A*sOrfeKfcW&  ^/l { ľW^iJH-i^

ANICONISM AND THE MULTIVALENCE OF EMBLEMS: ANOTHER LOOK
By SUSAN L. HUNTINGTON
[The author recommends that prior to reading this essay the interested individual read Susan L. Huntington, "Early Buddhist Art and the Theory of Aniconism," Artjournal 49, no. 4 (Winter 1990): 401-8; the letter to the editor by Michael Rabe in response to that article and Susan L. Huntington's reply to the letter in Art Journal 51, no. 1 (Spring 1992): 125-27; and then Vidya Dehejia's "Aniconism and the Multivalence of Emblems," Ars Orientalis21 (1991): 45-66. See also Susan L. Huntington, with contributions by John C. Huntington, The Art of Ancient India (New York and Tokyo: John Weatherhill, Inc., 1985),70-71,72-73,87,98-99,100.]
Introduction
At the invitation of the editors of Ars Orientaus, I am responding to the article by Vidya Dehejia entitled "Aniconism and the Multivalence of Emblems" published in Ars Orientalis, volume 21.' The stated purpose of Dehejia's article is to demonstrate and advocate that early Buddhist art was multivalent, thatjs, imbued with multiple meanings that could be conveyed by individual emblems (45). A second, unstated purpose of the article is to challenge some of the ideas I have presented regarding the long-held theory of aniconism thathas been used to interpret these early Buddhist materials for more than a century. At the same time that Dehejia challenges aspects of my work, she incorporates into her multivalency theory one of the most important contributions of mý analysis—my explanation of a certain type of composition as focusing on Buddhist sacred sites. Although her multivalency theory upholds the abiding belief in the aniconic theory, her interpretation of early Buddhist art is tempered by accepting my proposal that no t all of the works of art that have been interpreted as aniconic images are aniconic.
I agreed to write this response for two reasons. First, although Dehejia's theory of multivalency offers an interesting and clever amalgam of the traditional aniconic theory and my own ideas and seeks to move the interpretation of early Buddhist art in a new direction, I am not convinced that multivalency, as she has presented the case, is a viable alternative to the previously proposed
viewpoints. Second, I wish to respond to her criticisms of my work. Because I examine the theory of aniconism and related iconographical, historical, social, religious, and cultural issues in my forthcoming book on the early Buddhist art of India, I will not use this forum to present the extensive materials I have collected. Rather, here I will deal specifically with problems in the way my work has been interpreted and applied in Dehejia's article. Although I believe that Dehejia's reactions to my research are premature since my full study has not yet been published,2 it is important at this stage to correct misunderstandings and misrepresentations of my work. Further, since in a number of instances Dehejia actually argues for my viewpoint without acknowledging her indebtedness to my work, I welcome the opportunity to clarify the derivation of some of her interpretations.
Let me state at the outset that I have never claimed nor intended to claim that there are no "aniconic" works of art. In fact, I specifically allow for the possibility.3 My position is that the theory of aniconism is not valid as an all-inclusive explanation for the early Buddhist art of India and that the vast majority of artistic compositions that have been explained as aniconic scenes are not substitutes and do not portray substitutes for anthropomorphic representations of a Buddha. What I propose is that the explanatory power of the aniconic theory has been vastly overestimated and that the theory has been indiscriminately applied, much to the detriment of our understanding of early Buddhist art and its religious meanings and cultural contexts. My book will propose a series of new generalizations based on patterns of evidence, radier than anew absolutist theory. In light of what will be my reinterpreta-tion of the vast majority of reliefs, I believe that the monuments and the practices of early Buddhism will also require renewed study.
To clarify to readers where my work on early Buddhist art has been published, I first questioned the validity of the theory of aniconism as an all-embracing explanation for the art of early Buddhism in my Art of Ancient India.* In that context, I was not able to present more than a few
1*
J («I
jl
112
SUSAN L. HUNTINGTON
basic components of my larger research project since the theme of the Art of Ancient Indiavras the development of the art of the Indie world over many centuries, and a lengthy digression on die problem of aniconism would have been a detour from the main theme of the book. I also published an article in the Artjournal'm which I chose one specific type of image to discuss. I wish to emphasize "one specific type" because Dehejia has apparently misunderstood that article to represent a comprehensive, rather than selective, presentation of my ideas regarding the scenes appearing in early Buddhist art, an important pointto which I shall return. Since thatarticle was solicited to appear in an issue of the Art Journal devoted to the presentation of current research in the field, the paper had to fulfill certain editorial criteria that applied to all audiors whose work had been selected for publication. Specifically, I was limited in terms of the number of words and illustrations, which had to conform to die space allotted to the other authors, and I was asked to write the article in a way that would interest readers outside the specialized field of South Asian art. Therefore, for the Art Journal article I chose to present a component of my study that could stand alone. Interestingly, diis particular component of my larger study—the identification of reliefs that represent worship scenes at sacred Buddhist sites—is one that Dehejia adopts in her article (without, however, acknowledging me as the source of the notion).
In addition to these two publications, I have made a number of oral presentations on aspects of my work. These include papers at the 1988 and 1991 conferences sponsored by the American Committee for South Asian Art. These papers were stricdy limited in length and could only address very narrow aspects of my study; thus, some of Dehejia's misunderstandings of my research may have arisen from diese presentations as well.* While the 1988 talk emphasized the materials ultimately published in the Art Journal article, the 1991 presentation specifically targeted the historiography of the theory of aniconism, a subject that I believe is intimately related to die passionate advocacy of the theory diat my critics so commonly display. I have given lengthier presentations of my work at the conference of the International Association for Buddhist Studies in Paris during summer 1991 and in invited talks at Harvard University's Center for the Advanced Study of World Religions in spring 1990 and at
the University of Chicago in spring 1987, to name just a few.6
Critique ofDehejia's Article: Structural and Theoretical
MULTTVALENCY
The stated purpose of Dehejia's article is to demonstrate and advocate that early Buddhist art was imbued with multiple, simultaneous meanings that could be conveyed by individual emblems (45). The theoretical concept of multiva-lency is one that I not only endorse but applaud, as attested by my introduction of the term slesa, which Dehejia uses to explain the notion of mul-tivalency, in my Art of Ancient India (1985) ? The idea of multiple meanings in works of art is a concept diat I have published elsewhere, lectured about in public, and extensively incorporate into my teaching, as my students are well aware. John C. Huntington has also used the concept of multivalency to explain works of art. His fordicoming article entided "The Iconography of Barabudur Revisited: The Concept of Šlesa (Multivalent Symbology) and the Sarva-[buddha] käya as Applied to the Remaining Problems" has been in press since 1988, evidencing his subscription to the notion for at least the past five years.8 Further, he explains the term sleßa in detail in his article entitled "Pilgrimage as Image," which Dehejia cites in her Ars Orientalis essay for other purposes.9
Although I have advocated that multiple meanings are simultaneously inherent in many works of Indie art, I do not believe that the theory of multivalency that Dehejia proposes offers a viable alternative either to the traditional theory of aniconism or to die ideas diat I am presenting in my own work diat challenge die aniconic theory. A careful reading of her article reveals that the logic she has used in applying this principle to the early Buddhist artistic remains undermines the credibility of her scheme. She reasons in her first paragraph diat if the technique of double meanings was employed by writers, the concept would have been known to visual artists and employed by them as well.10 While this is possible, i t cannot be automatically assumed that there is a positive correlation between different media, such as literature and die visual arts, in every time and place. With regard to the early Buddhist
ANICONISM AND THE MULTIVALENCE OF EMBLEMS: ANOTHER LOOK
US
materials one must ask: Were literary techniques that employed double entendre or other multiplicities of meaning in vogue (or even in use) at the time the art monuments under discussion were created? And, if they were, can we document linkages between the literary and visual arts during that period? If Dehejia's multivalent interpretations are predicated upon an analogy widi literary techniques, a definite correlation must be found between the media at the time. If she is using the literary analogy to suggest a kind of generalized cultural milieu, then it is necessary that she unequivocally demonstrate the presence of multivalency within the art works themselves. My analysis of her article demonstrates that in a majority of cases (if not all), her proofs of multiple meanings are speculative or, if verifiable, do not apply to the degree and in the manner she suggests. Further, even if her interpretation of multivalency could be shown to be valid—and I suggest that it is not—her assertion regarding its importance for the theory of aniconism seems gready exaggerated. Her claim that "As soon as we accept the validity of such a system [of multivalency] ... aniconism ceases to be such a vexed problem" (47) vastly oversimplifies the complex religious, social^ cultural, and artistic issues surrounding the theory of aniconism and, at the same time, fails to offer a new, comprehensive explanation of what is seen—or not seen—in the art.
My chief concern regarding Dehejia's application of the theory of multivalency is that she does not distinguish deliberately intended multiple meanings from those that are naturally inherent in certain subjects. If my reading of Dehejia's article is correct, she is suggesting diat the creators of monuments, such as the stúpaš at Säňci, Bhärhut, and other early Buddhist sites, deliberately tried to invest the sculptural compositions with multiple layers of meaning. I contend that the compositions are intended to show a single, principal meaning but sometimes intrinsically carry with them additional layers of meaning as well.
To use a literary analogy, it is easily observed that a given word might have a variety of meanings. A writer can employ a word that has multiple meanings to convey only one meaning, as in most scientific writing, or can use the word to suggest more than one meaning, as in punning, off-color jokes, poetry, and a variety of other literary forms. The specific characteristics of some languages
have enabled writers to master the art of double entendre and create highly sophisticated literary forms. Sanskrit authors, inspired by the peculiarities of the Sanskrit language that facilitated the creation of simultaneous meanings, developed the literary mode of slesa, a deliberate literary mode that was predicated on the singular characteristics of the Sanskrit language. Specifically, the rules of Sanskrit grammar require the coalescence of certain sound combinations, with the result diat Sanskrit passages are often written continuously, without spaces between words. To give an example in English of how this can be done, one might use the sequence of letters a-t-o-n-e. Depending upon how one breaks up the sequence, one can read the letters as "a tone," "at one," or "atone." Thus, it is easy to see how writers could take advantage of the language to imbue their works with multiple, simultaneous meanings.11 Similarly, punning could be accomplished simply by choosing words that have alternative meanings.12 The word hide can mean "the skin of an animal," "to put out of view," "to give a beating to," and is also an old English unit of land area.13
It is important to establish, however, that the potential of a word (or, in Sanskrit, a sequence of                       *"
letters) to convey more than one meaning does                        J
not prove that an author who uses that word intends diat more dian one of them apply to a                        ^
particular situation. The word Springern be used                        ^
by a weather forecaster to refer to the season following winter and by a mattress sales clerk to                        m
refer to the support under a mattress, but to                       m
conclude diat by using the word spring each                        '1
individual is referring to the variant meanings is                       ^
simply incorrect.                                                                          J
What I believe Dehejia is suggesting about what she calls emblems (such as trees, pillars, and stüpas in the art of Säňcí, Bhärhut, and other sites) is                         1
that inherently they may have more than one                         1(
meaning and that these meanings are manifest in early Buddhist art in two ways. These two ways constitute what she calls her "two prerequisites." First, she suggests that individual "emblems" are imbued with multiple meanings and function in different contexts with different meanings (45, para. 2 and 3), but she uses neither internal evidence in the reliefs nor external evidence from Buddhological or other sources to verify that the same motif might be intended to convey alternative meanings. Second, she suggests that in a single composition, an "emblem"
114
SUSAN L. HUNTINGTON
might communicate more than one layer of meaning (45, para. 4).14 This assumes that the individual responsible for the selection of the subject matter in an artistic composition deliberately intended to communicate simultaneous, multiple meanings and apparendy intended that the alternatíve meanings be given equal or relatively equal emphasis.15 However, my research suggests that while additional layers of meanings might be implicit for some subjects, the compositions of the early Buddhist artistic repertoire were intended to focus on a principal meaning. I contend that while a representation of lay worshipers performing devotions at the site of the Buddha's first sermon is undoubtedly a reminder of the first sermon itself and a reference to it, the actual subject of the scene is still the worship at the site. In other words, I distinguish among a representation, a reminder, and a reference, while Dehejia seems to conflate these discrete notions.
To give an example from Western culture, a representation of a Christmas tree on a Christmas card is not a representation of the birdi of Christ, which is the main event celebrated by Christmas, although it may implicitly refer to it and be a reminder of it. But it also might not implicitly refer to the birth of Christ either: some people— for example, the many American Jews who install Christmas trees in their homes and businesses and send Christmas greeting cards—do not intend that the tree refer to the birth of Christ. Indeed, such individuals commonly shrink from the notion of associating their Christmas greetings with the event of Christ's birdi but instead intend to communicate to their family, friends, and associates simply a celebration of the season and good will.
