Ekamukha type; the realism is very much subdued (it is of the Gupta period), and it is curious to note that the third eye is horizontally placed in the centre of the forehead (in the heads of Śiva the position of the third eye is usually vertical, Indra's third eye being shown horizontally).

c18.s1.22.6

Though no text known to me describes a Dvimukha Liṅga, one sculpture in the collection of the Mathura Museum seems to me to represent this variety. The third eye (placed here also horizontally) and the jaṭāmukuṭa on the two Janiform heads on the shaft apparently establish the identity of the sculpture; the faces, curiously enough, have moustaches. One unique variety in red sandstone (also in the Mathura Museum collection) consists of four shafts joined together, with one face on the top section of each of them; there appears to have been some attempt to demarcate the nut from the shaft, but the result has not been sufficiently realistic. The hair of the four heads is arranged differently. Diskalkar describes it as a Pañcamukha Liṅga and dates it in the first century A.D., but there is no certainty whether there ever was another head on the top of the sculpture, and on stylistic and other grounds it can be assigned to the late Kushan period.¹

c18.s1.23

¹ Rao, op. cit., Vol. II, pp. 75-99.

c18.s1.24

¹ Some textual descriptions of pillars or columns of the late Gupta or early mediaeval period refer to their sections as Brahmakāṇḍa, Viṣṇukāṇḍa and Rudrakāṇḍa; cf. Mānasara, ch. 15. Stambhalakṣaṇam, vv. 10-11; the verses read *Caturasram Brahmakāṇḍam syādaṣṭāsram Viṣṇukāṇḍakam | Ṣoḍaśāsram tu vṛttamvā Rudrakāṇḍamiti smṛtam* (Acharya wrongly puts the word as *kānto*).

c18.s1.25

¹ Bṛhatsaṃhitā, ch. 57, 53-4: *Liṅgasya vittaparidhim dairghyenāsūtrya tat tridhā vibhajet| Mūle taccaturasram madhye traṣṭāsri vṛttamataḥ || Caturasramavānikhāte madhyam kāryam tu piṇḍikāśvabhre| Dṛśyocchrāyeṇa samā samantataḥ piṇḍikāśvabhrāt*. Utpala clearly explains the sense of the last line in this way—*Yāvatpramāṇam vṛttabhāgasya dairghyam tāvatpramāṇam samantataḥ pīṭhikāpṛthutvam kāryamiti* ||

c18.s1.26

¹ *Rūpamaṇḍana* as quoted by Rao—*Mukhaliṅgam trivaktraṃ syādekavaktraṃ caturmukham | Sammukhaṃ caikavaktraṃ syāt trivaktre pṛṣṭhake na hi || Paścimāsyaṃ sthitaṃ śubhraṃ kuṅkumābhaṃ tathottare | Yāmyam kṛṣṇakarālaṃ syātprācyaṃ dīptāgnisannibham || Sadyo Vāmam tath-Aghoram Tatpuruṣam caturthakam | Pañcamam ca tatheśānaṃ yogināmapyagocaram ||*

c18.s1.27

¹ For illustration of these three sculptures, cf. J.I.S.O.A., Vol. III, 1935, Pl. VII, Figs. 2-4.

c18.s1.28.1

The Guḍimallam Liṅga and the Mukhaliṅgas illustrate in a way the combined mode of representing Śiva both in his human as well as phallic form. Coomaraswamy illustrates an interesting Mathura sculpture of the late Kushan period, in which a four-armed standing figure of Śiva is carved on one side of a long pillar like emblem; the natural hands of the god are shown in *abhaya* and *kaṭyavalambita* poses, while the added ones are raised and placed on his *jaṭās* (op. cit., Fig. 68). References to the presence of Śiva on columnar altars are to be found in the earlier sections of the *Mahābhārata*. Thus, in its Kirātārjunīya episode, Arjuna is described as worshipping Bhava (Śiva) with a garland after making an earthen altar (as the divine symbol) (II. 39. 65—*Śaraṇyaṃ śaraṇaṃ gatvā bhagavantaṃ pinākinam | Mṛṇmayam sthaṇḍilam kṛtvā mālyenāpūjayadhavam*). Aśvatthāmā, on the nocturnal journey to the Pāṇḍava camp to murder the five Pāṇḍava brothers, is confronted by a gigantic figure at the gate; he then invokes the aid of Śiva, his patron deity, and there appears before him a huge golden altar with all-spreading flames of fire on it.¹ This concept of the sudden appearance of a flaming golden altar or pillar before Aśvatthāmā leads one to the consideration of the Liṅgodbhavamūrti of Śiva.

