Shiva (Siva) Nataraja: Re-examining the Origins of Nataraja Bronzes

One of the most recognizable Hindu icons, both inside and outside India, is the standardized depiction of Shiva Nataraja (Shiva king of dance) seen in places as far apart as Chidambaram, Tamil Nadu and the CERN nuclear research center in Geneva, Switzerland. This particular standardization of Shiva Nataraja seems to have arisen under the rule of the Chola queen Sembiyan Mahadevi, as the first fully three dimensional stone carvings in this style appeared during her reign, though questions have been raised about earlier origins (Srinivasan, 434). This standardized form is distinctive and easily recognizable in several ways. First, this particular style of Shiva Nataraja is distinct from not only depictions of other deities, but also other depictions of Shiva as cosmic dancer, by the raised left leg held high across the body at the level of the hip with the foot at knee level (Srinivasan, 433). The supporting right leg, and indeed all the limbs save the lower left arm, are deeply bent giving an appearance of movement paused in a single frame (Kaimal, 392-3). Though held straight, the left arm does faintly bend at the wrist and the hand is held in a relaxed gesture known as gajahasta or “elephant hand” (Kaimal, 393). His lower right hand is held, just above the wrist of the lower left, in abhayamudra, a gesture of fearlessness seen frequently in Indian and Indian-influenced art (Kaimal, 393). The two upper arms hold a damaru drum (right) and a flame (left) (Srinivasan, 433). The foot of the supporting right leg rests on a dwaf, Apasmara, the demon of ignorance (Srinivasan, 433). Finally, in the bronzes, though not in the stone depictions commissioned by queen Sembiyan Mahadevi, Shiva is surrounded by a ring of flames (Srinivasan, 433). The popularity of this image has far outlasted the Chola dynasty, and inspired many speculative interpretations of the iconography present.

Origin of the Image

It is generally accepted that the style of bronze Nataraja we see today originated, or at least rose to prominence, during the reign of queen Sembiyan Mahadevi of the Chola dynasty during the tenth century (Dehejia, 209). Mahadevi was a great patroness of the arts, she commissioned numerous pieces of art and even engaged in the refurbishment of several brick temples, rebuilding them in stone (Dehejia, 209). Mahadevi made the job of archeologists in our own time somewhat easier by re-inscribing previous information about donations and patrons in the temples she refurbished, providing a rich historical record (Dehejia, 209). Notable in regard to the Nataraja image is that it seems to have appeared first in bronzes and stone carvings during her refurbishments (Dehejia, 209). While the similarity of these Nataraja images to present depictions in this style is undeniable, the peculiar raised foot and four armed form being present, doubts have been raised recently about a definitively Chola origin (Srinivasan, 432).

There are certainly examples of images and sculptures which could have contributed to the present Nataraja image exemplified at sites like Chidambaram and CERN, so a pre-Chola origin is not out of the question. One of the earliest possible ancestors of the Chola-era Nataraja is a stone figure from the Harappan civilization, which shares the raised leg posture with the Chola-Nataraja (Dehejia, 32). Granted, a single oddity from a civilization that died thousands of years before the Chola rose is a tenuous connection at best, but Srinivasan points to numerous other examples which may indicate a continuous line of artistic evolution culminating in the Nataraja images we see today.

One of Srinivasan’s suggested precursors is a Satavahana statue, of Shiva as Lakulisa the ascetic, from Andhra Pradesh in southeastern India, the statue is dated to around the first or second century B.C.E. (Srinivasan, 434). What is remarkable about this statue is that, already as early as the first or second century B.C.E., we see the theme of Shiva trampling a dwarf which appears not only in Chola-era Nataraja images but in Pallava depictions as well (Srinivasan, 434).

The Pallava dynasty, in fact, is where Srinivasan asserts that the image of Shiva Nataraja we are familiar with today rose to prominence. Prior to the Chola overthrow of their dynasty around 850 C.E., the Pallavas ruled in the Tamil regions of south India from about 550 C.E., themselves having risen from the older Andhra dynasty (Srinivasan, 434-5). When the Pallava king Mahendravarman Pallavan converted from Jainism to Shaivism a burst of Hindu art in stone was produced (Srinivasan, 435). We can surmise that these stone icons were probably a distinctly Pallava innovation in the Tamil region by inscriptions at Mamallapuram praising Mahendravarman for building in “neither brick, nor timber, nor mortar.” (Srinivasan, 435).