Thus, a multivalency theory should not presume that the same motif (or "emblem") conveys the same meanings throughout its history and in all of its usages. New layers of meanings may have been added over the course of centuries; some may have pertained to certain populations only or may have been used only in certain regions, and others may have been employed in otherwise defined contexts. Without specific study one can-notassume that a motif with potentially multiple layers of meanings had all of those meanings at every period and in every instance of its usage. The task is to study the specific situation to determine which meanings might have been present and which were incorporated into the work by the artists. When I have applied the concept of
multivalency to art, I have looked for specific proof that in a given case the artistic intention was to present a duality or a multiplicity of meanings simultaneously.16 My suggestion that the Märavijaya image in Päla art might have served as a dual image referring to the Päla kings is supported by inscriptional evidence that indicates a widespread convention establishing a metaphor-ic relationship between the Päla kings and the image type.17 Further, I believe that there is generally, if not always, a principal meaning and that the other meanings are additional. Thus, in my discussion of the popularly portrayed Märavijaya as a dual image referring also to the Päla kings in Päla art, I have no intention of suggesting that the reference to the Päla kings is the principal or even an equivalent subject of the art but rather an additional layer of meaning that has been grafted onto the basic Buddhist image. Nor do I intend to suggest that the same subject, the Märavijaya, appearing in theartofotherregions—even those heavily dependent upon the Päla tradition, such as Tibet—was intended to communicate the allusion to the Päla kings.
Dehejia's interpretation of multiple meanings has led her to apply the principle of multivalency broadly, without proving that the principle is applicable to the cases she cites. Her interpretation fails to discriminate between primary and additional layers of meaning and to distinguish among representations, references, and reminders. Furthermore, in her discussion of emblems, she does not explain the sources for the various interpretations she provides for them, nor does she supply the reader with a means of interpreting a motif in a given context. I shall discuss these ideas in more detail below.
"Emblems"
A key term for Dehejia—used both in the tide of her article and throughout her presentation—is emblem.19 The concept of the emblem is the foundation stone of Dehejia's theory of multivalency. However, I believe that the most fundamental flaw in the multivalency theory is the assumption that certain motifs or subjects in the early Buddhist art of India necessarily function as emblems.
Dehejia defines an emblem as "a picture that represents something different from itselP (45). By using the term emblem, she assumes that what she calls "the major Buddhist emblems" of the tree, the pillar, and the stüpa (and, I assume,
ANICONISM AND THE MULTIVALENCE OF EMBLEMS: ANOTHER LOOK
115
some of the other subjects she addresses in her article) invariably represent something other than themselves. For traditional aniconists, this other subject is believed to constitute the "real" subject of the scene, namely, an anthropomorphic depiction of the Buddha that is absent from the composition. Dehejia modifies the traditional aniconic view and suggests that "emblems" might also refer to "sacred sites" and "attributes" of the Buddhist faith. As with proponents of the aniconic theory before her, Dehejia's list of emblems includes only nonfigurative motifs, although, theoretically, there is no reason that a human figure or an animal cannot also serve as an emblem since the only requirement of "emblem-ness" is that something stand for some thing other than itself.19
Dehejia's discussion of emblems is unclear because of the ambiguity with which the relevant emblems are named. Her generalization about multiple meanings is specifically applied to the tree, the pillar, and the stüpa, which she terms "the major Buddhist emblems" (45). However, it is not explained whether this triad is named to exemplify a variety of emblems or to stand as "the three major" emblems in early Buddhist art. If the former, this is'riot specified; if die latter, evidence to verify diat the tree, pillar, and stüpa are the major emblems is not provided. Therefore, the basis upon which the tree, the pillar, and the stüpa are prioritized over other Buddhist subjects, such as the cakra, the footprints, and the throne, is unclear. No literary, inscriptional, or other proof of this prioritization is provided, and it is unlikely that Buddhists and scholars of Buddhism would all agree that these three would be paramount among the possibilities in the Buddhist repertoire. Furthermore, Dehejia does not specify which tree she considers to be one of the three major emblems, nor does this issue become clear in her subsequent discussions: sometimes "the tree" clearly refers to the bodhi tree, but in other contexts Dehejia broadens die category to include other sacred trees in Buddhism.20
Regardless, I do not understand why nonfigurative elements in the compositions, like trees, pillars, and stüpas, might not simply represent themselves. If I were to go to Bodh Gayä today, I would find there a bodhi tree and a temple. These were not installed at die site as substitutes for images of the Buddha but are important Buddho-logical motifs in their own right. If I were to take a picture of the bodhi tree at Bodh Gayä and show
it to my friends and relatives at home, I would not say that this is a photograph of a tree that represents the Buddha but that this is a photograph of the tree (or the descendant of the tree) under which the Buddhaattained enlightenment. Therefore, this particular tree is a reminder of the event in the Buddha's life and a reference to it, but it is not a substitute for the event, nor is it a representation of the Buddha. If anything, it is a participant in the event that still remains at the site. Specifically, the tree marks the location of the Buddha's "sitting place" (pitha) but does not emblematically depict the Buddha himself. Nor, I argue, is it merely an indicator of the site or simply an "attribute of the faith."21 While it may suggest all of these things, the tree embodies its own meanings as well. Further, as I hope to demonstrate in diis article, I suggest that it is incorrect to look at a scene that might contain numerous figures and activities along with amotif like a tree or a throne and define the subject matter of the composition according to thatmotif alone.
Dehejia's own illustrations and analysis include a number of examples that illustrate my viewpoint. One relief depicts the story of the serpent king Erapattra (fig. I).22 Dehejia claims that the seat and the garlanded tree at the lower left indicate the emblematic presence of the Buddha, before whom the serpent king kneels in obeisance
fpj^nF^Tip-pi
Fig. 1 (Dehejia fig. 3). Story of the Serpent King
Erapattra. From Bhärhut, Madhya Pradesh, India.
Ca. 100-80 B.C.E. Indian Museum, Calcutta.
Photo: Archaeological Survey of India.
116
SUSAN L. HUNTINGTON

-*-* I
m
BEHŕ^TE^w
Fíc. 2 (Dehejia fig. 2). Pageantry Scene? On east gateway of Stflpa I at SäftcT, Madhya Pradesh, India. Ca. second-third decade of first century C.E. Photo: Archaeological Survey of India.
(48). She cites an inscription accompanying the scene, which she transliterates as "ErapatoNägaräja Bhagavato Vadate" and translates as "Serpent King Erapata adores the Holy One" (48), as evidence. However, the inscription does not prove that "the Holy One"—that is, the Buddha—is represented emblematically by the seat and the tree but only that Erapattra'is honoring him. The key term in the inscription is vadate, from the root vand, which has a variety of meanings, including to praise, celebrate, laud, extol, show honor to, do homage to, salute respectfully or deferentially, venerate, worship, adore, to offer anything, and to show honor to anyone.23 The physical presence of the Buddha is not implicitly required in most, if not all, of these meanings. In other words, the Buddha can be praised, celebrated, lauded, extolled, have honor shown to him, have homage paid to him, be venerated, worshiped, adored, have offerings made to him, and be shown honor without being physically present.
Based on the literary account I have found that most closely corresponds with the Bhärhut composition, I suggest that the seat and the tree simply indicate the Buddha's "sitting place" and are not emblematic of the Buddha.24 While I cannotyet explain why the Buddha is absent from the scene,25 I believe that his absence is simply that—an absence—rather than a presence indicated by the seat or the tree. If I am correct, then the seat and the tree are not emblems, and the symbolic/emblematic premise of the multivalen-cy theory, and the aniconic theory as it has been traditionally understood and applied, cannot be supported.
Although other examples illustrated by Dehejia can be used to demonstrate the idea of the
absent Buddha rather than the Buddha indicated by emblems, a scene on one of the gateways at Säňcí is especially pertinent (fig. 2). Dehejia and others before her have identified this composition as the Great Departure of the Buddha-[to-be], but this may not be the case. In her discussion of this scene, Dehejia questions how what she calls my "pageantry" theory can justify the fact that no Buddha (or, more correctly, Bodhisattva) is seated upon the horse. While the explanation for this scene that I propose provides an answer to this intriguing and important question, as I shall explain below, I am puzzled as to which emblem is intended to depict the absent Buddha-to-be in this composition. Simply, the horse does not represent the absent figure, nor does the umbrella, which even in so-called iconic depictions of the Buddha is held above his head. In other words, it is difficult to defend the belief that the Buddha-to-be is indicated here through emblematic means. If the scene depicts the Great Departure, the Buddha-to-be can only be indicated by his absence and not as the equivalent of a nonfigurative symbol. Therefore, the art shows not the "presence" of the Buddha-to-be but his absence.26
I suggest that unless there is specific internal evidence in an artistic composition to demonstrate that nonfigurative motifs are intended to serve as surrogates for something else, we should be cautious and not make this assumption. Alternatively or additionally, there must be external evidence to demonstrate that early Buddhist art employed such emblems if we are to use this mode of interpretation. However, Dehejia does not explain how she knows the meaning of any of the "emblems" she discusses.
aniconism and the multivai
Dehejia's "Three Vaud Categories" (or "Three Valid Aspects") of Emblems
Dehejia explains that the "interpretation of the emblem, with its inherent fluidity of meanings, will be considered under its three valid categories—first as an aniconic presentation of the Buddha, next as a sacred site, and finally as an attribute [of the faith]" (48)." Except for her brief conclusion (64), the preponderance of her article is devoted to discussion of the emblem under these three aspects (48-64). Although Dehejia claims that these are "diree valid categories," itis unclear how their validity is established, nor does the discussion of die works of art reinforce the claim of their validity. Further, it is not apparent from Dehejia's presentation whether these three categories are intended to explain all possible subjects in early Buddhist art or whether there are others as well. If these diree categories are intended to cover all contingencies, dien I disagree. Even examples cited in Dehejia's own article demonstrate that diis is not the case. I would argue, for example, diat the two Ašokan scenes discussed below are not merely "site" scenes, though the events take place at Buddhist sacred sites. Instead, they might be better classified as historical^or quasihistorical narratives intended to convey a number of messages to the Buddhist devotee, including, among odier things, reverence to the Buddha and his relics by living beings other than humans in one relief and a model of benevolent cakravartin-ship in the person of Ašoka in the other. Most importandy regarding these three categories, although Dehejia states that the "exact interpretation of the emblems depends on their visual context" (45), this statement is never clarified, nor is the reader provided with a means of determining the interpretation of motifs in individual cases.
The Emblem as Aniconic Presence. Under die first "valid category," Dehejia claims that emblems may denote the aniconic presence of die Buddha. She cites a dictionary meaning of the term aniconic as "symbolizing without aiming at resemblance" and aniconismas "worship or veneration of an object that represents a god without being an image of him" (45). In an endnote to these definitions, she provides additional meanings from other sources. She notes diat The Random House Dictionary explains aniconism as "opposition to the use of idols" (65 n. 5). Although Dehejia has understandably selected a definition
:E OF EMBLEMS: ANOTHER LOOK                                  117
that reinforces her interpretation and application of the term aniconic, alternative meanings cannot be ignored when examining the ways in which aniconism has been viewed in scholarship. Since die traditional aniconic theory has been intimately linked with the idea of a presumed Hlnayäna prohibition, an historical overview of the dieory of aniconism and its abiding interpretation in scholarship also must accommodate the definition that cites opposition to the use of idols. Further, the definition of aniconism that Dehejia uses—worship or veneration of an object that represents a god without being an image of him— is problematic for today's scholars of Buddhism. At the time the aniconic theory was first proposed, Western scholars assumed that the Buddha was a god, a factor that may have influenced the choice of the term aniconism to describe what they perceived in Indie art. But, as Helmuth von Glasenapp has since demonstrated, the Buddha was not a god.28 Therefore, while Dehejia's choice of a definition for the term aniconic selects an element of aniconism—the use of emblems— that serves the purpose of the theory of multiva-lence she is proposing, other components of the aniconic and aniconism should not be ignored in a diorough study of the issues.
The section of Dehejia's article called "The                       *
Aniconic Presence" is predicated on the idea that                       JS
early Buddhist artistic compositions contain "em-                       $
blems" that stand for what should be anthropo-                       ;r
morphic representations of the Buddha (48). However, in the case of the Erapattra relief from                       ,*
Bhärhut discussed above, I suggested that the                       j(e
Buddha may be indicated by his absence, not                      1'1
through the use of emblems. A similar case can be                       {,,
made for a relief from Bhärhut that shows King Ajätašatru, who is identified in the accompanying                        „f
inscription, kneeling in a reverential pose before a seat that has a pair of footprints before it and a                       ,
parasol above (fig. 3). Dehejia states unequivocally: "To ensure that the viewer correcdy identifies die footprints, throne, and parasol as the Buddha, the artist added the label 'Ajätasatu Bhagavato vamdate,' or 'Ajätashatru bows to the
Blessed One'___Itis difficult to misconstrue the
artist's intention" (50) .29 However, I suggest that die footprints, seat, and umbrella signify the Buddha's "sitting place" and are not emblems substituting for his physical person. If this scene t represents King Ajätašatru's visit to the Buddha himself, then I propose that the Buddha's presence is indicated by his absence. The throne, with its respectful parasol above, would mark
118
SUSAN L. HUNTINGTON
Fig. 3 (Dehejia fíg. 4). Veneration by King Ajätašatru.
From Bhärhut, Madhya Pradesh, India.
Ca. 100-80 p.CE. Indian Museum, Calcutta.
Photo: Archaeological Survey of India.
the Buddha's sitting place, and the footprints below would represent the place for the Buddha's feet. The inscription, which Dehejia believes so unambiguously proves that the footprints, throne, and parasol represent the Buddha can be as easily interpreted to mean that Ajätašatru bows to honor a being who is not shown.
