You are on page 1of 19

Religion Compass 3/3 (2009): 440458, 10.1111/j.1749-8171.2009.00141.

1749-8171 458??? 10.1111/j.1749-8171.2009.00141.x 0 Blackwell Oxford, Religion RECO Journal 141 March 4 Review The Benjamin REVIEW 40??? 2009 Form 2009 The Compilation Article UK Compass J. and ARTICLE Publishing Fleming Author Formlessness Ltd 2009 of Blackwell Siva Publishing Ltd

The Form and Formlessness of Siva: The Liga in Indian Art, Mythology, and Pilgrimage1
Benjamin J. Fleming*
University of Pennsylvania


The liga is well-known as the dominant emblem of the Hindu god iva. This article seeks to provide an accessible survey of the mythology and iconography of the liga, and the scholarly discussion about them. It considers some of the ancient objects that scholars have identified as ligas, reflecting on the methodological challenges involved in their interpretation. It also considers major narrative, theological, and pilgrimage traditions surrounding ligas, as preserved in the PurA{as, and some of the prescriptions for their construction and installation, as outlined in gamas and reflected in current practice. The author suggests, moreover, that liga worship may have played an important role in the trans-regional spread and consolidation of aivism as we know it.

Today, the Hindu god iva is worshipped throughout India, and devotees perform rituals to a liga, a cylindrical object typically made of stone, clay, wood, or metal. The liga is now the dominant emblem of the deity, and liga worship is important for aivite piety and ritual practice across the diverse regional cultures of India; thousands upon thousands of aivite temples across the subcontinent hold a liga in their central-most sanctum. Ancient and medieval evidence for ligas, however, suggests a long and varied history, in art and literature alike. It took centuries for the present form of the liga to evolve, and it also took centuries before liga worship took hold as the single most dominant form of devotion to iva. Nevertheless, the liga has been a constant source of inspiration in myths and liturgies celebrating the god. Arguably, in fact, the history of aivism its mythology, ritual, and theology is inseparable from the history of the liga. This article will survey some of the major literary, theological, and iconographic traditions surrounding ligas, with a focus on its place in the history and practice of aivism. What is our earliest evidence for ligas, and when did this form come to be associated with iva? What is the place of liga worship in aivite myth and ritual, past and present? What role did it play in the development and spread of aivism? In the process of exploring such questions, I hope also to provide an introductory guide
2009 The Author Journal Compilation 2009 Blackwell Publishing Ltd

The Form and Formlessness of Siva


Fig. 1. Liga in main shrine at Auda-Naganatha temple in Maharara. In Maharara, this liga is worshiped as one of the 12 jyotirligas. Photo, B. Fleming.

to scholarly and popular ideas about ligas as well as an entry-point into the study of aivism more broadly. The Form and Formlessness of Siva The liga is virtually unique among the myriad images of gods in South Asian religious art. Hinduism, after all, is famous for its rich artistic and ritual traditions surrounding the representation of divine figures in anthropomorphic forms (Banerjea 1956; Waghorne & Culter 1984; Davis 1999). Typical of the iconography of South Asian deities, moreover, are images with multiple arms, heads, eyes, and other attributes (rinavasan 1997, 1978). Although such features are found in some representations of iva, his dominant emblem, the liga, is distinguished by its extreme simplicity. We find this tendency attested already in the aivite gamas, a set of technical manuals on art, architecture, and ritual that encompass ancient and medieval traditions (especially during and after the ninth century ce). Although the gamas describe different shapes, sizes, kinds, and styles of ligas (see further below), they favor the geometrically simple, cylindrical form as the basic, normative design of this ritual object. Likewise, most ligas in Hindu temples today resemble posts or cylinders (Figs 12). Although they vary in size and material (see below) and are often decorated with garlands during rituals (Fig. 3), they share a simplicity that stands in striking contrast to the dominant emblems of other Hindu gods. The iconographic distinctiveness of the liga may parallel the theological differences between the devotees of iva and those of other Hindu deities.
2009 The Author Religion Compass 3/3 (2009): 440458, 10.1111/j.1749-8171.2009.00141.x Journal Compilation 2009 Blackwell Publishing Ltd

442 Benjamin J. Fleming

Fig. 2. A liga dedicated to Somanatha (a form of Siva) on a secondary platform outside of the main shrine at the Parali-Vaidyanatha Temple in Maharara. Photo, B. Fleming.

Fig. 3. Decorated liga and base in a secondary shrine at the Auda-Naganatha temple in Maharara. Photo, B. Fleming.

The very simplicity of the ligas form resonates with widespread concepts in aivite mythology and theology about the nature of this gods action and appearance in the world. Whereas Vai{avite traditions depict the avatras of Vi{u as acting directly in the world, for instance, aivite traditions portray iva as a distant god, who only participates in the world through his divine horde of followers (gaas). Phyllis Granoff (2004) has shown how the narrative distancing of iva is reflected in a range of medieval
2009 The Author Religion Compass 3/3 (2009): 440458, 10.1111/j.1749-8171.2009.00141.x Journal Compilation 2009 Blackwell Publishing Ltd