c18.s1.28.2

The mythology connected with this type of Śaiva image, a combination of the god's human as well as his columnar form (he is also called Sthānu), relates how Brahmā and Viṣṇu were at one time disputing their individual claims for the creation of the universe. Śiva suddenly appeared before them in the form of a blazing column of fire. Brahmā and Viṣṇu tried respectively to find its top and bottom, but they failed. Brahmā, however, falsely asserted that he had succeeded in his effort, for which falsehood Śiva cursed him never to have a cult of his own. Viṣṇu confessed his inability to find the ground of the column; Śiva, who had in the meanwhile become manifest in it, blessed him to have his own cult almost equal in importance to that of himself. This theme, undoubtedly evincing a sectarian bias, was very much popular with the Indian artists of the early and late mediaeval periods. Rao has illustrated three reliefs in illustration of this theme, from Conjeevaram, Ellora and Ambar Mangalam. The Chola sculpture being illustrated here is from the Bṛhadīśvara Temple of Tanjore; it depicts the story in the usual but a summary manner. The worshipping figures of Brahmā and Śiva on either side of the flaming pillar as in the sculpture from Daśāvatāra cave Ellora (Rao, op. cit., Vol. II, Pl. XIV, Fig. 1) are absent;

c18.s1.28.3

but the four-armed Candraśekharamūrti of Śiva with paraśu and mṛga in his back hands, the front ones being in the abhaya and kaṭya-valambita poses, is beautifully carved inside an elliptical cavity on the surface of the column which is decorated with a festoon design on its top. Brahmā is shown flying up in its top left corner, and the boar-faced Viṣṇu is depicted burrowing down below (Pl. XXXI, Fig. 4). The south Indian reliefs of Liṅgodbhavamūrti are usually of the Ellora and Tanjore types. The motif is not very common in Northern India, but reference may be made here to one very interesting sculpture illustrating the theme, which is now in the collection of the Rajputana Museum, Ajmer. The long slender column on which Śiva’s figure is depicted has the figures of Brahmā and Viṣṇu depicted on its sides in the act of soaring upwards and coming downwards respectively, and there are some accessory figures clustering round the central object; Brahmā and Viṣṇu are again shown as respectful attendants of Śiva (J.I.S.O.A., Vol. IX, Pl. X). The emblem itself enshrined in the main sanctum of many south Indian shrines looks like a huge column of a very wide growth (cf. those in Bṛhadīśvara Temple, Tanjore, Kṣitiliṅga at Śiva-Kāñchi, Jyotirliṅga at Tiruvanamalai, etc.).

c18.s1.28.4

It may be incidentally suggested here that the erection of columns or upright stones to commemorate one’s ancestors was a funerary practice common not only to the people of ancient India, but also to many other ancient nations of the world. It appears that this practice had also something to do with the growth and development of the phallic cult in India.

c18.s1.29

¹ *Mahābhārata*, X. 7. 13-14: *Iti tasya vyāvasitaṃ jñātvodyogāt svakarmanaḥ | Purastāt kāñcanī vedī prādurāsīnmahātmanaḥ || Tasyāṃ vedyāṃ tadā rājaṃścitrabhānurajāyata | Sa diśovidiśaḥ khañca jvālābhirabhipūrayan ||*

c18.s1.30

¹ *J.I.S.O.A.*, Vol. III, 1935, pp. 7-9. I endorsed there Fuhrer's explanation of the Bhita sculpture as 'the capital of a column'. The five faces on it were explained by me as symbolising the 'sons of Khajahuti' to commemorate whom the 'Column' was erected or installed (*Khajahutiputānāṃ lago patiṭhāpito*) by Nāgasiri, the son of Vāseṭhi. The phallic sign carved in outline on one part of it can be explained by referring to an observation of Grant Allan: 'On many grave-stones of early date a phallus marked the male sex of the occupant' and 'the stone being regarded as the ancestor of the family, it is not unnatural that early men should some times carve it into a phallic shape' (*The Evolution of the Idea of God*, p. 68).