An alternative interpretation may also be offered regarding this scene. Since Ajätašatru was converted by the Buddha only a year before the Buddha died, much of the king's devotions to the Buddha took place in the years following the Buddha's death, when the Buddha was no longer present. This scene might show die king venerating a place where the Buddha once sat but after the Buddha was already dead and therefore not to be expected in the composition. In this case as well, the footprints, throne, and parasol would be part of the paraphernalia installed at the place of veneration rather than symbols indicating the Buddha's person.
Another example Dehejia uses to illustrate the Buddha's aniconic presence is what she calls the "Enlightenment face of the Prasenajit Pillar"
from Bhärhut (fig. 4). Dehejia refers to the uppermost of the three panels on this pillar as exemplary of her second prerequisite (in which the "emblem" conveys multiple meanings). She claims that "the prime intention of this panel was to depict the historical event [of the Buddha's enlightenment]" (45) and states unequivocally that the "throne, which is surmounted by triratna emblems and stands beneath the bodhi tree encircled by a hypaethral shrine, represents, in this instance, the presence of the enlightened Buddha" (50). Yetshe is apparently also persuaded by the arguments I have proposed suggesting that this scene represents worship at the site of Bodh Gayä at a time after the enlightenment of the Buddha, for she offers a "site" interpretation of the scene as well.50 However, whereas I believe that the worship at the site is the principal theme of the relief, she claims that it is primarily a depiction of the eventof the Buddha's enlightenment. A key difference between our interpretations centers on the translation of the inscription, "Bhagavato sakamunino bodho," which she interprets as "enlightenment of the Holy One Šäkyamuni" (50). I believe, however, that the term bodhoxazy denote the tree and not the event of the enlightenment.31 Based on an examination of the narrative elements in the composition and this inscription, I suggest that the scene shows devotees (at a time subsequent to the Buddha's enlightenment) indicating their reverence to the sacred place of the event and the sacred tree underwhich the Buddhaattained enlightenment. While Dehejia claims that the Buddha is indicated emblematically by the tree, I propose that the throne marks the location where the Buddha sat beneath the tree and that the tree represents itself.
To explain the depiction of the building, which, as far as is known from archaeological and textual sources, was not present at the time of the Buddha's enlightenment, Dehejia suggests that the artists who created Ulis panel may have been unaware that a portrayal of a building erected after the time of the event might be anachronistic in a representation of the event itself. She cites what she considers to be a parallel example from the sixteenth century, but a theoretical analogy does not offer proof for the Bhärhut case. Further, the major basis for the interpretation of the scene as a depiction of the sacred site is the presence of the building. Then to argue that the building is there because the composition is a
ANICONISM AND THE MULTIVALENCE OF EMBLEMS: ANOTHER LOOK
119
depiction of a site seems to be circular reasoning. I suggest that a post-enlightenment structure is depicted because a post-enlightenment scene is intended.
The central and lower panels on the same face of this pillar show two other compositions normally associated with the Buddha's enlightenment. The central panel shows four groups of gods, separated, as Dehejia says, by the "compositional device of a tree" (51). The criteria for determining that the tree is not an emblem in this case are not specified. Inscriptions accompanying the scene name the groups of gods as beings from northern, eastern, and southern quarters, with those from the western quarter presumably also included, although an inscription for this group is lacking (50-51 ).32 Dehejia states that these beings came to honor the enlightened Buddha (51 and 66 n. 22) ,33 although at least one textual source records that they came prior to the enlightenment in order to help weaken Mara.34 The figure at the lower left of the composition is Mara, sitting beneath a tree and writing in the ground.35 Whether the gods are there to help defeat Mara or to honor the newly enlightened Buddha,36 I propose that the composition does not require the presence of a Buddha image. Instead, I suggest that the scene is not the Buddha's enlightenment but rather just what it appears to be and just what its inscriptions suggest: it is the representation of the role played by the celestial beings in this momentous event. If a camera had been present at the time of die Buddha's enlightenment, this scene might show what the camera would have found if it had turned to pan the audience. I suggest that these beings are immortalized in the artistic composition for their role as supporters and devotees of the Buddha. The emphasis on these figures correlates with textual sources, such as the Lalitavist-ara, which contains a lengthy and detailed description of the role of the devas, although it might be overlooked by those concerned primarily with what was happening to the Buddha.37
The lower panel on this pillar shows female musicians and dancers, who, according to Dehejia, arrive at the site of the enlightenment to honor the Buddha (51).38 As she notes, four of the inscriptions accompanying this scene identify four of the figures as specifically named apsara-sas. She claims that a fifth inscription reads: "music of the gods enlivened by mimic dance" (51) .39 There are no motifs in the scene that can
Fig. 4 (Dehejia fig. 5). "Enlightenment
Face" of "Prasenajit Pillar." From
Bhärhut, Madhya Pradesh, India.
Ca. 100-80 B.CE.
Indian Museum, Calcutta.
120
SUSAN L. HUNTINGTON
Fig. 5 (Dehejia fig. 9). Buddha's Descent from
Träyastrirrisa Heaven at Särikäsya? Ca. first century b.c.e.
From Butkara I, Swat, Pakistan. Saidu Sharif Museum.
Photo: Martha Carter.
Fie. 6. Devotion at Säiikäsya.
From Bhärhut, Madhya Pradesh, India. Ca. 100-80 b.ce.
Indian Museum, Calcutta. Photo: John C. Huntington.
be construed as emblems, and I propose that the subject matter be interpreted literally, based on the internal information contained in the inscriptions and visual composition. These tell us only that the apsarasas are dancing and performing music. To use the camera analogy again, it is as if the camerahad nowpanned to the orchestra and performers. And, while the occasion for dancing and singing may have been the remarkable transformation from Bodhisattva to Buddha that Šäkyamuni was undergoing, I suggest that the figures are shown as important devotees in their own right, like the devas in the scene above it
Taken together, what do the three panels on this face of the so-called Prasenajit pillar communicate? Dehejia discusses these scenes as part of her "Aniconic Presence" theme and summarizes her interpretation in this way: "The three panels on this face of the Prasenajit panel [sic for pillar] thus represent simultaneous events; they pertain to the enlightenment of the Buddha, as proclaimed by the inscription in the topmost panel, in which the Buddha's presence is indicated by aniconic emblems. Read thus, the presence of the gods of the four directions and of the heavenly apsaräs becomes meaningful" (51). (The reader is left to infer how the presence of the gods and the heavenly apsarasas become meaningful.)
However, I propose that the upper scene shows two pairs of devotees offering their veneration at the place where die Buddha sat, meditated, defeated Mara, and became enlightened and that this devotion takes place after the time of the Buddha's enlightenment. The two lower panels are also devotional scenes, but, in these, celestial beings offer dieir celebrations to the Buddha. Together, these reliefs reinforce the theme of devotion to the Buddha by his followers that I suggestisan importantmessage of early Buddhist art. While the central event being celebrated is the most momentous event of the Buddha's life and one of the most important subjects in Buddhism—Šäkyamuni's enlightenment—it is the celebrators, not the Buddha, who constitute the subject of these particular works of art.
Dehejia identifies a small relief found in recent excavations in the Swat region of Pakistan as an aniconic representation of the Buddha's descent from Träyastrirrisa Heaven at Särikäsya (fig. 5). Extrapolaüng from my interpretation of a relief from Bhärhut, which traditionally has been identified as the Buddha's Descent at Särikäsya (fig. 6) but which ľ believe depicts a worship scene at
ANICONISM AND THE MULTIVALENCE OF EMBLEMS: ANOTHER LOOK
121
Särikäsya,40 Dehejia assumes that I would apply the same interpretation to this Swät relief. She says that "Those who deny the existence of ani-conism41 would see the ladders as the actual set of stairs set up at the pilgrimage site of Sänkissa" (53).42 However, while both the Bhärhut and Swät compositions apparently refer to die Buddha's descent at Särikäsya, they may represent completely different themes. The one from Swät might show the event (with the Buddha missing) ,43 but the other, from Bhärhut, does not. In the Bhärhut carving, rows of devotees circumambulate the ladders and hold their hands in aňjali mudra as an expression of their devotion, and I contend for the reasons presented in my article that that scene represents the holy site of Särikäsya, with devotees honoring the site.44 If this is true, the two reliefs would convey completely different messages, one of which might relate to the Buddha life event, the other to die commemoration and celebration of that event and the place where it occcurred.
As Dehejia observes, Indra and Brahma are depicted in the Swät relief, but the two Vedic gods are absent from the Bhärhut composition. Because the anicbnic theory has never been extended to include' beings other than the Buddha, Dehejia recognized the need to provide an explanation for the absence of these two figures, who are crucial to the narrative of the Buddha's descent. She proposes that the absence of the Vedic gods, which she terms an avoidance, is "to be seen in the context of Hindu worship which, in the Vedic period, was concerned only with sacrifice and in which images of the gods played no part" (66 n. 27). She goes on: "Few early images of Hindu deities exist. It was mainly with the spread of Puränic Hinduism that images of the Hindu gods proliferated" (66n. 27). But Dehejia creates a very serious problem for her argument with this explanation. First, she associates the art of Bhärhut with the Vedic period, which concluded more than half a millennium before the Bhärhut stüpa carvings were created. Second, she implies that Vedic practices regarding die use of images, whether she is correct in her claims or not, had some bearing on a Buddhist monument. Since Buddhism reflects a thoroughly non-Vedic religious system—many might even say an anti-Vedic system—it is doubtful that the Vedic situation is pertinent here. While it is true that Indra and Brahma are Vedic gods, their appearance, or lack of appearance, in Buddhist art cannot be
expected to be governed by the rules of Vedism. Their subservient role in Buddhism—primarily as attendants to the Buddha—makes it clear that their status has been transformed from thatwhich they enjoyed in Vedism. Further, Dehejia's argument that it was only with the spread of Puränic Hinduism that images of the Hindu gods proliferated seems to be almost a Hindu equivalent of the Buddhist aniconic theory—in the Buddhist case, the earlier aniconic tradition of the Hinayänists was supposed to have been supplanted by the later Mahäyäna image tradition; in the Hindu case, she implies that the earlier Vedic lack of imagery was replaced by the imagery of the Puränic Hindu tradition. Since there is a decided lack of early imagery for both Buddhism and Hinduism (andinjainismaswell) even when the religions already flourished, I propose that there might be odier reasons for the apparent lack of early images.45
My interpretation of the Bhärhut panel offers a far more plausible explanation for the absence of Indra and Brahma in the composition. If the scene shows worship at the site of Särikäsya after the time of the Buddha life event, then it is unnecessary to include a depiction of the Buddha (who was no longer there), and there is no                      * need to include the figures of the two Vedic gods                       jJJ who accompanied him on his descent. It is only when the composition is viewed as a representa-                      ^ tion of die event itself that the absence of Indra                       m and Brahma becomes problematic.46
Dehejia then querieswhyarelief from Mathurä                      «
(fig. 7) that has die Buddha along with Indra and                      m
Brahma is "unhesitatingly accepted as a life scene                      J
. . .  [when] Xuanxang's [sic for Xuanzang's]                       ,a
account of the site of Sänkissa speaks of an image                      I
atop die steps. Why is it suddenly assumed that ' depictions of events of the life of the Buddha... became commonplace' at the precise stage when                       f
the iconic replaces the aniconic?" (53) .47 A care-                       ř,
ful look at die Mathurä relief reveals that the Buddha is present in each of the life scenes depicted on diis panel. The context of the descent vignette among a set of five Buddha life scenes, all shown with Buddhas, in a single composition offers compelling evidence that the episode represents a Buddha life event. I suggest that Dehejia is looking indiscriminately at a number of compositions that have been traditionally identified as representing one subject (in this case, the Buddha's descent), and, by ignoring the many distinctions among these
■
122
SUSAN L. HUNTINCTON
Fig. 7. Buddha Life Scenes. From Räjghät, Mathurä, Uttar Pradesh, India. Ca. second century C.E. Mathurä Museum, Mathurä. Photo: John C. Huntington.
Fic. 8 (Dehejia fig. 10). Buddha's First Sermon. From Bactro-Gandhära region. Ca. second-third century c.e. Present whereabouts unknown. Photo: Archaeological Survey of India.
ANICONISM AND THE MULTIVALENCE OF EMBLEMS: ANOTHER LOOK
123
reliefs, has erroneously concluded that they represent the same subject.
The defense of aniconism is stretched in Dehe-jia's discussion of two reliefs from the Bactro-Gandhära region (figs. 8 and 9). Figure 8 is a depiction of the Buddha's first sermon in which a seated Buddha reaches out a hand as if to turn a wheel. Dehejia claims that such a representation, aside from indicating the action of the Buddha turning the wheel of the law as he is said to have done with his first sermon, "may also be understood as the anthropomorphic Buddha giving his sanction to the earlier homage paid to the wheel-topped pillar" (55) .48 She reasserts this claim for figure 9, showing a standing Buddha touching or pointing to a pillar topped by three wheels, and concludes by saying that Those who believe in the primacy of emblems may suggest that the anthropomorphic figure derives its validity from the earlier established emblem!" (55). But Dehejia provides no evidence to support the assertion that the Buddha is "giving his sanction to the earlier homage paid to the wheel-topped pillar" in figure 9. For figure 8, what is the rationale for saying that the actual Buddha performing his sermon is gaining validity from the emblem? Is Dehejia suggesting that he is gaining validity from theemblem that symbolizes his first sermon but only became an emblem symbolizing his first sermon after the first sermon itself? What is the evidence that justifies these proposals in
Fig. 9 (Dehejia fig. 11). Buddha Touching or Pointing to
a Wheel-topped Pillar. From Bacto-Gandhära region.