The Form and Formlessness of Siva


Sanskrit sources, especially aivite PurA{as and gamas. These sources celebrate iva as a deity who is formless (nirgua) and without attributes (nikala).2 Such descriptions reflect their theological positioning of iva as the ultimate concept in the universe. They evoke the formless iva at the top of a cosmic hierarchy. Examples can be found in traditions about ivas presence in VArA{as}, a city celebrated as sacred to the god.3 In the Jna-saMhit (49) of the Siva Pura tradition, for instance, it is said that all extant matter, divine beings, earthly creatures, etc., emanated in progressive stages of manifestation and materialization from ivas state of formlessness. Near the bottom of this hierarchy, VArA{as} itself emerged out of the tip of ivas trident. It is said, moreover, that his special liga in the city (called Avimukta; lit. the one who never leaves) was set in its place by the god BrahmA. In the continuum linking the highest god with specific sites on earth, the liga is thus lauded as emblematic of ivas formlessness. This physical, ritual object stands in direct relationship to the gods abstract, elevated presence at the top of the universal hierarchy. Scholars such as Wendy Doniger (1973) have pointed to this paradox and have suggested that it is central for aivite theology more broadly. If so, then it is all the more significant that the liga serves as an embodiment of the paradox. According to the Jna-saMhit (49), the highest reality of iva is eternal and without worldly qualities. Yet the liga can be approached and can serve the world in which it now exists. A cylinder is still a form, after all, even if it might seem formless in comparison with anthropomorphic images. The liga is thus an ideal symbol and/or earthly surrogate of the higher reality, from which it is said to derive and with which it is said to retain a connection. The ligas status as a representative of the otherworldly iva resonates with the literal meaning of the Sanskrit term liga, namely, mark or sign.4 Some texts explicitly describe ivas ligas as his avatras a designation typically reserved for the corporeal manifestations of Vi{u, as richly represented in Hindu art and literature. The term avatra, for instance, is applied to ivas manifestation in the form of jyotirligas (lit. ligas of light) in the Satarudra-saMhit (42.160) of the Siva Pura tradition. In aivite mythology, this form of iva is sometimes envisioned as a formless orb of light, floating in mid-air (Granoff 1993, p. 69, verse 40), or through which the god interacts with his devotees (e.g., giving them special weapons). ivas jyotirliga form is also, however, associated with 12 specific sites where the god once appeared and to which devotees can make pilgrimage (Feldhaus 2003, pp. 12830; Yamaguchi 2008; Fleming 2009). In this case too, the liga both represents and mediates the paradox of a formless god who can be specially manifested in specific places on earth. That this particular articulation continues to be compelling is suggested by the popularity of literary, liturgical, and pilgrimage traditions about ivas 12 jyotirligas, beginning almost 1000 years ago and continuing today.
2009 The Author Religion Compass 3/3 (2009): 440458, 10.1111/j.1749-8171.2009.00141.x Journal Compilation 2009 Blackwell Publishing Ltd

444 Benjamin J. Fleming

The Early History of the Liga Thus far, we have considered the form and meaning of the liga from a synchronic and synthetic perspective, examining modern aivite traditions alongside the medieval sources from which they draw (e.g., PurA{as and gamas). This approach reflects one major strand of scholarship on aivism, as exemplified by the work of Doniger (for other synchronic approaches to aivism, see Kramrisch 1981). In her book Siva: The Erotic Ascetic (1973), for example, she explores the internal contradictions within aivite tradition, drawing on a range of diverse materials to evoke an image of the god as a synthetic whole. She proposes that devotees of iva encounter the god as a unity of opposites, orchestrated between the extremes of ascetic distancing and sexual indulgence. Even as Doniger (1973, p. 35) acknowledges that the apparently contradictory strains of ivas nature may well have originated at different times and places, she thus stresses that they have resulted in a composite deity who is unquestionably whole to his devotees. Other scholars, such as Hans Bakker and Phyllis Granoff, take a different approach to investigating the diverse traditions now encompassed within aivism. They seek to reconstruct the origins and development of specific traditions, by taking seriously the differences between our ancient evidence, on the one hand, and our medieval and modern evidence, on the other (e.g., Bakker 1996, pp. 32134; Granoff 2003, pp. 23). Consequently, such scholars tend to focus on the most ancient attestations of iconographical and literary motifs that later became prominent in medieval and modern aivism. Such an approach proves particularly valuable for the study of the liga, due to the striking differences between ancient liga-like objects and the form of the liga familiar today. The earliest history of the liga cannot be told through textual sources. Rather, it must be reconstructed entirely through the analysis of artifacts. Whereas our literary evidence for ligas is largely medieval, the relevant material evidence dates back to as early as the third century bce. In medieval and modern materials, as noted above, iconographical and literary depictions of the liga tend to emphasize a simple, amorphous or cylindrical form. When we look to the older archaeological and sculptural data, however, we find a somewhat different picture. Some of the most ancient sculptural artifacts that both Indian and Western scholars identify as ligas are phallic in form, and some of these phalli even have human faces on them. Comprehensive surveys of the relevant materials can be found in arthistorical studies such as Gerd Kreisels Die Siva-Bildwerke der Mathur-Kunst (1986) and Doris rinavasans Many Heads, Arms and Eyes (1997). Here, it will suffice to consider a few prominent examples, which are representative of the different types of ancient artifacts that scholars identify as ligas. In the relevant materials dating from the second century bce to the sixth century ce, there is no one dominant or consistent form. Nevertheless, one may see a general trend in the extant evidence: phallic imagery is
2009 The Author Religion Compass 3/3 (2009): 440458, 10.1111/j.1749-8171.2009.00141.x Journal Compilation 2009 Blackwell Publishing Ltd