c18.s1.31

The aniconic and aniconic-iconic forms of Śiva have been discussed at some length in the preceding pages, and it is now time to study some of his representations in his two-, four- or multi-armed human forms, occasionally showing more than one head. These images can first be classified under two broad heads, those depicting his *ugra* or *ghora* (terrific) aspect, and the others illustrating his *saumya* or *śānta* (peaceful) one. Each of these groups again can be generally subdivided under two broad heads on the basis of their illustrating myths commonly associated with the god, or not. Those that are not primarily associated with any particular Śivaite story retain the early form, an idea about which we can have from the many representations of the god on ancient Indian coins and seals. Such images of the peaceful category have been given various names mainly of a descriptive character in the iconographic texts collected in the Śaiva Āgamas. Thus names like Candraśekhara, Umā-sahita, Āliṅgana Candraśekhara, Vṛṣavāhana, Sukhāsana, Umā-Maheśvara, Somā-Skanda, etc., explain the different varieties of Śaiva images. The first three in this list depict the god standing either alone or in company with his consort Umā (actually embracing her), while the second three depict him as seated in ease either alone or accompanied by Uma and sometimes by both Umā and Skanda (*sa-Umā-Skanda* : Somā-Skanda). Other graceful or placid forms of Śiva are described in the various Śaiva Āgamas under such names as Dakṣiṇāmūrti and Nṛtyamūrti, etc., which portray the god as the master in the various arts of dancing, playing on musical instruments, expounding the śāstras, and practising the Yoga. Śiva Naṭarāja (the ‘king of the dancers’) dances various types of dance, such as Nādānta or Tāṇḍava, Lalita, Lalāṭatilaka, Kaṭisama, Talasamphoṭita, etc., the karaṇas of which are described in Bharata’s Nāṭyaśāstra. Śiva as the greatest teacher of yoga, of vīṇā (lute or gourd with strings), of jñāna (knowledge) and vyākhyāna (expounding the śāstras) is known by such names as Yoga-Dakṣiṇāmūrti, Vīṇādhara-Dakṣiṇāmūrti, Jñāna-Dakṣiṇāmūrti and Vyākhyāna-Dakṣiṇāmūrti. Another group of images depicting his placid aspect, but not any particular mythology, are his Ardhanārīśvara and Hari-Hara or Haryarddha forms which attempt to emphasise in a way the syncretism of the central deities of different cults, Śaiva, Śākta and Vaiṣṇava. Some Śaiva images, mostly evolved in south India in mediaeval times, are intended to illustrate in an esoteric manner some of the principal tenets of Āgamānta Śaivism or Śuddha Śaivism; they are known by such names as Sadāśivamūrti, Mahā-Sadāśivamūrti and Maheśamūrti. Lastly, mention should also be made of the image of his ‘twenty-eighth incarnation’, Lakulīśa who, as we have already shown earlier, was responsible for systematising the Pāśupata creed. Lakulīśa images are very seldom found in Southern India, but they are frequent in Orissa in the east and Gujrat and Kathiawar Peninsula in the west.

c18.s1.32

The images of the ghora or ugra form, which are not associated with any particular story narrating the exploits of Śiva, can be called Bhairava, Aghora, Raudra-Pāśupata, Vīrabhadra, Virūpākṣa, and Kaṅkāla. Bhairava is sometimes described as Brahmaśiraśchedakamūrti (of Śiva) in some of the Purāṇas and Śaiva Āgamas; these texts try to explain this aspect of Śiva as cutting off one of the heads of the polycephalous Brahmā, for his alleged sins and iniquities. But there is very little tangible connection between these varying myths and iconic types. The Āgamic texts enumerate as many as sixty-four Bhairavas divided in eight groups of eight each, the leaders of these groups being Asitāṅga, Ruru, Caṇḍa, Krodha, Unmatta-Bhairava, Kapāla, Bhīṣaṇa and Saṃhāra respectively. They are the consorts or guardians of the sixty-four Yoginīs mentioned in the Tāntric texts. The particular type of Bhairava usually found in Northern India goes by the name of Baṭuka (‘youthful’) Bhairava. He is nude, terrific in appearance with protruding fangs, rolling and round eyes, and his hands hold such objects as a sword, a khaṭvāṅga, a śūla or a kapāla; he usually wears wooden sandals, and is often shown accompanied by a dog. A less terrific type of image of the deity is his Kaṅkālamūrti, in which he carries on the prongs of his trident, the skeleton of Viṣvaksena, the gate-keeper of Viṣṇu, who was killed by Śiva for his refusal to admit him into the presence of Viṣṇu. Such images of the late mediaeval period are comparatively common in Southern India. Bhikṣāṭanamūrti of Śiva, mythologically associated with the Kaṅkālamūrti, is, however, of a placid type. It shows the god as a wandering youth of the untouchable order, usually nude, holding a kapāla in one of his hands, and is sometimes accompanied by a frisking deer.¹

c18.s1.33

¹ Rao has described and illustrated most of these varieties of Śaiva images in his monumental work, Vol. II, pp. 105 ff. and plates. He has not given any account of the Lakulīśa images, probably for the reason that they are not common in Southern India.