Ca. second-third century c.E. Private Collection, Japan.
Photo: Martha Carter.
terms of Buddhist practice, literature, or other sources? From what evidence are these ideas drawn, and how are they supported?
I suggest that figure 9 depicts a Buddha pointing to a wheel-topped pillar that has been erected, perhaps to honor the place where a sermon had been given.49 He is accompanied by monk devotees at the left and Vajrapäni at the right, while flying celestials hover above. Figure 8 probably shows a literal rendering of the Buddha's turning of the wheel of law. The claim that the Buddha is here gaining his validity from the emblem is unsubstantiated and, as far as I can determine, does not make sense either Buddho-logically or in relation to the artistic developments of the period.
As an aside, Dehejia identifies the configuration in figure 9 as a triratna. While this configuration shares a visual resemblance to the triratna form that she illustrates (compare with fig. 17, Dehejia's fig. 22), the triratna component is absent in this composition, and I suggest that the configuration depicts a pillar topped by a capital that has a design incorporating three wheels. The assertion (55) that the. form refers to the Buddha, the Dharma, and the Samgha (the traditional interpretation of the triratna) is therefore untenable. Because three wheels are shown rather than the triratna, it is more plausible that they are intended to suggest three teachings, or three levels of teachings, or something else pertaining to the number three and potentially conveyed by the wheel.
Further developing the theme of the "Aniconic Presence," Dehejia claims that while the artists of the Bhärhut and Säňd monuments portrayed simple, and often single, emblems, the AmarävaS artists used a series of emblems combined to build up an "emblematic body" (55)50 for the Buddha. She illustrates a composition that shows a crowd of figures surrounding a central throne with a number of accoutrements (fig. 10). But upon what basis is the extraordinary suggestion made in the interpretation of the subject of this composition that the Buddha's feet are represented by the footprints (this is the most plausible of the equations); that his limbs are represented by the cushioned throne (how is it determined that it was not his buttocks that was represented by the throne?); that his torso is represented by the pillar [of radiance]; and that his head is equivalent to the triratna above? What is the foundation for this interpretation? Where is
124
SUSAN L. HUNTINGTON
Fig. 10 (Dehejia fig. 12). Buddhist Devotional Scene.
From AmarävatT, Andhra Pradesh, India. Ca. second
century c.e. AmarävatT Site Museum.
Photo: AmarävatT Site Museum.
the Buddhological or other evidence of these equivalents? Why is this not another example, like those discussed above, of the Buddha's throne and its accoutrements being shown not as emblems of the physical person of die Buddha but rather as part ofthe setting of the scene?
The Emblems of the Tree, the Pillar, and the Stüpa as a Sacred Site. The second of the "valid categories" concerns the depiction of sacred spots and the devotions performed at them (56-61).51 Dehejia says that in "their second aspect, the emblems of the tree, pillar, and stüpa, seen in relief sculptures, may represent sacred spots, or ärthas,52
and the devotions performed there. Thus, the bodhi tree may be intended to represent Bodh Gayä, site of the enlightenment; the wheel-crowned pillar may represent Särnäth, site of the first sermon; and the stüpa may represent one of the sacred relic mounds built at a variety of sites" (45) .M
Dehejia discusses two reliefs at Säňcl that depict the Emperor Ašoka's visits at Buddhist sacred sites (figs. 11 and 12).54 The first composition shows a visit by Emperor Ašoka and his entourage to the bodhi tree at Bodh Gayä; the second scene shows his visit to the Rämagräma stüpa. Although Dehejia does notspecifically state this, the premise of her article and the subheading for this section of her text imply that she would interpret the bodhitree and/or the building in the center ofthe Bodh Gayä scene and the stüpa in the Rämagräma scene as emblems standing for something else, presumably the sites of Bodh Gayä and Rämagräma, respectively. However, I suggest that these subjects do not serve emblematic roles in these scenes: they are major "actors" in the narratives and, therefore, represent themselves and not something else.
With respect to the Bodh Gayä depiction, the Ašokävadäna makes it clear that Ašoka was visiting the tree when he went to Bodh Gayä and not a Buddha in the form of a tree or as symbolized by a tree—and I see no justification for reading additional layers of meaning into the tree in the A'sokävadäna or in an artistic composition such as this.55 The only odier possible motif that might serve as a nonfigurative emblem in this composition is the building and its altar. But if Dehejia would claim that the building with its altar serves
Fig. 11 (Dehejia fig. 14). Emperor Ašoka's Visit to the Bod/»'Tree at Bodh Gayä during the Quinquennial Festival. On east
gateway of Stüpa I at SäňcT, Madhya Pradesh, India. Ca. second-third decade of first century c.e.
Photo: Archaeological Survey of India.
é
ANICONISM AND THE MULTIVALENCE OF EMBLEMS: ANOTHER LOOK
125
as an emblem for the site, then I disagree. I suggest that it is included because it was present at the site at the time of Ašoka's visit and was an important element in the Ašokan theme. Dehejia claims that the building was not present at the time of this visit by Aioka, citing a passage (57) in the A'sokävadäna where Ašoka is said to have "mounted a platform that he had had built on all four sides of the Bodhi tree."56 She interprets this statement about the platform to mean that there was no shrine present at the time Ašoka lustrated the tree. However, the mention of a platform at the base of the tree does not preclude the simultaneous existence of a building. In fact, the sequence of events recorded in the A'sokavadana reveals that on an earlier visit (his first visit) to Bodh Gayä, Ašoka had built a "caitya" there.57 Since the term caitya can refer to a building associated with a sacred relic, such as a bodhi tree, it is possible—even likely—that the building shown in the relief is a caitya and, specifically, the very one built by Ašoka.58
Dehejia identifies the event depicted in the Säňcí relief as Ašoka bathing the bodhitree due to his desire to revive the tree after it had been cursed by Ašoka's chief queen, Tisyaraksitä. She does not cite die source for the version of the story she uses to explain the scene, but her citation about the platform from which the bodhitree could be watered occurs on page 266 of Strong's translation of the A'sokävadäna. On the preceding page, the section heading clarifies that the event she notes was part of the Quinquennial Festival, duringwhich Ašoka presented the Buddhistcommunity with all of his property except his state treasury and promised to bathe the bodhi tree with milk, which he dien did. Dehejia has, therefore, somehow confused the Quinquennial Festival with the Tisyaraksitä story.59
Dehejiamay have been following Foucher, who cites the version of the Tisyaraksitä story in the
Divyävadäna, which relates that the tree was revived by Ašoka after his wife tried to kill it60 But if she is relying on the Divyävadäna for one portion of her interpretation—that is, for the story—how can the A'sokävadäna—and a different story—be used to decide whether abuilding should orshould not be present in the composition? Regardless, it is clear that the tree and the building are important elements in the story rather than emblems. Indeed, the centrality of the tree and structure indicate their preeminence as foci of devotion in their own right and not as emblems.
Tli e second Ašokan composition referred to by Dehejia (57) depicts Ašoka's visit to Rämagräma stüpa (fig. 12).61 After the Buddha's cremation, his relics were divided into eight portions, and one portion was given to each of the eight kings of the time. Each portion was considered to be a drona (literally, a bucket[ful]). Each king then built a stüpa over these relics. These stüpas were called drona stüpas (because of the drona measurement and also because a brahman named Drona had performed the division of the relics). Additional stüpas were erected over the bucket that had been used for measuring the relics and over the embers of the cremation fire. At some point, the drona stüpa that had been built at Rämagräma was supposedly flooded by the Ganges River, and the relics were swept down to the underwater palace of a näga king. There, the näg-os not only worshiped the relics but did so in such a grand fashion that when Ašoka came to Rämagräma to retrieve this portion of the relics, he was convinced to allow the nägas to retain possession of them.62 The composition at Säňcl shows a stüpain tlie center, with a group of nägas on the left appearing in devotional poses. To the right of the stüpa, a human king, presumably Ašoka, arrives in his horse-drawn chariot with his entourage, apparently with the aim of retrieving the relics.
Fig. 12. Emperor Ašoka's Visit to the RSmagräma Stüpa. On south gateway of Stüpa I at Säňcl, Madhya Pradesh, India. Ca. second-third decade of first century c.e. Photo: John C. Huntington.
126
SUSAN L. HUNTINGTON
Like Ašoka's visit to Bodh Gayä for the Quinquennial Festival, this composition seems a straightforward illustration of a story that is recorded in literature in a forthright fashion and is not characterized by multiple meanings.631 submit that the siüpais intended as a representation of the actual structure that had been erected to house the Buddha's relic, rather than as an emblem, and therefore serves as part of the narrative. Simply, the scene does not include an emblem, and it is not multivalent.
Dehejia's classification of the two Ašokan scenes under her "Sacred Sites" category does not do justice to their Buddhological content. Although I authored the sacred site idea and am pleased that she applies it in constructing her own theory (albeit without direct acknowledgment of my work), I do not feel that these Ašokan scenes are explainable solely as sacred site images. Their message is something greater. Of course, Bodh Gayä and Rämagräma are sacred sites, but much more is implied by the scenes. One message behind the selection of these subjects may have been to show aparadigmatic king demonstrating the way to honor the Buddha through his relics at the sacred sites.64 Strong notes regarding the Rämagräma story that the emphasis in the Asokävadäna is "at least nominally on the value of devotion to the relics, whether it be die devotion of Asoka or of the nägas."65 The two compositions at SäňcT showing Ašokan themes portray his reverence for the relics and are a paradigmatic demonstration of the highest level of lay worship. As Strong explains, one of Ašoka's most important actions was to redistribute the Buddha's relics into 84,000 stúpaš, which, I might add, reflects the desire to make them more accessible to devotees.66 It was after this act that he was no longer called Ašoka-the-FiercebutDharmäšoka.67 Such an analysis melds perfecdy with the evidence I have offered regarding the importance of relics in Buddhism.68 That such a composition is illustrated adorning a monument that is itself a stupa—a stupa founded by Ašoka and almost certainly housing a portion of the relics retrieved from the other seven drona stüpas—may not be coincidental.
These two scenes do not contain emblems; rather, the tree shrine and the Rämagräma stúpa seem to serve as themselves in the narrative and not as surrogates. If the Bodh Gayä tree shrine and the Rämagräma stúpa are intended to indicate their respective sites, they do so as both key elements in each narrative and as participants in
the story, not as symbols. I see litde reason to bring in a discussion of multivalency—or of ani-conism—in the interpretation of these two compositions. Further, to distill the content of the scenes, with their emphasis on Ašoka's activities on behalf of Buddhism, to simple representations of sacred sites indicated by a tree or a stüpa fails to acknowledge the underlying message of piety and devotion expressed by these narratives. Dehejia's discussion of her figures 15,16, and 17 (figs. 13, 14, and 15, respectively) corroborates the theories I have proposed, particularly in my Art Journal article. She notes that some reliefs show the stúpa and the pillar as memorials erected at sacred pilgrimage sites and that in these cases the stúpa and the pillar69 do not indicate the presence of the Buddha.70 This section is illustrated by a relief at Säňcí showing the worship of a stúpa by people dressed in non-Indian garb (fig. 13), which I published in Art of Ancient Indiavňůi the same interpretation;71 a relief that she identifies as showing Särnäth due to the presence of a wheel-topped pillar being worshiped by people and the presence of deer below (fig. 14); and a
Fie. 13 (Dehejia fig. 15). Devotion at a Stüpa. On north
gateway of Stüpa I at SäňcT, Madhya Pradesh, India.
Ca. second-third decade of first century c,E.
Photo: American Institute of Indian Studies.
ANICONISM AND THE MULTIVALENCE OF EMBLEMS: ANOTHER LOOK
127
relief that she identifies as a depiction of Bodh Gayä for it shows the bodhi tree and a temple beneath it (fig. 15). Because her interpretation derives from mine, it follows that I am in general agreement with this portion of Dehejia's discussion. However, while Dehejia interprets die stüpa in figure 13, the wheel-topped pillar in figure 14, and the bodhi tree and temple in figure 15 as emblems, I contend—unless it can be shown otherwise—that they are simply depicted as themselves, as any other landmarks or sacred foci of a site might be.
While discussing these three compositions (figs. 13-15), Dehejia addresses a point I made in my Art Journal article, although she misunderstands the implications of my remark. In summarizing my proposal that many of the so-called aniconic compositions are worship scenes at Buddhist sites, I state that these scenes express concepts central to the practice of Buddhism during the period of the creation of the art works, "particularly relating to the exaltation of lay worship."72 Dehejia takes issue with the emphasis on lay worship as opposed to worship by monks at sacred sites (58-
Fic. 14 (Dehejia fig. 16). Devotion at a Wheel-topped Pillar at Särnäth. On south gateway of Stüpa I at Säňcl, Madhya Pradesh, India. Ca. second-third decade of first century c.E. Photo: American Institute of Indian Studies.