The Form and Formlessness of Siva


gradually downplayed and abstracted over time. Early examples of ligas from the Katrapa period (ca. second to first centuries bce) to the KuA{a period (ca. first to third centuries ce) are also less homogenous than those that come after them. During the course of this period, ligas evolved into forms that were highly abstracted as well as increasingly systematized, perhaps in interaction with the theological discourse on form and formlessness noted above. As the earliest evidence for liga worship, scholars most often point to the Gu|imallam-liga (Fig. 4), which has been dated variously between the third and first centuries bce. Scholarly attention was first drawn to this sculpture by T.A. Gopinatha Rao, who published it in his seminal Elements of Hindu Iconography (1914, vol. 2.1, pp. 6571). When Rao discovered it, in the early twentieth century, it was still being actively venerated. The phallic image stands five feet high, is highly graphic, and is erect. In front of the phallus is a human figure. He stands on a squat dwarf and holds a water pot and a deer; an axe rests against his left shoulder. A ninth-century ce inscription at the temple identifies the sculpture as ParaurAmevara (Lord of RAma with the axe; Sarma 1994, pp. 45). An association with iva is suggested by the -vara suffix, a common feature in epithets of the god. The inscription attests one interpretation of the image: at least by the ninth century the sculpture was thought to represent ivas mastery over the Vai{avite avatra ParaurAma. Since the inscription is a medieval addition to the more ancient object, however, it does not explain the original intent of the sculpture, which remains unknown. Indeed, despite the usual assumption that phallic imagery alone suffices to identify any early object as aivite, more caution may be warranted; there is much about the Gu|imallam-liga that continues to puzzle scholars and much that remains debated about early aivism. One intriguing aspect of the Gu|imallam-liga is its provenance. Its iconography is unique to the region in which it is found; no similar sculptures have been identified in Andhra Pradesh. However one wishes to date this sculpture, it is clearly the earliest of its kind in South India. Its southern provenance is intriguing inasmuch as iconographic features of the human figure recall contemporaneous traditions in the North, particularly the images of yakas (demons or sprites) found on Buddhist stpa railings at Bharhut (ca. second century bce) and other sites (Coomaraswamy 192831, pp. 8, 41; von Mitterwallner 1984, pp. 1218).5 It is possible that further comparative study might help us to make sense of this ancient image and its original meaning. While the Gu|imallam-liga stands in some isolation in South India, there appears to have been a large output of liga-like, phallic sculpture in the North, especially during the KuA{a period (ca. first to third centuries ce). Many such works were produced in MathurA (Kreisel 1986; rinavasan 1997, pp. 26081, 30524). This ancient city was a center for sculptural production, including but not limited to ligas. It
2009 The Author Religion Compass 3/3 (2009): 440458, 10.1111/j.1749-8171.2009.00141.x Journal Compilation 2009 Blackwell Publishing Ltd

446 Benjamin J. Fleming

Fig. 4. Liga at Gudimallam (pre-excavation) dated variously from the third to first centuries bce. Photograph courtesy of John C. Huntington with the permission of the Archeological Survey of India.

seems to have been a highly creative environment; much of the sculpture from MathurA is unique, displaying a variance of forms that contrasts with the systematized iconographical consistency of later periods. Among the known pieces identified as ligas, for example, we find a variety of shapes, sizes, and decorative motifs (von Mitterwallner 1984).
2009 The Author Religion Compass 3/3 (2009): 440458, 10.1111/j.1749-8171.2009.00141.x Journal Compilation 2009 Blackwell Publishing Ltd

The Form and Formlessness of Siva


Fig. 5. So-called Pacakamukha-liga from Mathura, ca. 200 ce. Housed at the Government Museum Mathura, no. 516. This sculpture combines four singled-faced ligas and, according to some scholars, may have had another head on the top, now broken (hence, paca five). Photo courtesy of John C. Huntington.

A significant iconographical form that may have developed in MathurA is the mukha-liga, a phallic-like form featuring one or more human faces. Examples of mukha-ligas have been discovered with one, two, and four faces (e.g., Fig. 5). The single-faced (ekamukha) ligas, in particular, make up perhaps the largest single corpus of early liga evidence. Scholars have long sought to explain the meaning of mukha-ligas with reference to literary evidence, but corroborating textual material is later and more limited than one would like. The motif of iva with multiple faces is presented in one of our earliest manuals of Hindu iconography, the Viudharmottara-Pura (ca. fifth to eighth centuries ce). The possibility of older allusions to this iconographical motif has been discussed by N.P. Joshi (1984, p. 52) and Bakker (1997, pp. 7274; 1999, p. 339), with reference to passages in the Mahbhrata that describe iva with four faces (1.203.26; 13.17.74; 13.128.4; 14.8.30). Such a connection, however, remains tentative, especially as this epic also describes the god BrahmA as having four heads (3.194.12; 3.275.17; 12.335.18). In addition, none of these early literary sources makes any direct connection between this multiface iva and ligas. Although the mukha-liga later became associated with iva, its original meaning remains unclear.
2009 The Author Religion Compass 3/3 (2009): 440458, 10.1111/j.1749-8171.2009.00141.x Journal Compilation 2009 Blackwell Publishing Ltd

448 Benjamin J. Fleming

Granoff (2003, p. 2) rightly cautions that there are problems with the iconography and identity of some of the earliest ligas, the [scholarly] interpretation of which has been based on late sources that clearly show the hands of the theologian-systematizer. It may be tempting to project later traditions backwards onto earlier evidence or to interpret ancient iconographical data through the lens of later literary sources. Yet, it is also important to take seriously how little we know, from the surviving evidence, about the early development of aivism. The precise historical locus for the association between the liga and the god iva, for instance, remains uncertain. None of our earliest examples of ligas and liga-like objects are explicitly, unambiguously, or unequivocally aivite. Accordingly, we may miss something significant about the early history of aivism when we automatically connect such images to iva or conflate them with later traditions. It is possible, for example, that the association of the liga with iva alchemized over a period of several centuries, perhaps continuing to evolve even into the medieval period. That the liga only gradually rose to the status of ivas main emblem is suggested by our ample evidence for anthropomorphic images of iva. The earliest examples date from the same period as most of the earliest ligas and liga-like figures, as discussed above (i.e., KuA{a period), and they continue into the Gupta period and beyond. Such evidence points to regional, ritual, and iconographical diversity within early aivism (see further Goldberg 2008; Falk 1994; Younger 1995, pp. 8789, 1068; Bakker 1997, pp. 6679; 2001). There seems to have been a time, in other words, when liga worship was not yet dominant at a pan-Indian level. The evidence from MathurA, in particular, also points to sustained developments in liga-related sculpture, as the popularity and acceptance of such images grew. In the centuries following our earliest evidence for ligas, we see the influence of aivite literature and theology on its form and iconography. Whereas the iconography of early ligas lacks consistency, our later evidence speaks to efforts at systematization. Innovations and standardization in the iconography of the liga seem to have occurred particularly during the Gupta period (ca. 3rd6th c. ce). This development appears to have been marked, moreover, by concerted attempts to eliminate any overt resemblance of the liga to the human phallus. Jitendra Nath Banerjea (1935, pp. 3644; 1956, pp. 44556), Gritli von Mitterwallner (1984, p. 18, n. 33), and Hans Bakker (1997, pp. 7576) have pointed to the discomfort of the Brahmanical tradition with realistic, phallic imagery.6 They propose that the older form may have been modified due to this discomfort; this abstraction, in turn, allowed for the incorporation of the liga into the broader tradition. One might further suggest that elite theologians may have wanted nothing at all to do with liga worship until it was sufficiently abstracted from phallic realism in the Gupta period and could be absorbed into the increasingly pro-Vedic strand of aivism that was beginning to take form at the time (Fleming 2007, pp. 14083).7
2009 The Author Religion Compass 3/3 (2009): 440458, 10.1111/j.1749-8171.2009.00141.x Journal Compilation 2009 Blackwell Publishing Ltd