What is interesting about these Pallavan stone icons is that the depictions of Nataraja among them show the four-armed Shiva with the raised leg and dwarf, of which there are no prior examples outside the Tamil region in stone or metal (Srinivasan, 435). Examples of Shiva Nataraja from outside the Pallava-controlled Tamil region show Shiva in the chatura tandava posture with both feet touching the ground and knees splayed outward, as opposed to the bhujangatrasita karana posture in which one leg is raised at hip level across the body which we see in the Chola bronzes (Srinivasan, 435). In addition, the dwarf is not present in any of these chatura tandava examples (srinivasan, 435). The number of arms also differs from the four-armed depictions seen in the Pallava and Chola examples, we see eight arms in Gupta examples from the Sirpur region of central India dating to the fifth century, and sixteen arms in a Chalukyan example from Badami in south-west India dated to the sixth century (Srinivasan, 435).

The earliest clear approximation of the Chola style Nataraja we see is on a Pallava pilaster from a cave temple at Siyamangalam, dated to the seventh century (Srinivasan, 436). This icon stands in the bhujangatrasita karana posture, although with the right leg raised, his lower right hand is in abhaya mudra with his upper right hand holding a lamp or bowl with a flame (Srinivasan, 435-6). This statue does differ additionally from the Chola examples in that its lower left arm extends out away from the body rather than across the body, though it retains the gajahasta gesture (Srinivasan, 435-6). Furthermore, the upper left hand holds an ax and the dwarf is not present under the foot of the supporting leg (Srinivasan, 435-6). This is paralleled in an eighth century cave painting from Ellora in Maharashtra, attributed to the Rashtrakuta dynasty, as well another Pallava stone icon in the Tirukkadaimudi Mahadeva temple in Tirucchinampundi (Srinivasan, 436).

While evidence seems to suggest that, in the cave temples constructed by Mahendravarman stucco and wood images are most likely to have been the norm, a seventh century verse by the poet Appar mentions Shiva’s “sweet golden foot raised in dance”, so we can not rule out bronze processional icons (Srinivasan, 436). In addition, the mention of Shiva holding a drum in the image worshipped at Tillai (now Chidambaram) from the same seventh century verses by Appar seems to indicate that this aspect of the standardized Nataraja icon was already incorporated during the Pallava dynasty (Srinivasan, 436).

Hindu bronzes have not often been attributed to the Pallavas, due largely to a lack of inscriptions on the bronzes themselves, however there is no definite way to date solid metal artifacts with any known method (Srinivasan, 436-8). What we can do, however, is group metal artifacts by shared ore sources based on lead isotope content (Srinivasan, 437). There are some metal artifacts which have been attributed to the Pallavas, for instance a bronze of Shiva dancing in the urdhvajanu pose found in Kuram (Srinivasan, 440). This bronze is attributable to the Pallavas in part because of the forward facing dwarf, as opposed to the sideways facing dwarf in the Chola Natarajas, in addition it shares a metallurgical profile with other artifacts from the reign of Paramesvaravarman Pallavan I (Srinivasan, 440).

This Pallava metallugical profile becomes interesting in regard to two Nataraja bronzes previously attributed to the Chola dynasty, which share the lead isotope content of the Pallava bronzes and the left legged bhujangatrasita karana posture and four armed form of the Chola bronzes, with the hands of each arm bearing the same gestures and implements (Srinivasan, 440). The first, from Kunniyur, differs from Chola images in that it lacks the flying locks of hair found in the Chola bronzes, though the ring of fire is surprisingly present, a date around 850 C.E. is suggested (Srinivasan, 440-1). The second, a small bronze from the British Museum, differs in several ways; the raised leg does not cross the body, the dwarf faces forward, and both the flying locks and circle of flame are not present (Srinivasan, 440-1). This second bronze has been dated to around 800 C.E., making it the oldest known Pallava bronze of Shiva Nataraja (Srinivasana, 440-1). This may indicate that the ring of flame was the latest addition to the Nataraja icon.