59). However, she develops the idea beyond what I stated or intended. I do not believe thatworship of the Buddha's relics was the exclusive domain of lay worshipers and have never claimed nor intended to claim that the clergy did not practice veneration of the relics.73 But I do maintain that the many artistic renderings of devotion to relics—seen in the reliefs from Säňcí, Bhärhut, and related sites—show lay worshipers, not clergy. In fact, monks and nuns are conspicuously absent in this artistic repertoire. When I write in my Art Journal article that "Furtlier, the scene is clearly an exaltation of Buddhist devotion, specifically lay devotion, since the figures are lay worshipers, as indicated by their secular garb,"74 I am not excluding monastic worship at Buddhist sites—I am merely explaining what is portrayed in the art. When I suggest that the message of these particular monuments, as envisioned by the artists, patrons, and donors, emphasized lay worship, this does not mean that the contemporaneous religious practices did not include the veneration of relics by the clergy. Dehejia's statement that "It is strange that lay
Fig. 15 (Dehejia fig. 17). BodhiTree and Temple, possibly
at Bodh Gayä. On south gateway of Stüpa I at Säňcl, Madhya Pradesh, India. Ca. second-third decade of first century c.E. Photo: American Institute of Indian Studies.
128
SUSAN L. HUNTINGTON
worship alone should be highlighted by those who wish to explain emblematic reliefs as portrayals of pilgrimage sites" (58) is followed by an endnote number. From the sentence, it might be expected that the endnote would contain a reference to "those" who "wish to explain." Instead, without citing me, her note refers to my statement that the "emphasis on sacred plfhas and pilgrimage to them in Buddhism has never waned" (66n. 37) .7sNotonlyis she apparently unaware of the extensive artistic evidence that proves this point, which I will discuss in detail in my book, but she assumes that proof for something in Buddhist practice would be manifest in the art—and the surviving art at that.76 But are we to understand that Gandhära, Ajantä, and the Gupta monuments—which Dehejia names—represent all of Buddhism and its art and that what is manifest in other artistic traditions constitutes aberrations? A glance at the history of Buddhism documents what I have said quite thoroughly.77 Why representations of lay worship of trees, pillars, and stüpas do not occur prominently in Gandhära, at Ajantä, or on Gupta monuments I cannot at this point say, although I contend that these artistic traditions emphatically do incorporate lay worshipers, who, though usually very tiny, frequently appear as devotees in Buddhist reliefs.78
In the same note, Dehejia changes die subject again and quotes my Art Journal article, claiming that I take a "less rigid stand on aniconism" in one of my endnotes. While I reserve the right to change my stand in the future—as any good scholar and ethical human being should—I repeat what I have stated before in this article: I have never said that there are no aniconic images. I have only argued that the aniconic theory is inadequate as a single, all-embracing theory to explain the art of early Buddhism. Clearly, there is much more occurring in the art than simple avoidance of Buddha images. Even Dehejia's own work, which accepts that there are stories about Asoka, diat there are representations of sites, jätakas, and other subjects not focusing on the life of the Buddha, demonstrates that not all of the art requires an aniconic or life-of-the-Buddha explanation. What I stated in my endnote about the possibility that there might be some aniconic compositions does notrepresenta change in die position I propose in the body of my text; it is a statement about what my future research might discover.
Although Dehejia maintains it is strange that lay worship is highlighted by those explaining emblematic reliefs as pilgrimage sites,
she recognizes the fact that so many of the reliefs do show lay worshipers. But her explanation of the prevalence of lay worshipers in these compositions is unsatisfying. She says: "These depictions of lay worship may have occurred by chance in the course of portraying, on each monument, no more than six to eight scenes of worship" (58-59). While I disagree with her number,79 it is more important to ask whether it is defensible to argue that such depictions may have occurred by chance. If the creation of these subjects is a matter of chance, then could it be "by chance" that the Buddha image is absent in the scenes she identifies as demonstrating the Buddha's aniconic presence? In other words, a position must be taken on whether the elements included in the artistic renderings are there by choice (whether it is the choice of the artist, the patron, or some other authority) or by chance. I prefer to begin with the premise that the works of art and the elements they include reflect deliberate choices unless proved otherwise. And Dehejia does not prove otherwise; the popularity of the lay worshiper motif in itself seems to be evidence that it í was intentionally included. Concomitandy, the scarcity of clergy in the scenes must reflect a choice on die part of the makers. If Dehejia is implying that the depictions of lay worshipers represent the chance survival of some reliefs, I maintain that it is still puzzling that among the hundreds of surviving examples there is an apparent lack of compositions representing the clergy.
While agreeing that sacred places are important, Dehejia says that "it is questionable whether, in die early Buddhist edios, pilgrimage to ürthas ever took priority over the life of the Buddha80 and, equally, whether artistic depictions of pilgrimage sites took precedence over portrayals of events from the life of the Buddha" (56). If she is referring to my work, and I assume this is the case, let me clarify my point. I have said that "I suggest that die early Buddhist art of India was not primarily concerned with the biography of Säkyamuni Buddha,"81 but this is not the same as claiming that artistic depictions of pilgrimage sites took precedence over portrayals of events from the life of the Buddha. By suggesting that life scenes of Säkyamuni Buddha are not the primary subject matter of the early Buddhist art, I have not implied that scenes depicting sacred sites are the most important. As the reader of diis article will see, I believe that there are a number of subjects in the early Buddhist art of India, which include
ANICONISM AND THE MULTIVALENCE OF EMBLEMS: ANOTHER LOOK
129
but are not restricted to sacred "site" scenes and whatDehejia calls "pageantry" scenes. But I stand by my claim that, counting these varieties of subjects, those that are clearly and solely scenes of the Buddha's life events are in a minority, if they exist in any numbers at all.
As proof for her statement, Dehejia writes that "Asoka's pilgrimage emphasized the physical form in which the Buddha had lived the life that was being experienced, not merely the sanctity acquired by the site of an event" (56) .82 But the A'sokävadäna suggests another interpretation, for it records that when Ašoka decides to undertake his pilgrimage, he is said to have fallen at the feet of the elder Upagupta and said: "Elder, I want to honor the places where the Blessed One lived, and mark them with signs as a favor to posterity."83 That the events and the Buddha's physical presence are an important part of a site, its history, and ambiance is undeniable. But this does not change what I believe is being shown in much of the art. People who live after the time of the Buddha cannot see him at the sacred sites associated with him, nor can they witness the events of his life; they can only get as close to the Buddha as possible,by being at die place where the events occurred, and especially being in the presence of his bodily relics. It is always a substitute but a necessary one. Dehejia says that the eyewitness accounts of those beings who saw the Buddha "stress the person of the Buddha and his charismatic qualities" (56). I do not dispute diis, but when she concludes from this premise that "Even in a pilgrimage cycle, the emphasis is on the desire to experience the Buddha himself in all his glory" (56), then I question this leap of logic. That is, although there are "eyewitness" accounts describing the glorious physical being that was the Buddha, it is difficult to understand how those accounts can be used to draw conclusions about the nature of pilgrimage cycles. As has become clear from the many studies that have recendy emerged on the practice of religious pilgrimage, the purposes of pilgrims are generally multifaceted. While the ostensible purposes might include the "desire to experience the Buddha himself in all his glory,"84 this seems far too simplistic an explanation for a complicated cultural phenomenon. Dehejia's statement almost seems to argue against her position since any individual who went on a Buddhist pilgrimage to a site where the Buddha had once been and expected to see the Buddha diere would be disappointed. What they would find would be relics or
reminders, such as the bodhi tree. Indeed, this is precisely my point: the Buddha is gone, but the places (and the relics and monuments at them) preserve something of his presence.
Dehejia then moves to the statement that "It is difficult to sustain the argument that early Buddhist artists and devotees were not primarily concerned with the biography of the Buddha" (56-57). I suggest that early Buddhist artists and devotees were probably very concerned with the biography of the Buddha, but the scenes that I discussed in my ArtJournalarticle do not show the biography of the Buddha; they show people being concerned with the biography of the Buddha, if you will. Any comments I have made about the relative importance of the life of Säkyamuni in the art should not be misconstrued to mean that I do not believe in the Buddha's centrality to the Buddhist religion. My analysis is intended to describe only what is visible in the surviving art works under discussion.
Two sculptures from Amarävaü" depict images of Buddhas installed as objects of worship at what may be sacred sites (her figs. 20 and 21, discussed on 60). The first of these compositions formed an important basis of my Art Journal article and was crucial to my comments regarding time and place (fig. 16) .85 The second relief, known today only from a drawing, shows a similar subject (Dehejia's fig. 21).86Dehejia says that the sculptures are "noticed here as possible representations of worship at local sites" (60), but she does not clarify how this relates to her theory of multivalency or how it relates to the theory that rejects aniconism as an all-embracing explanation for the early art of India. In the "Sites" subsection of her text she explains neither the implications of local site worship nor the idea of showing a representation of an image that was being venerated at such a local site. She says that "Since the Amarävati stüpa railing itself is renowned for its circular carved medallions, the scene appears to represent a local site where one such medallion was worshipped" (60). These two thoughts do not follow.87
Dehejia expresses concern over the depiction of footprints beneadi the throne in the Amarävaü" sculpture. Elsewhere in her text, she claims that the depiction of footprints in the Bhärhut panel that I identify as showing a worship scene at the site of Särikäsya (fig. 6) and in the Swät panel (fig. 5) are substitutes for a Buddha image (53). She claims that "the portrayal of footprints in place of the bodily image can only suggest a
w m
«I
I'
130
SUSAN L. HUNTINGTON
Fig. 16 (Dehejia fig. 20). Composition Showing
Devotion to an Image of a Buddha.
From AmarävatT, Andhra Pradesh, India.
Ca. second century c.e. AmarävatT Site Museum.
Photo: AmarävatT Site Museum.
tradition of aniconism" (53). Yet regarding the AmarävatJ composition (fig. 16), she claims that footprints are shown "to reinforce the sanctity of the medallion" (60). The basis for this suggestion is not explained. Is this assumption drawn from Buddhological, inscriptional, literary, or other evidence? And if the problem of the presence of footprints can be dismissed so easily here, in a scene that clearly does not require a Buddha figure since it includes an image of a Buddha, how then can Dehejia be so stringent in her puzzlement over the presence of footprints in other compositions, such as the two so-called "Descent" scenes from Swät and Bhärhut (53)? In her discussion of these two AmarävatI reliefs, Dehejia dien cites Jain evidence regarding the installation of plaques on thrones beneath trees (60-61). However, she does not clarify the date of the Jain evidence nor its implications for the study of early Buddhist art. She concludes this section by saying that "Further studies may clarify the exact significance of these two intriguing site-related portrayals" (61). Here, it seems, she has chosen to ignore my work.
The Emblems of the Tree, the Pillar, and the Stupa as an Attribute [of the Faith].88 Dehejia's use of the term attribute, which is the basis of her third "valid" category, is not explained in her text. Does Buddhism have attributes? If so, what constitutes an attribute? Her explanation that "Not infrequently, the main intention of the artist depicting a pillar, tree, or stüpa was to emphasize the Buddhist truth to which it attests" (63) does notsubstantiate an equation between a Buddhist truth and an attribute. Dehejia states that sometimes "the tree89 is intended to recall the divine wisdom of the Buddha, while the pillar suggests his sacred doctrine " (45) .90 She explains that the stüpamzy not be the great decease nor a sacred íču^asite but rather the Buddha's achievement in finally severing the bonds of rebirth (63). How is it known specifically that "the tree" is meant to recall the divine wisdom of the Buddha? And, assuming that the (bodhi) tree is intended to recall the Buddha's divine wisdom, can the Buddha's wisdom be categorized as an "attribute"? Some would say that it is the very essence of the religion. Upon what basis is it surmised that the pillar suggests the Buddha's sacred doctrine? Wouldn't the meaning of a pillar depend in part at least upon the motifs with which it is decorated? And should the Buddha's doctrine be characterized simply as an "attribute of the faith"? How
ANICONISM AND THE MULTIVALENCE OF EMBLEMS: ANOTHER LOOK
131
Fie. 17 (Dehejiafig. 22). Devotion to Triratna. From
Bactro-Gandhära region. Ca. second-third century c.E.
Peshawar Museum. Photo: Peshawar Museum.
isitknown that the stüpa represents theBuddha's severance of the bonds of rebirth? Is there textual, inscriptional, or other evidence to document these speculations? In what context are these ideas supported Buddhologically, and how is it known that they apply during the period under discussion? A statement that the equivalents cited are true does not make them so.
As examples of artistic renderings that she believes depict Buddhist attributes, Dehejia refers to the "emblems" carved on the "dies" or blocks between the architraves and on the uprights between the architraves of the Säňcí toranas.91 Without further explanation, it is difficult to understand how these representations might have a different purpose than, for example, the one Dehejia illustrates as her figure 17 (fig. 15). She further notes that "due to their capacity for multiple references, emblems intended to indicate the attributes of the Buddhist faith also serve to remind the viewer of the Buddha himself and die site with which an attribute is associated" (63). But how are we to know when and where the emblem means what?
Dehejia illustrates only two sculptures under the category of emblems as attributes of die faith. The first (her fig. 22, discussed on 63) shows the worship of the triratna (fig. 17). This carving is a Kusäna-period work from the Bactro-Gandhära region and is not representative of the early Buddhist art at sites like Säňcí and Bhärhut; its implications for the earlier artistic tradition under discussion are unclear. Dehejia discusses the triratna as a Buddhist emblem referring to the "three jewels of Buddhism," that is, the Buddha, the Dharma, and the Sarrigha. While die triratna's principal meaning may be a reference to the
triple gem of Buddhism, it is shown in this composition in a way that does not differ materially from the ways that trees, stüpas, and other motifs are sometimes used in the art of Bhärhut, Säňcl, and other early sites. Installed as an object of devotion that is being revered, in this case by members of the clergy, this representation depicts not merely an emblem with a literal meaning but a scene in which that emblem is under worship.92 Therefore, I propose that such a scene gready resembles many of the reliefs I have analyzed.