The Form and Formlessness of Siva


The Liga in Saivite Myth and Pilgrimage As we have seen, our ancient iconographical evidence raises many questions about the origins of liga worship and the association of the liga with iva. This problem is not found in medieval literary traditions, which explicitly identify it with iva. Interestingly, some of these traditions take up the very question of the origins of liga worship. Perhaps the most famous and influential examples are the myths of the Druvana (Pine Forest) and the Ligodbhavamrti (form of the arising liga). One version of the Druvana myth occurs in the Krma Pura (2.36.492.37.164; Davis 2002, pp. 15061), a work compiled around the eighth century ce. It tells of how iva was once wandering as an ascetic through a pine forest, when he sexually aroused and tempted the wives of a group of formidable sages. In retaliation, the sages cursed iva, causing his penis to fall to the ground. When it fell, it erupted into flames and threatened to destroy the entire universe. In order to appease iva and halt the impending doom, the sages began to worship the phallus; it is said that they were ordered to this fate as a punishment for their transgression against iva. According to this popular medieval tale, this was how liga worship first originated. Above, we noted how the phallic imagery of early ligas was transformed into more abstract forms that resonates with theological ideas about iva as distant, formless, and transcendent. By contrast, the Druvana myth explicitly presents the liga as the phallus of the god. As such, it may point to the reception and resonance of the older phallic iconography, even into the medieval period. That this myth was widespread, for instance, is suggested by the integration of versions into a number of different PurA{as.8 By contrast, the Ligodbhavamrti myth explains the mysterious origins of the liga with an emphasis on its cosmic significance. This story also occurs in various versions. Typically, it begins with a fight or argument between the gods BrahmA and Vi{u, each of whom claims to be the highest lord in the universe ( jagatprabh; Skanda Pura Then, between the two deities, a splendid column of fire or light suddenly appears; it is so enormous that its limits lie beyond their vision or perception. Vi{u, in the form of a boar, thus digs underground to search for the base of the column; BrahmA, in the form of a swan, attempts to fly to its apex. Both are unsuccessful, as the column is infinite without beginning or end (e.g., Fig. 6). Vi{u thus comes to realize that the column is a form of iva; he praises the god, recognizing him as the one who is truly the highest lord in the universe. BrahmA, by contrast, contrives a lie; he claims that he had, in fact, reached and seen the top of the column. iva himself immediately appears, and BrahmA is exposed as a fraud. As a result, the god is subordinated to iva.
2009 The Author Religion Compass 3/3 (2009): 440458, 10.1111/j.1749-8171.2009.00141.x Journal Compilation 2009 Blackwell Publishing Ltd

450 Benjamin J. Fleming

Fig. 6. A scene depicting the Ligodbhavamrti on the back of the main shrine at the Ramesvara temple in Tamil Nadu; ca. fourteenth century ce. The gods Viu and Brahma are depicted on either side of the central column. Photo, B. Fleming.

Whereas the Druvana myth recounts the origins of liga worship in relation to ivas body, the Ligodbhavamrti myth instead emphasizes its timeless and cosmic significance. The column of light is here described as ivas jyotirliga (liga of light; Skanda Pura, and it is said to have no beginning or end. The image of iva, in this myth, is that of a distant and abstract god. Different versions of the story describe the cosmic column in various ways; it is notable, however, that many emphasize that the liga is beyond the senses, thus evoking the view of iva as formless and without attributes. The Druvana and Ligodbhavamrti myths are famous to this day, and they seem to have been popular among medieval storytellers and writers across the Indian subcontinent as well. Whereas these traditions are pan-Indian and speak to the origins of the liga in a general sense, there are also many medieval story-traditions that focus on local landscapes, often celebrating the ligas attached to specific temples and religious centers by recounting their origins. In Sanskrit and vernacular sources from across the subcontinent, we find almost countless examples of tales about the fame, power, and benefits of individual ligas. Although likely oral in their ultimate origins, such stories are frequently collected into texts called mhtmyas, a term suggesting the special efficacy
2009 The Author Religion Compass 3/3 (2009): 440458, 10.1111/j.1749-8171.2009.00141.x Journal Compilation 2009 Blackwell Publishing Ltd

The Form and Formlessness of Siva


or virtue of god or sacred place. Over time, mhtmyas become woven together and integrated into larger compilations of mythic and pilgrimage traditions, such as the PurA{as. As a result, purA{ic literature often preserves a diverse cross-section of local traditions about sacred sites. This literature also provides important evidence for the role of liga worship in connecting different locales, by means of pilgrimage and storytelling alike. As early as the sixth century ce, for example, the early Skanda Pura lists a grouping of aivite pilgrimage sites, which apparently extended across the HimAlaya mountains (Bisschop 2006, pp. 1214). Other regional pilgrimage routes featuring ligas are richly attested in later PurA{as (see, passim, Feldhaus 2003, pp. 12756; Dyczkowski 2004, pp. 93174). Concurrent with the rise of regional pilgrimage routes, it seems that stories about ligas traveled throughout the subcontinent, contributing to the emergence of a relatively homogenized liturgy surrounding li ga worship. The variety in early forms of aivite ritual practice gave way to increased systematization and standardization, sometime after the seventh or eighth century ce. Gradually but progressively thereafter, liga worship would come to supersede other types of aivite ritual practice, and it became established as the dominant, trans-regional mode for worshipping iva (Younger 1995, pp. 8789, 1068; Fleming 2009). As liga worship became widespread in medieval times, hymns were composed and stories collected in praise of ligas not just locally and regionally but even at a pan-Indian level. One of the most dominant groupings of pan-Indian ligas are the 12 jyotirligas. This is perhaps the most wide-ranging set of ligas: one is located in the HimAlaya mountains, another on the southern tip of the subcontinent, while the others are spread in western, central, and eastern locales (Feldhaus 2003, pp. 12830; Fleming 2009). Here is a stotra (hymn of praise) about the 12 jyotirligas and their locations:
(The liga) SomanAtha is in SAurAra, and MallikArjuna is in r}aila. MahAkAla is in Ujjain; Amalevara is in OmkAra. In Parali is (the liga) VaidyanAtha and in the DAkini (Forest) is (the ligas) Bh}maakara. At the Bridge (to LAkA) is RAmea, and in the DArukA Forest is NAgea. In VArA{as} is (the liga) Vivea, while on the shores of the Gautam} (River), Tryambaka is (found). In the HimAlaya (mountains) there is Ked Ara, and at the ivAlaya (Pond), Ghumea is (found). The person who recites from memory these ligas of light ( jyotirligas), every morning at dawn, annihilates the sin accumulated during seven rebirths.9