It may be that these two Pallava images show an evolution from wood carvings of Shiva Nataraja due to their compactness and lack of flowing locks, both indicative of the limits of wood’s tensile strength, we see these same limits in modern wood carvings of Nataraja (Srinivasan, 440). This may explain the increasingly flared out and circular nature of the icon in Chola times as the tensile strength of bronze was understood to allow for these stylistic changes.

These issues of tensile strength may also indicate that properly three dimensional stone carvings of this style of Nataraja came later than the bronzes and were, in fact, modelled on pre-existing bronzes. We see the emergence of three dimensional stone Natarajas in this style during the reign of Sembiyan Mahadevi, and these images bear the signs of a struggle to represent the style found in the bronzes in a medium with lesser tensile strength (Srinivasan, 441). For instance, in the stone Nataraja from Manavalesvarar temple at Tiruvelvikudi, we see a strut disguised as clothing supporting the lifted leg and crossed left arm to allow for a more expansive image which would make more sense in a bronze casting (Srinivasan, 441-2). The lifted leg of an eleventh century Chola sculpture at the Gangaikondachalapuram temple is propped up by a rough basal strut, while in several other examples the lifted leg is completely broken off (Srinivasan, 442). These struts may even have been inspired by the runners which facilitate lost-wax casting, though they are usually removed from the finished product (Srinivasan, 442). All of this seems to indicate that the style of Nataraja statue attributed to the Chola dynasty was already well developed as such, and likely in bronze, during the Pallava dynasty.

Iconographic Interpretation

An influential, and enduring, interpretation of the Nataraja icon was offered close to one hundred years ago by Ananda Coomaraswamy in “the Dance of Shiva” (Kaimal, 390). While Coomaraswamy’s interpretation is certainly compelling, and likely responsible for the popularity of the Nataraja icon in the west and its interpretation by Western scholars for the last hundred or so years, there is some reason to doubt its accuracy in reflecting the way that the Pallavas and Cholas interpreted this icon when they developed it (Kaimal, 391). Kaimal offers three fairly compelling reasons for questioning Coomaraswamy’s interpretation. First, the question of if it is even possible to properly recover the original meaning of these objects, given the fragmentary evidence from medieval India (Kaimal, 391). Second, Kaimal questions whether a single interpretation is sufficient, noting that objects of art take on different meanings during different times and in fact live multiple symbolic ‘lives’ (Kaimal, 391). Finally, Kaimal draws attention to the fact that Coomaraswamy based his interpretation on texts written several centuries after this style of Nataraja rose to prominence (Kaimal, 391). On this last point, Kaimal also reminds us that there is no simple equivalence between text and sculpture, both mediums have their own “spheres of eloquence” which do not always overlap entirely (Kaimal, 391).

Kaimal is cautious not to completely reject Coomaraswamy’s interpretation however, as it does reflect the significance of the icon to devotees in the thirteenth century and later (Kaimal, 392). While elements of the thirteenth century interpretation could have, and in all likelihood did, derive from earlier interpretations, Kaimal offers three different interpretations which may reflect the meaning of this icon for devotees in the tenth century and possibly earlier (Kaimal, 392). The first interpretation, that Nataraja was used as a kind of emblem of the Chola dynasty is certainly compelling and well argued by Kaimal. Though, while it could serve as the subject of a book in its own right, this interpretation does not tell us much about the symbols within the icon or their origin, which are the primary foci of this paper.

Kaimal’s second interpretation deals with the origin, or synthesis, of this Nataraja icon in Chidambaram (previously Tillai). When Appar wrote about Tillai in the seventh century, it was already an ancient and well established center of many sects, including sects devoted to Vinshnu and the goddess (Kaimal, 391). Kaimal points to earlier interpretations of Nataraja from Tillai which see the tandavam as a dance much more associated with Shiva’s destructive aspects than with the lofty philosophical interpretation of Coomaraswamy (Kaimal, 401).