Curiously, Dehejia links her discussion of the triratnav/ith mention of the Buddha's halo, saying that "Gandhäran panels centering on the halo, or triratna, depictneither the worship of the Buddha in aniconic form nor worship at any specific site; rather they represent homage paid to the Buddhist faith itself through adoration of its attributive emblems" (63). Dehejia's equation of the halo to the triratna is puzzling since the triratna is an abstract symbol not associated with the Buddha's physical person and is not a relic. The halo, like the Buddha's robe, his begging bowl, the bodhi tree, and other objects he used, is considered a paribhogika relic and would have been worshiped as such. Justification for claiming that the halo is one of the "attributive emblems" of Buddhism is not provided. Further, if the halo is an attribute, what does it signify?
Dehejia identifies what she calls the "most impressive use of emblems as attributes" (63) in the decoration of the dome of the great stüpa at Amarävafl and illustrates an example (63-64) (fig. 18) .93 She explains that in the lowest register these carvings illustrate either a tree sheltering a seat or the Buddha himself; the central register usually portrays a wheel upon a pillar, although sometimes it has a representation of a preaching Buddha; and the uppermost register invariably shows a representation of a stüpa.9* By looking at the diree compositions occupying the three registers of the Amarävatí dome slab, I cannot determine how they differ from a number of scenes that Dehejia might identify as representing sacred sites or the aniconic presence of the Buddha. Each composition contains a central motif flanked by worshipers (lay worshipers, I might note). Yet Dehejia claims that "On this occasion, the tree or the image beneath the tree represents neither the event of die enlightenment nor the site of the event; the wheel or the preaching Buddha represents neither the first sermon nor its site; and the stüpa stands neither for the great
132
SUSAN L HUNTINGTON
Fio. 18 (Dehejia fig. 24). Three-tiered dorne slab showing,
from bottom: Devotion to Throne and Bodhi Tree;
middle: Devotion to Throne and Wheel-topped Pillar;
top: Devotion to Stüpa. From AmarävatI, Andhra Pradesh,
India. Ca. second-third century c.e. British Museum,
London. Photo: Trustees of the British Museum.
decease of the Buddha nor fora stüpa site" (64). Instead, she claims, the "intention seems to lie in quite another direction" (64), suggesting that the emblems (motifs) "repetitively reiterate the belief in the threejewels of Buddhism, that is, the Buddha (lowest register), the Dhamma or doctrine (central panel), and the Sarhgha ... (upper register)" (64). In other words, Dehejia equates these three motifs with the meaning of the trirat-na. As support for this proposal, she provides an endnote (66 n. 43) in which she suggests I. K. Sarma's agreement with such an interpretation. Apparently, Sarma told Dehejia that recendy unearthed dome slabs from AmarävatI have the syllables bo and dha inscribed on them. Dehejia does not explain where these inscriptions occur in relation to the compositions on slabs, or whether they occur only or primarily on slabs that show the bodhi tree and the dharmacakra. Since, as Dehejia herself notes, bo might stand for bodhi tree and dha might stand for dharmacakra, I am uncertain how these inscriptions therefore prove that the bodhi tree and the dharmacakra appear in these scenes as attributes standing for the Buddha and the Dharma.
Dehejia's equation of the stüpa and the Sanigha is given the following explanation: "I suggest it was because the relic mound was indeed the center of the monastery and the very raison d'etre for the formation of the monastic community that congregated around a stüpa" (66 n. 43). In my view, this thinking requires a vast leap of faith, for the stüpa might stand for other things as well. Further, such a claim about the centrality of the relic mound needs verification and explanation regarding whether this is always true or only at some sites and for some sects. If this is correct, I am delighted, since this information provides further support to my theory that places great emphasis on the relic cult and not simply the narration of the life of the Buddha.
Dehejia continues that monks and lay worshipers who would have circumambulated [the AmarävatI stüpa] would have had on their lips the "refuge" creed, taking refuge in the Buddha, the Dharma, and the Sanigha, but she does not explain how she knows what they would have been saying. Would these same words be spoken during the ritual practices at other stüpas, and, if so, then why does the configuration occur here alone? Dehejia further claims that die reliefs would have been a visual affirmation [to the largely illiterate people] of their faith in the Buddha, the Dharma, and the Sanigha. Again, this speculation is stated
ANICONISM AND THE MULTIVALENCE OF EMBLEMS: ANOTHER LOOK
133
without external proof. Most importantly, Dehe-jia does not define what it is about these particular compositions that forces a different interpretation from the aniconic presence or site interpretations (64) .95 Not only is the slab from Amarävaö similar in type to scenes that she herself might interpret as Buddha life events or site scenes, but she does not provide the Buddholog-ical sources of documentation for her explanations of the meanings of what she calls the emblems. Her suggestion that in this case the tree or the "image" (I suggest absence of an image) beneath the tree represents the Buddha, the wheel represents the Dharma, and die stüpa denotes the Sanigha, is, therefore, unfathomable.
Dehejia's Two Prerequisites
Dehejia's belief in the use of emblems in early Buddhist art is the cornerstone from which she builds her argument for multivalency and is the crux of what she calls the "two critical and complementary prerequisites for the accurate interpretation of early Buddhist art" (45). The first of her two prerequisites is "an awareness of die multiple meanings conveyed by the major Buddhist emblems of the tree, die pillar, and the stüpa" (45), and the "second crucial prerequisite for interpreting the emblems is to acknowledge their multilayered significance" (45). The similarity in the way Dehejia defines her two prerequisites makes it difficult to establish and understand their distinctiveness. In other words, what is the difference between having "an awareness of the multiple meanings" of the emblems and "acknowledging their multilayered significance? "In addition, when the concept of the two prerequisites is introduced, it is explained that they are for the accurate interpretation of early Buddhist art; when the second prerequisite is introduced, it is identified as a prerequisite for interpreting "the emblems" (45, para. 4). Therefore, it is unclear whether the prerequisites are for the interpretation of early Buddhist art or for interpretation of emblems, or whether a distinction is implied at all.
From the context of Dehejia's article and her discussion of works of art, I have assumed that the prerequisites are to be used to interpret the art, not the emblems.951 infer that in the case of her first prerequisite a single motif can have multiple meanings, but, depending on circumstances, one or another of the meanings would be intended (45). The second prerequisite apparently means that references to multiple meanings were made
simultaneously by an emblem within a single work of art (45). My understanding, then, is that the difference between her prerequisites is that in one case the motif carries the possibility of multiple meanings but in a specific context manifests primarily one of them, while in the other case the motif might carry multiple meanings simultaneously.
In addition to this lack of clarity in definition, which makes it difficult to know how to apply the two prerequisites to the art, itis doubtful whether the two prerequisites are applicable to all examples of early Buddhist art.97 If these two concepts cannot be applied to every work of art in this corpus, then diey are not prerequisites. For example, the lower panel of what Dehejia calls the "Enlightenment Face" of the "Prasenajit Pillar" from Bhärhut (fig. 4) does not even contain a nonfigurative motif that can be interpreted as an emblem; the scene above it contains a tree, but even Dehejia does not believe that the tree serves as an emblem in this context; and the two Asokan reliefs (figs. 11 and 12) contain nonfigurative elements, but it is difficult to sustain the notion that their principal role is emblematic. Without the presence of emblems, the prerequisites do not apply to these and similar works of art.
Second, if my interpretation of the two prerequisites is correct, diey are mutually exclusive: a single emblem cannot suggest only one meaning and more than one meaning simultaneously. In order for both prerequisites to be present in a single artistic composition, the work of artwould probably have to contain a minimum of two emblems, one of which would stand for an aniconic representation of a Buddha, or a sacred site, or an "attribute of the faith," while the other would embody two, or perhaps even all three, levels of meanings simultaneously.98 Since I suspect that the Indie artists would not be so contrived as to create an artistic composition containing one emblem that had only one meaning and a second that had as many as three, I would argue that it is unexpected that a single work would manifest both prerequisites. Therefore, it is impossible for a single "emblem" to embody both the first and second prerequisites and unlikely that a given work of art would do so.
Finally, I suggest that unless there is documentary evidence that the artists and patrons responsible for die creation of these artistic compositions were aware of and applied these principles to dieir art, they cannot be considered prerequisites, for prerequisites belong to the creators, not
134
SUSAN L. HUNTINGTON
Fig. 19 (Dehejia fig. 1). King Prasenajit and His
Entourage Showing Devotion at a Shrine with a Wheel.
From BhSrhut, Madhya Pradesh, India. Ca. 100-80 b.cxe.
Indian Museum, Calcutta.
Photo: American Institute of Indian Studies.
the interpreters of the art. If they are instead general principles that reflect patterns found in the art and that can be used to help interpret the art, then this is another matter altogether.
Methodology
Unless there is agreement about the way in which knowledge about works of art is constructed, there is litde chance that different scholars will agree in their conclusions. To understand the different approaches that scholars, such as Dehejia and I, might take to interpret the early Buddhist art of India, it is necessary to address some methodological issues, particularly the discrimination between internal and external evidence. I shall use one of the sculptures Dehejia illustrates, a composition from Bhärhut, and her related discussion as a case study (fig. 19). This rectangular composition shows a railing
and gateway in the foreground that form the forward boundary of the precinct of a shrine in a courtyard. In the courtyard, an entourage of figures, some in a chariot, others on horseback and elephant, moves in a clockwise direction around the central structure. The shrine contains a representation of a Buddhist wheel and two male figures with hands in respectful postures venerating the wheel. A pair of palm trees flanks the upper portion of the scene. There are two inscriptions contained within the composition of diis relief: the one on the roof of the gateway reads "räjäPasenaji Kosalo" (King Pasena-ji [Prasenajit] the Kosala [Kaušala]);" the second, which appears on the roof of the main building, reads "bhagavato dhamachakath" (The Wheel of the Doctrine of the Holy One).100
From the internal evidence contained within the relief, that is, the pictorial information as well as the two inscriptions, we can surmise the following:101
1.  that one (or more; see below) of the figures in the composition represents King Prasenajit of Kosala;102
2.  that the wheel within the upper shrine is a dharmacakra (Buddhist wheel of law) "oF the holy one (bhagavato);
3.  that an entourage of figures, including King Prasenajit, is circumambulating the dharmacakra shrine in a clockwise fashion, as may be determined from the sequencing of the figures around the shrine, and that the entourage will exit the precinct of the shrine through the gateway at the lower right, as suggested by the Figure on horseback coming forward through the gateway;
4.  that two figures (possibly the same person depicted twice; see below) are inside the shrine venerating the wheel.
But a number of questions are not specifically answered within the composition. Some can be answered through the use of sources external to the relief, while others may never be answered. Most importantly, simply from looking at the composition and translating the inscriptions, we do not know who King Prasenajit is in terms of his historicity—such as when he lived, where he lived, what his life stands for Buddhologically, and why he is shown in the art; we do not know the identification of the shrine or its location; we do not know the occasion that is immortalized in the composition (that is, why King Prasenajit is visit-
ANICONISM AND THE MULTIVALENCE OF EMBLEMS: ANOTHER LOOK
135
ing this shrine); we do not know who built the shrine, when it was built, or why it was built; and we do not know the specific identity of any of the other figures in the scene, though we can determine from their costume and physical features that all of them are male. Further, we do no t know how this composition fits into its archaeological and artistic context at Bhärhut, and ultimately into the whole corpus of early Buddhist art, that is, who made it, why it was made, and the purpose of the makers (including both artists and patrons) in portraying this scene as opposed to other possible subjects that were known to them. Although Dehejia does not cite the external sources she uses to interpret the scene—a problem that persists in her discussion of virtually all of the works of art in her article103—she makes a number of suppositions, interpretations, and claims (47), which include:
1.  that King Prasenajit built a temple in honor of the Buddha (she seems to imply that die temple was built at Šrävastl but does not state this outright);104
2.  that King Prasenajit visited this temple (presumably with an entourage);
3.  that the site being shown in the scene does not represent Särnäth, the location of the Buddha's first sermon;
4.  that the label identifying King Prasenajit suggests whatshe infers to be "the actual historical event" in which the monarch visited the Buddha at Prasenajit's capital of Šrävastl and listened to his sermon;
5.  that the wheel is portrayed "as an object of worship in the shrine erected by King Prasenajit and to recall the sermon given there rather than to indicate the actual presence of the Buddha" (47);
6. that there is an "inevitable and surely intentional" conflation of meanings conveyed by the relief: "The shrine was built by King Prasenajit at the spot where the Buddha had preached to him; undoubtedly, the artist intended that the relief should also recall diat event" (47);
7.  that "As a nonfigural emblem, the wheel emphasizes the Law and also refers to the Buddha as the Giver ofthat Law" (47) ;105
8.  that most early Buddhist visual narratives contain this double (Dehejia actually provides a triple) layer of meaning; and
9.  that once this concept of multiple meanings is accepted, "aniconism ceases to be such a vexed problem" (47).