This stotra has its origins in medieval times, but it is sung, to this day, by school-children, devotees, and priests alike (Fleming 2007, pp. 28 35, 2009) a poignant example of how local and regional traditions about ligas and liga worship eventually came to shape views of iva throughout India.
2009 The Author Religion Compass 3/3 (2009): 440458, 10.1111/j.1749-8171.2009.00141.x Journal Compilation 2009 Blackwell Publishing Ltd

452 Benjamin J. Fleming

Unity and Diversity within Saivism For understanding ligas and liga worship in modern Hinduism, medieval developments are arguably pivotal. It was then that widespread, pan-Indian aivite traditions emerged, linking a diverse complex of local sites and regional pilgrimage traditions together. By that time, the iconography of the earliest artifacts of the Katrapa and KuA{a periods had been displaced by the geometrically simple ligas found in thousands upon thousands of temples in India today. These cylindrical forms, as we have seen, have much in common with the theological conceptualization of iva developed by Brahmanical elites, which portray him as a distant and formless deity with little direct contact with the world. Even in medieval and modern times, however, diversity persists. Ligas today, for instance, are of various sizes, and they can be made of stone, clay, wood, metal, or other materials. Some are built for permanent installation, while others are moveable, employed in seasonal festivals or other specific rituals. Some moveable ligas are made from perishable materials like wood and unfired clay, and discarded at the end of a ritual in which they are used. One example is a practice at the marevara temple in the city of OmkAra in Madhya Pradesh. There, priests make hundreds of small clay ligas every single day and place them across seven large wooden palettes. After performing a series of rituals, the priests throw the clay ligas into the nearby NarmadA river. From purA{ic stories, it is clear that rituals of this sort have been practiced since at least medieval times.10 Some medieval Sanskrit texts describe the use of different materials to create ligas, even alluding to ligas made of gems or precious stones.11 In addition, in a passage from the Vidyevara-saMhit (19.78) of the Siva Pura tradition, different materials are correlated with the cosmic cycle of yugas (ages):
A liga consisting of gems is best in the Kta age, gold is best in TretA age, silver is best in the DvApara age, and clay is best in the Kali age.

This particular list may be an attempt to account for the varieties of types of ligas known when it was composed (ca. ten to twelfth centuries ce), and it may emphasize the importance of clay ligas, which are especially renowned in the Siva Pura tradition. Other sources associate the liga with bone. In a number of passages in the Skanda Pura tradition,12 for instance, skulls are either equated with ligas or are depicted as transformed into them. Some association of ligas with relics is also suggested by inscriptional evidence related to the medieval practice of entombing and celebrating the memory of ones teacher by erecting a liga (Bhandarkar 1931, pp. 18). Medieval prescriptions about the construction and installation of ligas have also influenced practices to this day. The instructional literature for making ligas for installation in temples is extensive, contained primarily
2009 The Author Religion Compass 3/3 (2009): 440458, 10.1111/j.1749-8171.2009.00141.x Journal Compilation 2009 Blackwell Publishing Ltd

The Form and Formlessness of Siva


in gamas but also in some PurA{as (Kramrisch 1946, pp. 23943, 41415). Ligas, as presented in these sources, are typically divided along the cylindrical column into three parts of equal size: a square base, a circular mid-section, and a rounded top. Each of these parts is further subdivided according to certain conditions, which can include the ligas placement in the temple as well as the shape and size of the temple in which the liga is to be installed. The base portion is typically set into the floor, while the mid-portion is either surrounded by stone slabs or set into a stone base (phik). The remaining, top section is left exposed for propitiation by worshippers. Only a third of the actual liga is above the ground. The differences in details between types of ligas are overwhelming in their variation and complexity. Indeed, to all except ritual specialists, those who compose and study the technical manuals, and perhaps the artisans who create these religious objects, the details are likely unknown. Nor is knowledge of such details necessary for participation in popular rituals involving ligas or for understanding the rich body of mythological, hymnic, and theological traditions about them. Each stylistic variation, however, operates according to an internal logic of equivalences between the ritual specialist, the liga, and the space in which the liga is housed. In this sense, the liga is meant to be understood as a microcosm of the universe itself. One of the most dominant types of ligas, for example, is called a mnua-liga (lit. the liga derived from [or made by] man); this type is typically proportioned in direct correspondence to the physical proportions of sacrificer or priest who oversees the construction of the liga. Often, a liga may be proportioned in relation to the shrine in which it is to be housed, taking into account the length of the shrine, the height of the doorway, and so on. In turn, the shrines themselves are constructed according to specific rules laid out in the gamas. Their size and shape are chosen in relationship to cardinal directions and to geographical formations, such as mountains and rivers. Accordingly, the ligas therein are in alignment with the macrocosmic universe. Conclusion It is impossible to do justice to the richness of the literary, theological, and iconographic traditions surrounding ligas in one brief article. Above, I examined some of the most prominent traditions as well as some of the major issues of scholarly discussion and debate. I also attempted to point to some persistent dynamics and dichotomies, found in both material and literary evidence, and to highlight some of the continuities and discontinuities between ancient, medieval, and modern traditions about ligas. It is important to stress, however, that these represent only a small selection of the many and varied traditions about ligas in India and beyond. Our survey of textual materials, for instance, was limited to
2009 The Author Religion Compass 3/3 (2009): 440458, 10.1111/j.1749-8171.2009.00141.x Journal Compilation 2009 Blackwell Publishing Ltd