c18.s1.34.1

Before a brief account is given of the other broad group of Śaiva icons, both of the saumya or ugra type, which very characteristically illustrate stories connected with Śiva performing acts of anugraha (grace) or saṃhāra (destruction), it is necessary to study a few reliefs of the early and late mediaeval period, representing the ‘non-mythological’ groups of Śaiva icons. Gopinath Rao describes three types of Candraśekharamūrtis as the Kevalamūrti, the Umāsahitamūrti and the Āliṅganamūrti on the basis of Amśumadbhedāgama and other Śaiva Āgamas, and illustrates them mostly by late mediaeval bronze and stone sculptures from Southern India (op. cit., Vol. II, Pls. XV-XX). Kevala Candraśekhara denotes those images of this type, in which the god is shown alone; in the Umāsahita- and Āliṅgana-Candraśekharamūrtis, the god either stands with Umā by his side or he lovingly embraces his consort. Śiva is called by this name for the presence of the crescent moon on his jaṭās (Candraśekhara, Śaśāṅkaśekhara), and in these mediaeval south Indian images he holds paraśu and mṛga in his back right and left hands respectively, his front ones being shown in the abhaya and the varada poses (cf. the passage in his dhyāna : paraśumṛgavarābhītihastam). Such images are usually decorated with ornaments.

c18.s1.34.2

Two interesting sculptures from Eastern and Northern India are being illustrated here, which, though not tallying closely with the south Indian group just mentioned, may yet be considered along with them. The beautiful four-armed figure of Śiva from Khiching shows the god standing gracefully with a slight bend in his body (ābhaṅga), holding a kapāla in his front left hand; his back right hand holds a rosary; the front right one is broken, while the back left hand, the palm of which is gone, must have held a triśūla (its three pronged top is quite distinct on the top corner). The god stands on the pericarp of a double-petalled lotus with two gracefully poised female attendants carrying jars (of wine or poison) on his either side. The finely carved ornaments, the beautifully arranged jaṭāmukuṭa, the very elegantly displayed halo, the well-balanced scroll designs on the two sides and the lower part of the prabhāvalī, the exquisite modelling of the main figure and its attendants, and, above all, the sublime expression of pleasant contemplation on the divine face mark the sculpture as one of the finest examples of Indian art (Pl. XXXII, Fig. 1). In marked contrast to this remarkable specimen of sculptural art of mediaeval Orissa, stands the Hara-Pārvatī from Kosam, of a far earlier (Gupta) date, strikingly simple and unconventional. The ithyphallic god stands facing with a slight bend of the body, holding a flask in his left hand (cf.

c18.s1.34.3

the nectar flask in the hand of the earlier Maitreya figures) with Umā on his left; the goddess holds a darpaṇa in her left hand, and though the ornaments on the figures are sparse, the Gupta sculptor took particular care to give a character to the head-gear of the divine couple. Their faces are not devoid of expression, but none of the sublime introspection of the Khiching sculpture is present there. The inscription on the pedestal bears a date in the reign of Kumāragupta I (Pl. XXXVIII, Fig. 2). A still earlier red sandstone relief of Hara-Pārvatī or Umā-sahitamūrti of Śiva is to be found in the Mathura Museum. It shows the divine couple leaning on Nandī; Śiva is ithyphallic, and both the god and the goddess hold *nīlotpala* buds in their hands. This is one of the earliest sculptural representations of the god in his aspect of Umāsahitamūrti. It will be of interest to compare and contrast these Hara-Pārvatī images with the south Indian Āliṅgana-Candraśekharamūrti from the Bṛhadīśvara Temple at Tanjore, reproduced here. The four-armed god stands to front with a slight bend, his front left hand clasping Umā, the other three hands holding a *ṭaṅka* (?), *paraśu* and a *mṛga* from the lower right onwards; the goddess stands demurely on his left, her right and left hands being in the *vismaya* and *kaṭihasta* poses. Though there appears to be a somewhat hard expression on the Devī's face, the god's face bears a deep penetrating expression;

c18.s1.34.4

the head-gear, ornaments and garments of the two divine figures are convincingly displayed. This is a good specimen of the Chola art in its best days (Pl. XXVII, Fig. 3). The Aihole sculpture of Vṛṣavāhanamūrti of Śiva shows the four-armed god standing in a *dvibhaṅga* pose and reclining on his mount, his back right hand holding a *triśūla* and the front left hand, a snake, the front right hand rests in the *varada* pose on the horn of the vivacious bull, the back left hand being indistinct. The facial expression of the god is not very distinctive but its lack of character seems to be much compensated by the vigorous expression of the bull mount (Pl. XXXIV, Fig. 3). The three-faced four-armed Śiva seated on his mount Nandī in the *sukhāsana* pose in an intaglio in the Pearse Collection, now in the Indian Museum, is one of the earliest *āsana* type images of this god. The hair arrangement on two sides of the central face, the *jaṭā* knobs on all the three heads, the flames issuing from the shoulders, the halo encircling the three-heads and the pronouncedly ‘muscular treatment’ of the body,—all characterise the figure as having been carved in the best Gandhāran tradition, and it may belong to the late Kushan period. The back right and left hands of the god hold a triśūla and pāśa respectively, while the front right and left ones rest on the butt of the gadā and the neck of his mount; the animal is shown in a suggestively restful pose (Pl. XXXIV, Fig. 1).

First 49 of 138 passages. Read on

Cites (48)