Some of Dehejia's claims probably derive from secondary sources, that is, writers in recent or relatively recent times who have sought to interpret the art and who have in turn based their interpretations upon presuppositions, such as the belief in aniconism. In her analysis of this composition Dehejia adopts a position that draws upon that first presented by Alexander Cunningham more than a hundred years ago and which has been generally accepted since that time. Specifically, Cunningham identifies the building housing the wheel as the "Punya Sála" (Hall of [Religious] Merit) that he claims was built by King Prasenajit at Šrävastl for the use of the Buddha.106 One of the earliest believers in the concept of aniconism,107 Cunningham suggests that in this composition the wheel is a symbol that takes the place of the Buddha himself (though why he would expect the Buddha to be sitting perpetually in the temple builtbyPrasenajitisnot evident). He claims that the wheel was intended "as a type of the advancement of the Buddhist faidi by preaching, and thus becomes an emblem of Buddha the Teacher, in the same way that the Bodhimanda, or seat on which Sákya Muni sat for six years108 in meditation, is used as a symbol of Buddha the Ascetic in all the Bharhut Sculptures, where the figure of Buddha himself is never represented."109
Some claims made by Dehejia, and others before her such as Cunningham, are unverified from Buddhist literature, but others can be supported. A brief search through textual sources reveals that:110
1.  King Prasenajit was a contemporary of the historical Buddha Šäkyamuni, which allows us to infer that the "bhagavato" referred to in the inscription is Šäkyamuni and not another Buddha;
2.  King Prasenajit was a devotee of Šäkyamuni and was converted by him when the king heard a sermon by Šäkyamuni;
3.  The conversion of King Prasenajit by the Buddha took place at the Jetavana in Anätha-pindada's park, near Šrävastl;111
4.  King Prasenajit and the Buddha had an ongoing discourse and King Prasenajit made frequent visits to the Buddha to ask questions and receive the Buddha's wisdom.112 These conversations took place at Šrävastl and possibly other venues as well.113
5.  Särnäth, the place of Šäkyamuni's first sermon, was situated within the Kosala kingdom
136
SUSAN L. HUNTINGTON
of King Prasenajit, and there is no reason to suppose that the king never visited Särnäth.114 Dehejia's claim that this relief does not represent Särnäth, therefore, cannot be assumed a priori. Archaeological sources reveal that there was a temple at Särnäth at least since Ašokan times and that it had a wheel as the main object of worship; it is possible that another early patron had built such a temple there during the lifetime of the Buddha and that dierefore Prasenajit could be shown visiting Särnäth.115 6. King Prasenajit may have built a shrine at Šrävastí (perhaps to honor the Buddha), and it may have had a wheel as the main object of devotion. The seventh-century Chinese pilgrim Xuanzang describes some ruins at Šrävastí, including a "Great Hall of the Law" (Saddhar-ma MahasaLa) built by Prasenajit.116 However, the inscription in the relief that refers to the "Holy One's Wheel of Law" is not enough information to identify the site. (A detailed examination of clues in the relief, such as the two trees, might be helpful.)
Even a more exhaustive search or the discovery of new materials is unlikely to provide the type of detail that Dehejia, Cunningham, and others have accordedthis relief. Even if it is assumed that the Buddhist wheel is intended to serve as a symbol of the Buddha's teachings, it cannot be determined that only one temple housing a wheel as the main object of devotion existed during King Prasenajiťs tíme—whether built by him or not The existence of the temple that Xuanzang refers toatSrävastI doesnotnegate the possibility that other similar temples had been built at other places where the Buddha taught.117 The inscriptions only identify the king and the Buddhist wheel but do not specify the site or the event being commemorated.
Further, the relief does not contain internal evidence to support the claim that the wheel is a symbol replacing the Buddha or that it serves as an emblem of the Buddha as a teacher. Cunningham's statement must have been based upon a belief in the concept of aniconism as a presupposition for interpreting the art. Because Cunningham believed that the wheel served as a substitute for a figure of the Buddha, he titled the section of his text dealing with this relief "Visit of Prasenajita to Buddha" rather than "Visit of Prasenajit to a Temple with a Wheel" or something similar.118 While Dehejia modifies this aspect of Cunningham's analysis, she retains the notion that the
wheel is an emblem serving both as a reminder of the sermon the Buddha gave to Prasenajit and as a reference to the Buddha as Giver of the Law (47). She says that a conflation of meanings is conveyed by the relief and that this conflation "is inevitable and surely intentional" (47) and, further, that: "The shrine was built by King Prasenajit at the spotwhere the Buddha had preached to him; undoubtedly, the artist intended that the relief should also recall the event. As a nonfigural emblem, the wheel emphasizes the Law and also refers to the Buddha as the Giver of that Law" (47).
While I concur that the wheel might be a reminder of the sermon (or one of the sermons, possibly the conversion sermon) the Buddha gave to Prasenajit and at the same time a reference to the Buddha as Giver of the Law, a "reminder of" and a "reference to" are not the same as a "representation of a subject. I suggest that this relief is a representation of King Prasenajit performing veneration at a shrine that contains a Buddhist wheel and that these other meanings, while possibly implicit in the concept of the wheel, are not represented directly in the scene. The multivalent meanings that Dehejia identifies may be inherent layers of the wheel's meaning but are not what is being depicted in the relief. Thus, I contend that die relief portrays a single subject—Prasenajiťs visit—and that, while the wheel is imbued with multiple layers of meaning, the additional layers do not constitute interchangeable subjects for the relief.
Such a scene expresses what I suggest is a popular and important theme on early Buddhist monuments—the devotion and piety of followers of the Buddha.119 Devotees might include humans (both royalty and commoners), celestials, and animals. The Buddhological message of a relief like the Prasenajit composition, I propose, may not be the life of the Buddha in the biographical sense but rather the good works and spiritual evolution of someone like King Prasenajit that occurred through the teachings of the Buddha.120 The frequent appearance of compositions with similar messages among the reliefs at Bhärhut, Säňcí, and other sites suggests that emphasis is being placed on the king. When shown on the early Buddhist monuments, such scenes are not intended to record events in the Buddha's life but rather to highlight the actions of his paradigmatic, faithful devotees. That a Tibetan monarch many centuries after the lifetime of King Prasenajit would claim descent from him suggests that
ANICONISM AND THE MULTIVALENCE OF EMBLEMS: ANOTHER LOOK
137
King Prasenajit was a model of kingly Buddhist worship.121 If Prasenajit's importance was this great to the Tibetans many centuries after the creation of the Bhärhut carvings, it can be inferred that his renown was also considerable at the time the Bhärhut monument was created.122
Buddhist piety is manifest in many of die scenes that Dehejia refers to as "site scenes." But it is also manifest in other scenes that traditionally have been viewed as aniconic representations of events of the Buddha's life. The Bhärhut roundel showing Anäthapindada covering the ground with pieces of gold to provide the Jetavana monastery for the Buddhais also an example of this theme.123 While the Buddha was the recipient of Anätha-pindada's generosity, what is being shown in the composition is not the Buddha receiving the gift but Anäthapindada paying the exorbitant price of the gift. The scene is, therefore, a demonstration of the generous piety of one of the Buddha's followers, not a biographical subject in the Buddha's life.
By not distinguishing between motifs as references, reminders, and representations, Dehejia concludes that the Prasenajit relief was intended to convey multivalent meanings. But, if her assertion that "Most early Buddhist visual narratives contain this double [sic for triple?] layer of meaning" (47) is based on the type of evidence she presents for this composition, then her claim may be insupportable.
This methodological discussion distinguishes between internal evidence contained within an artistic work and the external evidence that can be used to amplify the interpretation of that work. The limitations of external sources are, unfortunately, extensive, not only because of their scarcity but because, frequently, they provide conflicting information. This discussion also attempts to demonstrate that it is necessary to reexamine the art unfettered by the lore and presuppositions thathave been passed down from generation to generation as truths. Because the theory of aniconism is so deeply embedded within a matrix of more than a hundred years of scholarly thinking, any challenge to that theory is scrutinized with careful detail and exceptions and apparent contradictions may seem glaring. However, I suggest that the burden of proof should rest with those who seek to add layers of interpretation and symbolism external to what is shown in the art. If a relief seems to show the literal depiction of devotion by an early king such as Prasenajit, or other subjects, then the burden
of proving that it also is intended to communicate symbolic, emblematic, or other layers of meaning should belong to those who make the claim.
Dehejia and the Issue of Aniconism
The Theory of Aniconism
The issue of aniconism plays two roles in Dehe-jia's article. First, the presupposition of aniconism is incorporated into her discussion of emblems and multivalency, particularly in the "Aniconic Presence" section. Second, her defense of the traditional theory of aniconism is a subtheme woven into her discourse. I have already addressed the first aspectand will only discuss Dehe-jia's treatment of the general question of aniconism here.
As mentioned above, Dehejia had two purposes in writing her article, one stated and one unstated. This duality is manifest throughout her article, as her text weaves between her stated purpose of demonstrating multivalency in the early art of Buddhism and her unstated purpose of defending the theory of aniconism against the ideas I have presented. Her discussion of aniconism is selective, and die transitions between her trains of thought are often left unstated, rendering the sequence of ideas difficult to follow. For example, after her discussion of her figure 1 (fig. 19) there is a digression about the problem of aniconism, which treats a number of separate topics and which is not clarified in relation to the theme and organization of her paper. I shall address the most significant of the issues raised in her article here.
The "Pageantry Theory"
Dehejia discusses what she calls the "pageantry theory" in relation to a composition on one of the gateways at Säňcí that I have published (fig. 2). She says that: "The pageantry theory proposed as an alternative to aniconism is riddled with complications; in particular, there is little evidence, if any, that Buddhism had a tradition akin to that of the Christian passion plays, in which events from a sacred biography were staged" (47). This statement must be clarified immediately. Neither I nor my husbandjohn has ever hinted, much less claimed, that the Buddhists had a pageantry tradition akin to Christian passion plays.124
My own suggestion relates to the idea of celebrations of the main occasions of the Buddha's
138
SUSAN L. HUNTINGTON
life, some of which may have taken place at the actual sites of the events, rather than formal dramatic productions equivalent to passion plays.125 In my discussion of the Säňci relief (fig. 2), I have called the scene "a processional celebration."126 Although later in the same discussion I use the term pageant—I say that some of the figures in the scene are not devas but rather actors in a pageant that recreates aspects of the event of Säkyamuni's departure—I used the term pageant in the dictionary sense, not as the equivalent of a passion play. Specifically, although the literal translation of the Middle English pagyn or padgeant is "scene of a play," the current English definitions are: 1) a mere show or pretense; 2) an ostentatious display; 3) a show or exhibition, especially an elaborate, colorful exhibition or spectacle, often with music, that consists of a series of tableaux of a loosely unified drama; or 4) a procession, usually with floats.127 Any of diese definitions could suit what I intend. Dehejia's extrapolation and exaggeration of my suggestion cannot be justified. Further, when I used the term actorva. reference to the figures carrying the horse in the relief, I intended to suggest that these individuals were participants playing a role in this circumstance and not that these individuals were necessarily professional actors. What I intended to convey was that there may have been reenact-ments as part of the celebration of events at sacred sites, possibly on anniversaries of Buddha life events. While Dehejia has not found in her "detailed search through Buddhist texts . .. any evidence of pageants" (65 n. 13), she has no means of concluding that her search and my search were identical and that I have not found evidence to support my proposal. It is curious that, withoutwaiting to see whatmaterials I present in my book, Dehejia felt confident that her own search would produce the same results as mine. Further, without knowing that every relevant literary work has survived, is available, and has been examined for this information, it is impossible to substantiate a claim that there is an absence of literary evidence.
Dehejia maintains the long-held view that the scene is a representation of the Great Departure of the Buddha (actually Bodhisattva) prior to his enlightenment. The traditional interpretation assumes that the relief represents the actual event and that the absence of the human figure of the Buddha-to-be on the horse is evidence of ani-conism. As I have discussed above, Dehejia believes thatin this scene the Buddha-to-be is shown
by emblems, but I have demonstrated that the Buddha-to-be's presence, if it is intended in the scene, would have to be indicated by his absence, not by an emblem. Dehejia observes that if the scene is a reenactment, the fact that there is no human actor on the horse is anomalous (48). But, if the scene represents another subject, the absence of the figure might not be problematic. Further, if it is necessary that there be no anomalies for a scene to be correcdy identified, then the anomalies in the composition that undermine Dehejia's interpretation of the scene as an aniconic rendering of the Great Departure must also be considered.