454 Benjamin J. Fleming

Sanskrit literature, and our survey of historical developments focused on pan-Indian developments. Much more could be said about the practices, ideas, and developments distinct to individual cities and regions, and expressed in vernacular languages. Likewise, space did not permit a consideration of the views of ligas specific to certain groups, such as the Ligayats or V}raaivas; in contrast to the pro-Vedic tendencies noted above, this South Indian aivite sect wear small ligas around their necks and are traditionally anti-Brahmanical and anti-Vedic (Rao 1990, pp. 37). Tantric literature also offers its own broad and rich array of ritual traditions focusing on liga worship; at times, moreover, tantric ritual has had a strong and dynamic influence on the more mainstream practices portrayed in the PurA{as (e.g., akti traditions). Above, we noted how the PurA{as preserve traditions about myth, pilgrimage, and sacred geography that may help us to understand the rise of a pan-Indian aivism focused on the liga. Tantric traditions of liga worship may be similarly central to the spread of aivism outside of the Indian subcontinent, particularly into Nepal and Southeast Asia (Sanderson 20032004). Although the precise origins of the liga remain a mystery, the diverse traditions about this enigmatic representation of iva make clear its enduring power to inspire stories, art, pilgrimage, theological reflection, and vigorous debate throughout the centuries. Short Biography Benjamin J. Flemings research focuses on aivite ritual, myth, and iconography, with a particular concern for traditions about pilgrimage and sacred geography. He holds a BFA, BA, and MA, from the University of Regina and a PhD from McMaster University. His dissertation, Cult of the Jyotirligas and the History of aivite Worship (2007), investigated the relationship between ritual, storytelling, and pilgrimage in medieval aivism. His publications include an article, Mapping Sacred Geography in Medieval India: The Case of the 12 Jyotirligas, in the International Journal of Hindu Studies. He has presented papers on PurA{as and on inscriptional materials at the annual meetings of the American Academy of Religion, American Oriental Society, and Canadian Society for the Study of Religion, as well as at the Fourth International Vedic Workshop, Oriental Club of Philadelphia, and the Humanities Forum and Religious Studies Colloquium at the University of Pennsylvania. He has been awarded grants and fellowships from the Shastri Indo-Canadian Institute and the Social Sciences and Humanities Council of Canada. He is presently a Mellon Postdoctoral Fellow in the Department of Religious Studies at the University of Pennsylvania, where he teaches courses on Hinduism. Notes
* Correspondence address: Dr Benjamin J. Fleming, University of Pennsylvania, 249 S. 36th Street, Cohen Hall #234, Philadelphia, PA 19104, USA. E-mail:
2009 The Author Religion Compass 3/3 (2009): 440458, 10.1111/j.1749-8171.2009.00141.x Journal Compilation 2009 Blackwell Publishing Ltd

The Form and Formlessness of Siva


1 All translations in the article are those of the author. I am grateful to the following for editorial comments, suggestions, and references: Annette Yoshiko Reed, John C. Huntington, Richard Mann, Steven R. Reed, and Shayne Clarke. I would also like to thank my anonymous reader for his/her thoughtful reading of the article. Research for this article was supported by the Andrew W. Mellon Foundation, the Social Science and Humanities Research Council of Canada, and the Shastri Indo-Canadian Institute. 2 E.g., Liga Pura 1.28.1314; Siva Pura, Jna-saMhit 49.315. For references in various gamas see Davis 1991, pp. 11314, 12122. 3 For the rich complex of traditions about iva in VArA{as}, see e.g., Eck 1983; Bakker 1996; Fleming 2009. 4 The impulse to understand the liga in terms of the gods formlessness is also evident in a popular etymology of the term, which links it with fire; see Tagare 2002, pp. 2021. 5 See also rinavasan 1997, p. 227, for an attempt to situate this work in a continuum of early pilgrimage routes cited in the Mahbhrata. 6 A similar discomfort is evident in modern Hinduism, particularly in elite scholarly and theological discourse (see, e.g., Tagare 2001, p. 23). This long-standing tendency within the tradition may also play a part in critiques of the fixation on sexual imagery in some Western scholarship on iva and the liga. 7 Mann (2007, p. 734) notes a similar process with Skanda, a figure identified in medieval sources as the son of Siva: A central means the Brahmanical tradition had of drawing Skanda into the orthodox fold was to link him with Vedic ideas and figures. See further Mann 2003. 8 On this myth cycle, see further Doniger 1980, pp. 13754; Handelman & Shulman 2004; Davis 2002, pp. 15052; Granoff 2003; Filliozat 1961, pp. 28994. 9 Stotraratnval, Sivastotra, 1976. 10 E.g., Siva Pura: Jna-saMhit 53.5660; 58.3739; Koirudra-saMhit 25.5356; 33.4448. 11 E.g., Siva Pura, Koirudra-saMhit 17.35; MAnasAra 52.32833 (52.15255); see Acharya 1934a, p. 541; 1934b, pp. 34950. 12 E.g., Skanda Pura, Avanti-khaa: Avantiketra-mhtmya 23.4; Caturatiliga-mhtmya 8.30 31.