Many of the less obvious symbols built into the Tamil Nataraja sculptures do indeed point to an association with the destructive aspects of the creative cycle, and many of these symbols appear on depictions of other wrathful aspects of Shiva all over India (Kaimal, 401). For instance, the skull often present in the hair of Nataraja icons and the serpents which encircle his limbs often receive special emphasis in images of Shiva’s destructive aspects, such as the ‘enraged’ face on the giant three-faced Shiva at Elephanta (Kaimal, 402). These often indicate Shiva as Aghora, associated with cremation grounds and destructive ecstasy, as well as drawing an association with similarly adorned goddesses such as Kali, Chamunda, and Nishumbhasudani (Kaimal, 401). These wrathful goddesses also share the characteristics of deeply bent supporting legs and multiple arms splaying out in an explosive and energetic fashion (Kaimal, 402). That these symbols were present in earlier forms of Shiva and other gods/goddesses may indicate that they were redeployed to allow this icon to participate in a symbolic conversation which was already ongoing, and this interpretation would fit nicely with a gradual evolution of the form from the Pallava dynasty through the Chola standardization (Kaimal, 404).

The association with goddesses is interesting in regard to another possible origin of the icon. One of the origin myths laid down in the Chidambaramahatmya, a tenth century text reflecting the Sanskritization of the Tamil cult at Tillai into a pan-Indic cult, tells of a dance competition in which the goddess already resident at Tillai, Tillai Amman, resented Shiva’s encroachment and challenged him to a dance competition (Kaimal, 407). Shiva won the competition by taking a raised leg posture, which modesty prevented the virginal goddess from copying (Kaimal, 407). This loss split the goddess in two, the wrathful virginal aspect retreated to a shrine outside the temple walls, while her benign aspect became Shiva’s wife and remained in the temple where her worship continued. This may reflect an earlier tradition being replaced by, or syncretized into, a more pan-Indic cult rooted in Upanishadic Hinduism rather than the local Tamil culture. This Sanskritization of a local cult may reflect political or social changes brought about as a result of empires growing larger and larger which had to unify disparate belief systems without abolishing them.

Another myth, also presented in the Chidambaramahatmya support the hypothesis that symbols present in the Nataraja icon derive from earlier cults which where absorbed in, and Sanskritized by, the Nataraja cult. The “Pine Forest myth” relates the story of Shiva visiting several sages who were living in a pine forest to punish them for their devotional inadequacies (Kaimal, 406). Shiva arrives in the form of a nude and mirthful ascetic, Bhikshatana, who was sexually irresistible to the wives of the sages, he was accompanied by Vishnu in his female form, Mohini, who proved distracting to the sages themselves (Kaimal, 406). When the sages realized their humiliation they became infuriated and attacked Shiva with various objects which he incorporated into his dance (Kaimal, 406). After incorporating the objects hurled at him by the sages, Shiva’s dance intensified until it encompassed all of creation (Kaimal, 406). As the sages saw this dance they became enlightened by the cosmic proportions of Shiva’s true form and instituted the worship of Shiva in an aniconic form as the linga, which we see carried on at Chidambaram today (Kaimal, 406).

It is the particular items thrown at Shiva, and their incorporation into his dance, which interest us here. The items were: a skull, which Shiva wears in his hair; serpents, which adorn Shiva’s limbs and hair; a dwarf, which he tramples underfoot; a tiger, to which are attributed the shredded appearance of Shiva’s flowing garment; and the fire and drum which we see in Shiva’s two upper arms as well as the flaming ring within which he dances (Kaimal, 406). It certainly is not out of the question to see this legend as a possible reference to earlier Tamil cults, represented by the items, being displaced by and absorbed into the cult of Shiva as a pan-Indic god. This interpretation would further support the idea of a unification of disparate local cults as the empire grew to incorporate, and accommodate, more cultural groups. This is by no means the last word on the origins of the Nataraja icon, but it may indicate that a reappraisal is in order.

Works Cited

Dehejia, Vidya. Indian Art. Phaidon, 2011, London.

Kaimal, Padma. “Shiva Nataraja: Shifting Meanings of an Icon” in The Art Bulletin, 81, 3. College Art Association, 2009, New York.

Srinivasan, Sharada. “Cosmic Dancer: On Pallava Origins for the Nataraja Bronze” in World Archaeology, Vol. 36, No. 3. Taylor & Francis, 2004, Abingdon.

Article written by Logan Page (Dec. 2018), who is solely responsible for its content.