1. primary sources
Acharya, PK, (ed.) 1934a, Manasara on Architecture and Sculpture: Sanskrit Text with Critical Notes. Repr. ed. Oriental Books Reprint Corp, New Delhi. Adriaensen, R, Bakker, HT & Isaacson, H, et al. (eds.) 1998present [ongoing], The Skandapura. Egbert Forsten, Groningen. BhaarA}, K, (ed.) 1988, Skandapurasya Ambikka. Mahendraratnagrantha-mAlA, 2. Kathmandu. Cds, G, (eds.) 193766, Inscriptions du Cambodge. 8 vols. EFEO, Paris. Dhal, UN, (ed.) 2000, Siva Pura (Uttarakhaa). Nag Publishers, Delhi. Jyotirligamnasa-pjastotraM. 1941, Nirmayasagar Press, Bombay. Kumar, P, (ed.) 1996, Sivamahpura. 2 vols. Nag Publishers, Delhi. Jna-saMhit, n.d Bhandarkar Oriental Research Institute Library. Library cat. no. 61352. arnAcarya, S, (ed.) 1966, Siva Pura. Samskti SamsthAna, Uttara Pradea. Shastri, JL, (ed.) 1980, Liga Pura. Motilala Banarasidasa, Delhi. Siva Pura, 1884, Ga{apatik{Ajis Press, Bombay. Siva Pura, 189596, Vekaevara Press, Bombay. Siva Pura, 1908, Va{gavAs} Press, Calcutta. Skanda Pura, 1962, Mansukharaya Mora, Calcutta. Vaidya, PL, (ed.) 196768, Mahbhrata. Bhandarkar Oriental Research Insititute, Pune.

2. english translations
Acharya, PK, (trans.) 1934b, Architecture of Manasara. Repr. ed.; Oriental Books Reprint Corp, Delhi.
2009 The Author Religion Compass 3/3 (2009): 440458, 10.1111/j.1749-8171.2009.00141.x Journal Compilation 2009 Blackwell Publishing Ltd

456 Benjamin J. Fleming

Doniger (OFlaherty), W, (trans.) 1980, Hindu Myths: A Sourcebook translated from the Sanskrit. Penguin, Middlesex, UK. Roy, PC, (trans.) 1970, Mahbhrata. Munshiram Manoharlal, New Delhi. Tagare, GV, (trans.) 1992present [ongoing], The Skanda Pura. Ancient Indian Tradition & Mythology, 4966. Motilal Banarsidas, Delhi. The Siva Pura, 1969, translated by a Board of Scholars. Ancient Indian Tradition & Mythology, 1 4. Motilal Banarsidas, Delhi. The Liga Pura, 1998, translated by a Board of Scholars. Ancient Indian Tradition & Mythology, 5 6. Motilal Banarsidas, Delhi.

3. secondary literature
Bakker, HT, 1996, Construction and Reconstruction of Sacred Space in VArA{as}, Numen, vol. 43, no. 1, pp. 3255. , 1997, The Vkakas. Egbert Forsten, Groningen. , 1999, Review of D. rinivasan, Many Heads, Arms and Eyes, Artibus Asiae, Vol. 58, pp. 33943. , 2001, Sources for Reconstructing Ancient Forms of iva Worship, in Les Sources et le Temps Sources and Time: A Colloquium, Pondicherry, 1113 January 1997, F Grimal, et al. (eds.), pp. 397419. Institut Franais de Pondicherry, Pondicherry, India. Banerjea, JN, 1935, The Phallic Emblem in Ancient and Mediaeval India, Journal of the Indian Society of Oriental Art, vol. 3, pp. 3644. , 1956, The Development of Hindu Iconography. 2nd revised ed. University of Calcutta Press, Calcutta. Bhandarkar, DR, 1931, Mathura Pillar Inscription of Chandragupta II, Epigraphia Indica, vol. 21, no. 1, pp. 19. Bisschop, P, 2006, Early Saivism and the Skandapurana: Sects and Centres. Egbert Forsten, Groningen. Coomaraswamy, AK, 192831, Yakas. 2 vols. Smithsonian Institution, Washington D.C. Davis, R, 1991, Ritual in an Oscillating Universe: Worshiping Siva in Medieval India. Princeton University Press, Princeton. , 2002, The Origin of Liga Worship, in Religions of Asia in Practice, D Lopez (ed.), pp. 15061. Princeton University Press, Princeton. , 1999, Lives of Indian Images. Princeton University Press, Princeton. Doniger (OFlaherty), W, 1973, Asceticism and Eroticism in the Mythology of Siva. Oxford University Press, London. [re-published in 1981 under the title Siva the Erotic Ascetic] Dyczkowski, MS, 2004, A Journey in the World of the Tantra. Indica Books, New Delhi. Eck, D, 1983, Banaras, City of Light. Penguin Books India, New Delhi. Falk, H, 1994, Von Gtterfiguren und menschlichen Gttern, in Festschrift Klaus Bruhn zur Vollendung des 65, N Balbir & J Bautze (eds.), pp. 31333. Verlag fr Orientalistische Fachpublikationen, Reinbek, Germany. Feldhaus, A, 2003, Connected Places: Region, Pilgrimage, and Geographical Imagination in India. Palgrave Macmillan, New York. Filliozat, J, 1961, Les images de iva dans lInde du sud: Limage de lorigine du liga (Ligodbhavamurti), Arts Asiatiques, vol. 8, no. 1, pp. 4256. Fleming, BJ, 2007, Cult of the Jyotirligas and the History of Saivite Worship. Unpubl. PhD dissertation, McMaster University, Hamilton, ON, Canada. , 2009, Mapping Sacred Geography in Medieval India: The Case of the 12 Jyotirligas, International Journal of Hindu Studies, vol. 13, no. 1, pp. 5181. Goldberg, E, 2008, ArdhanArevara: What We Know and What We Do Not, Religion Compass, vol. 2, no. 3, pp. 30115. Granoff, P, 1993, Halayudhas Prism: Temples and Worship in Medieval North India, in Gods, Guardians and Lovers: Temple Sculptures from North India, V Desai & D Mason (eds.), pp. 6793. Asia Society, New York. , 2003, MahAkAlas Journey from Ga{a to God, Revista degli Studie Orientali, vol. 77, pp. 135.
2009 The Author Religion Compass 3/3 (2009): 440458, 10.1111/j.1749-8171.2009.00141.x Journal Compilation 2009 Blackwell Publishing Ltd

The Form and Formlessness of Siva


, 2004, Saving the Saviour: iva and the Vai{ava AvatAras in the Early SkandapurA{a, in Origin and Growth of the Puranic Text Corpus: With Special Reference to the Skandapurana, H. Bakker (ed.), pp. 11139. Motilal Banarsidas, Delhi. , 2006, iva and his Ga{as: Techniques of Narrative Distancing in the PurA{as. Festschrift U. Dhal. University Bhuvanevar, Bhuvanevar, India. Handelman, D & Shulman, D, 2004, Siva in the Forest of Pines. Oxford University Press, Oxford. Joshi, NP, 1984, Early Forms of iva, in Discourses on Siva: Proceedings of a Symposium on the Nature of Religious Imagery, M Meister (ed.), pp. 4761. University of Pennsylvania Press, Philadelphia, PA. Kramrisch, S, 1946, The Hindu Temple, vol. 1. University of Calcutta, Calcutta. , 1981, The Presence of Siva. Princeton University Press, Princeton. Kreisel, G, 1986, Die Siva-Bildwerke der Mathur-Kunst. Fran Steiner Verlag Wiesbaden GMBH, Stuttgart. Mann, R, 2003, The Early Cult of Skanda in North India: From Demon to Divine Son. Unpubl. Ph.D. dissertation, McMaster University, Hamilton, ON, Canada. , 2007, Skanda in Epic and Puranic Literature: An Examination of the Origins and Development of a Hindu Deity in North India, Religion Compass, vol. 1, no. 6, pp. 725 51. Rao, TAG, 1914, Elements of Hindu Iconography, vol. 2, no. 1. The Law Printing House, Madras, India. Rao, VN, 1990, Introduction, in Sivas Warriors: The Basava Pura of Plkuriki Somantha, pp. 338. Princeton University Press, Princeton. Sanderson, A, 20032004, The aiva Religion among the Khmers, Part I, Bulletin de lecole franaise detrme-orient, vol. 9091, pp. 349462. Sarma, IK, 1994, Paraurmevara Temple at Guimallam. Dattsons, Nagpur, India. rinavasan, D, 1978, The Religious Significance of Divine Multiple Body Parts in the Atharva Veda, Numen, vol. 25, no. 3, pp. 20915. , 1997, Many Heads, Arms and Eyes: Origin, Meaning and Form of Multiplicity in Indian Art. Brill, Leiden, the Netherlands. von Mitterwallner, G, 1984, Evolution of the Liga, in Discourses on Siva: Proceedings of a Symposium on the Nature of Religious Imagery, M Meister (ed.), pp. 1231. University of Pennsylvania Press, Philadelphia, PA. Waghorne, JP & Culter N, (eds.) 1984, Gods of Flesh/Gods of Stone: The Embodiment of the Divine in India. Columbia University Press, New York. Yamaguchi, S, 2008, The Twelve Jyotirlinga Temples in Modern Hinduism, Journal of Indian and Buddhist Studies, vol. 57, no. 1, pp. 26875. Younger, P, 1995, The Home of Dancing Siva: The Traditions of the Hindu Temple in Citamparam. Oxford University Press, New York.

4. further reading
Acharia, PK, 1946, An Encyclopaedia of Hindu Architecture. Oxford University Press, London. Agrawal, RC, 1965, Four-faced iva and Four-Faced Vi{u at Mathura, Vishvesharand Indological Journal, vol. 3, no. 1, pp. 10710. , 197273, Four-faced iva-Ligas in National Museum, Journal of the Oriental Institute of Baroda, vol. 22, pp. 36568. Bakker, HT, 2008, Mansar: The Discovery of Pravarevara and Pravarapura Temple and Residence of the Vkaka King Pravarasena II. Proceedings of a Symposium at the British Museum (London, 30 JuneJuly). Library of the University of Groningen, Groningen. Brahm PuraM, 1954, Gurumandal series, 11. Mansukharaya Mora, Calcutta. Collins, CD, 1988, The Iconography and Ritual of Siva at Elephanta. State University of New York Press, New York. Dyczkowski, MS, 1988, The Canon of the Saivgama and the Kubjik Tantras of the Western Kaula Tradition. State University of New York Press, Albany. Goudriaan, T, 1992, Ritual and Speculation in Early Tantrism: Studies in Honor of Andre Padoux. State University of New York Press, Albany.
2009 The Author Religion Compass 3/3 (2009): 440458, 10.1111/j.1749-8171.2009.00141.x Journal Compilation 2009 Blackwell Publishing Ltd

458 Benjamin J. Fleming

Granoff, P, 1998, RAmas Bridge: Some Notes on Place in Medieval India, Real and Envisioned, East and West, vol. 48, pp. 9396. Handelman, D & Shulman, D, 1997, God Inside-Out. Oxford University Press, New York. Kramrisch, S, 1984, The Great Cave Temple of iva in Elephanta, in Discourses on Siva: Proceedings of a Symposium on the Nature of Religious Imagery, M Meister (ed.), pp. 44368. University of Pennsylvania Press, Philadelphia, PA. Lorenzen, DN, 1972, The Kplikas and Klmukhas: Two Lost aivite Sects. University of California Press, Berkeley Meister, M, (ed.) 1984, Discourses on Siva: Proceedings of a Symposium on the Nature of Religious Imagery. University of Pennsylvania Press, Philadelphia. Reddy, YG, 1972, The Image of Lingodbhavamurthi at Panugal, Journal of Indian History, vol. 50, no. 1, pp. 914. Scheuer, J, 1982, Siva dans le Mahbhrata. Presses Universitaires de France, Paris. Shulman, D, 1986, Terror of Symbols and Symbols of Terror: Notes on the Myth of iva as SthA{u, History of Religion, vol. 28, no. 2, pp. 10124. , 1993, The Hungry God: Hindu Tales of Filicide and Devotion. University of Chicago Press, Chicago. Smith, TL, 2007, The sacred center and its peripheries: aivism and the Vras Sthala-Puras. Unpubl. Ph.D. dissertation, Columbia University, New York. Vogel, JP, 1930, La sculpture de Mathur. Les ditions G. van Oest, Paris.

2009 The Author Religion Compass 3/3 (2009): 440458, 10.1111/j.1749-8171.2009.00141.x Journal Compilation 2009 Blackwell Publishing Ltd