Let us examine this relief in detail. At the left is a palace, or perhaps a city, peopled by men and women who are awake, not asleep. According to the generally known accounts regarding the Bud-dha-to-be's departure, he left the palace quietly atnightwhen the inhabitants were asleep.128 The secrecy of the departure is a prevailing emphasis in textual accounts of the event.129 Thus, if this composition represents the Great Departure, this aspect of the scene is incompatible with textual accounts. In front of the palace/city, a woman bows in apparent reverence before a pond with lotuses and ducks in it, again a theme that is not part of the usual Great Departure accounts. Ap-parendy exiting from the city or palace gate is an entourage that includes a horse carried aloft by bearers (the horse's visible front right leg is clearly above the ground). The manner in which the horse is carried is clarified the second and third times it appears; apparently, six bearers (three on the visible side of the horse and presumably three on the other side) bear the animal aloft. Textual accounts of the Great Departure usually mention four devas carrying the horse's hooves.130 The representation of six human males rather than devas not only contrasts with literary evidence but differs from what is found in Bactro-Gandhäran depictions of the Great Departure, where devas are clearly indicated.131 Included in the entourage is a small male figure carrying a vessel; he appears just in front of the horse the first two times the horse is shown.132 After the horse is shown the second time, the entourage arrives at a tree enshrined in a vedika; the liquid contained in the vessel is apparently being used to lustrate the tree.133 From the arrangement of figures around the tree, it may be surmised that the entourage has stopped to honor the tree and that the tree is not incidental to the subject matter. Indeed, since the tree occupies the central posi-
ANICONISM AND THE MULTIVALENCE OF EMBLEMS: ANOTHER LOOK
139
tion in the composition—comparable to that of the bodhi tree and the Rämagräma siüpa in the Asokan compositions—it may be inferred that it is a crucial element Yet this component of the scene—the veneration of a tree—is not part of the traditional versions of the Buddha-to-be's Great Departure known to me, nor is it mentioned by Dehejia. The procession apparendy then continues, with the horse being shown a third time borne by the same six figures (one of the figures on the unseen side of the horse, the most forward one, is partially visible). The entourage arrives at the site of a large pair of footprints. The horse, no longer being carried, and a small male figure (perhaps the vessel carrier?) are placed direcdy in front of the footprints, widi the male human kneeling in veneration. Just below this scene the horse is shown again, walking rather than being carried by bearers and accompanied by members of die entourage, all of whom are turned back as if returning in the direction from which they had come. In light of die importance of the horse that I suggest below, it is notable that the human male figures at the right gesture respectfully to the horse.
Dehejia does not discuss the presence or significance of the tw"o prominent nonfigurative motifs in the composition—the central tree and the footprints at the right, which she might claim serve as emblems. Yet I suggest that these are crucial clues to the correct interpretation of the activities in the scene, as I will discuss below.
Another important element in the composition that I have not mentioned is the umbrella, which appears five times. Dehejia asks why the artist would have "resorted to the extraordinary device of portraying a parasol hovering at an appropriate height above empty space over the horse [if the scene represents a reenactment]" (48). Acareful examination of the position of the umbrella the first four times it appears suggests that it is not being held above an invisible rider at all. The first time the umbrella appears it is held in front of the horse, the second time it is held to the side of the horse (and by a woman, not a deva), the third time it is tilted so that it could be read as being above the head of the horse or still to the side of the horse, and the fourth time the umbrella is lowered so that it seems to preclude enough space for a rider. The fifth and final time the umbrella appears it has been placed in the ground or into a stand next to the venerated footprints. When the horse is shown for the fifth time turning back towards the palace/city, the
umbrella is not being carried along with it but has been left at the place of footprint veneration.
If one accepts my suggestion that one of the main themes of India's early Buddhist art was acts of piety (towards the Buddha especially) by devotees, such as kings like Prasenajit and Ašoka as well as other devotees, then the subject I propose for this representation may be another demonstration of this theme. My present thinking regarding this composition is that it may be intended to honor the Buddha-to-be's horse, Kanthaka, who not only carried him away from the palace at the time of the Great Departure but who had been a lifelong companion and devoted servant of the young prince from the moment of their simultaneous births. Buddhist literature talks of a "turning around place," which the Buddha-to-be indicated was to become the Kan thakanivattana shrine (honoring the "turning around of Kan.thaka") .1M According to Thomas, the shrine of the Turning Back of Kanthaka was probably a real shrine that was known to the authorities upon which the Pali commentator of the story had based his text.lss The centerpiece of the composition is a tree shrine like those often used in India to mark sacred spots. The Säňcí scene clearly seems to emphasize the horse, not the Buddha-to-be, and since an important element of the composition is that the horse turns around, the scene may reen-act Ran.thaka's role in the departure in order to emphasize Kanthaka's devotion and loyalty to the Buddha. The place of the footprints, which turn back toward the palace/city, may represent a commemorative spot, like that of the tree, perhaps in this case marking the place where the Buddha and the horse parted company.1*6 That the humans accompanying the horse the final time the horse appears display gestures of respect to the animal furdier suggests the important role of die horse in diis composition.
Because of her abiding belief diat the biography of the Buddha was the primary subject of the early Buddhist art, Dehejia rejects the notion that an artistic rendering might show a re-creation of an event rather than the historical event itself. She claims that it "does a disservice to the notion of die religious devotion of the many hundreds of monks, nuns, and lay worshippers who contributed towards the decoration of the Sänchi stúpa... to suggest diat they would build the immense stone structure and then decorate it merely with pictures of a pageant! Surely it was unnecessary to depict the enactment of an event when the artist could easily circumvent that middle step and
140
SUSAN L. HUNTINGTON
depict the event itself" (48). But, if the celebrations themselves had special meanings—as paradigms or reminders, for example—they might have had great importance and priority. Why might a Christmas card show a group of people sitting around a Christmas tree looking happy rather than depict the birth of Christ? Because something different is being communicated— something about the celebration and not the event If one can infer from literature as well as the living traditions of South Asia, the ceremonial and celebratory life in ancient India was extraordinarily rich, and it is possible that these activities inspired some of the subject matter in art. When Dehejia says that it was unnecessary to depict the enactment of an event when the artist could circumvent that middle step and depict the eventitself, she makes a claim that can be substantiated only by proving (not merely asserting) that the life of the Buddha was more important than all other subjects and that other subjects were not even permissible because ofthat prioritization.
StDpas and Relic Worship
Dehejia then embarks on a discussion of the theme of stüpas and their role as repositories for relics (48), a topic I also discuss in my Art Journal article.157 She begins by saying that the "whole138 purpose of going to a stüpav/as indeed to experience the presence of the Buddha through proximity with his enshrined relic" (48). Incorrectly extrapolating from what I have said about the role of relics in Buddhism and Buddhist art, she further states that "the presence of the relic does not thereby preclude the need for stories from the life; in fact, reliving the historic life through viewing narrative sculptures recounting those events would enrich the experience of going to a stupa" (48). While I agree that the presence of a relic does not preclude the inclusion of works of art portraying the life of the Buddha, neither are such life events required; while depictions of Buddha life scenes might "enrich the experience of going to a stüpa," so might the presence of other subjects as well. An "either/or" situation does not exist where a relic of the Buddha was enshrined.
The richness of the role of relics and their implications for Buddhist practices are no t addressed by Dehejia, who selects only one aspect of relics to discuss, namely, whether the presence of a relic obviates a need for Buddha images. In particular, she does not discuss the important distinctions among the principal types of relics recognized in
Buddhism, namely, saňraka, paribhogika, and udde'saka."9 Her generalized reference to relics seems to be concerned specifically with the 'saňraka type, that is, the remains of the Buddha's physical person. A pivotal aspect of my work is my emphasis on the importance in Buddhist practice of paribhogika relics, which include all places the Buddha had ever been and all things he had used or touched. This second type of relic is central to my interpretation of many of the works of art at sites like Säňci and Bhärhut as scenes associated with pilgrimage practices. In other words, it is because paribhogika relics, such as the bodhi tree or the site of Särnäth, have such crucial importance in Buddhist practice that I believe that they are shown in the art—they are not substitutes for preferred Buddha images but important in their own right. Further, Dehejia ignores the third type of relic, udde'saka, which are images or representations of the Buddha, and the notion that, at least in some Buddhist traditions, udde'saka are considered to be less important than the other two types.140
It should also be noted that the veneration of relics in Buddhism is not limited to those of Šäkyamuni Buddha. Relics of other past Buddhas, of Bodhisattvas (such as the Dalai Lama, who is revered as an incarnation of Avalokites-vara), of venerated members of the clergy, and others are also treasured and honored. Looking at the phenomenon of Buddhism in its myriad forms and in the many regions where it flourished, it is apparent that the veneration of relics is one of the most important aspects of the religion. It is manifest not only in the early Buddhist culture of India but in later traditions (such as that of the Päla dynasty of eastern India, with its emphasis on the paribhogika sites of the eight major life events of the Buddha), in the Bactro-Gandhära region, in Myanmar (Burma), in Thailand, in Sri Lanka, in China, in Mongolia, in Nepal, and in Tibet.141 Relics and their veneration provide a vitally important focus in the religion, and much of the art and architecture must be understood in light of this truth.142
Aniconism and the Depiction ovJatakas
Using the same line of reasoning—that the presence of a relic does not obviate the need for representations of the Buddha's life events— Dehejia turns to the subject of jatakas in early Buddhist art. She says: "Equally, it is not valid to assume that the prevailing religion during this early period emphasized the perfection of virtues
ANICONISM AND THE MULTIVALENCE OF EMBLEMS: ANOTHER LOOK
141
narrated in the jätakas, thereby obviating the need for life scenes" (48).143
However, it is incorrect to presume that I would claim that jätakas were a substitute for Buddha life scenes, thereby setting up an "either/or" situation. In fact, my view is quite the opposite. Instead, it is previous scholars attempting to understand early Buddhist art in lightof the prevailing aniconic theory who have viewed jätakas as surrogates selected because the presumed primary subject matter—scenes from the life of the Buddha showing him in human form—was assumed to be forbidden. I contend that the jätakas convey a series of important Buddhological messages in their own right, particularly since the literary conventions used for jätakas emphasize their didactic importance rather than their role solely as sources of biographical information for the Buddha. While on a literal level they might be seen as biographical, on a didactic level they are paradigms of the Buddhist pilgrim's progress toward an enlightened state. Each jätaka consists of three sections: the first is the explanation of the catalyst that causes the Buddha to tell that particular story as a related lesson; the second is the story itself; and the third and final section is the meaning or lesson of the story. Western scholars have traditionally emphasized the middle of the three, that is, the narrative, and, when interpreting jätakas on art monuments, have considered their tasks fulfilled when they have identified the main actors in die scene and recounted the events of the story. But I believe that, Buddho-logically, it is the underlying lesson that is most important. Thus, the jätakas, along with scenes of Buddhist piety, the devotional scenes atBuddhist sacred sites, and other subjects, served to deliver the communicative messages of the monuments. They are not substitutes for biographical scenes of the Buddha's life, and, therefore, Dehejia has inverted the point. Further, her reluctance to acceptan emphasis on the perfection ofvirtues in the "prevailing religion" almost argues against the importance of one of the central goals of the Buddhist religion itself—the spiritual perfection of the individual (48) .As paradigms of the Bodhi-sattva's perfections mastered during his many lifetimes, the jätakas are among Buddhism's most poignantremindersofthe ultimate pursuit of the religion.
The apparent theme of Dehejia's paragraph, which began with a discussion of stúpaš as repositories for relics and then turned to the issue of the role of jätakas, becomes clear at the end when
she states: "The proposition that the early art of India144 was not primarily concerned with the biography of the Buddha is difficult to sustain" (48).145 Yet Dehejia does not demonstrate this either in her discussion of relics and jätakas or in the course of her article. My research suggests that there are many cases that, while usually seen as representing life scenes of the Buddha, are depictions of something else altogether. The scene from Bhärhut showing King Prasenajit at a temple with an enshrined wheel (fig. 19), I have argued, represents a story about King Prasenajit, not an event in the life of the Buddha. Likewise, the two Ašokan compositions are not biographical scenes of the Buddha, the jätakas are not scenes of the Buddha's last life, and I suggest that there are many other compositions among the corpus of early Buddhist art that are not depictions of life events of the Buddha. That Dehejia confuses this issue is clearly demonstrated when she precedes her discussion of the two reliefs about King Ašoka with the statement that "It is difficult to sustain the argument that early Buddhist artists and patrons were not primarily concerned with the biography of the Buddha" (56-57).146 She then follows that sentence with examples of two sculptures that portray the life of Ašoka, not the Buddha. By discussing the two Ašokan examples, she is inadvertently supporting my viewpoint—that the early Buddhist art of India is not solely concerned with the biography of the Buddha. If Dehejia is implying that Asoka's visits to the sites of Bodh Gayä and Rämagräma are part of Buddha's biography, not the story of Ašoka, then I cannot agree. Certainly, Ašoka visited these places because of the Buddha and his devotion to the Buddha, but that is not the same as saying that his visits at sacred sites or scenes showing these visits are demonstrations of the life of the Buddha. She continues this theme by stating that "On the contrary, scenes from the Buddha biography took pride of place in the decorative scheme of the first extensively decorated stüpa at Bhärhut" (48)147 and then enumerates where and how many such scenes she identifies on the Bhärhut monument. However, if her evidence that these are life scenes is no stronger than that which she presents in her article, then I propose that her numbers are likely to be inaccurate.
Time and Place
Without clarifying where the idea of "time and place" was introduced and how it applies to the
142
SUSAN L. HUNTINGTON
early Buddhist art of India, Dehejia addresses (52) the following proposal I have made about some reliefs:148 "Two observations may be made about reliefs that actually portray Buddha's life events: (1) the place being shown is the place where the event occurred, and (2) the timeoi the activity depicted in the composition is the time of the event itself. These two conditions generally are not present or even implicit in reliefs of the 'aniconic' type."149 In response to this idea, Dehejia states that "firtha proponents150 maintain that these elements [time and place] are always explicitly indicated in iconic life scenes but generally absentin aniconic renderings, which are hence, presumably, to be understood as sacred sites" (52) .1S1 She tries to prove the incorrectness of my position by showing an "aniconic" scene in which she sees references to time and place. With regard to a relief at Säňcl she states: "If place and time are key clues indicating a life scene, both elements occur in the aniconic portrayal at Sän-chi"(53).
My intention was not to claim that time and place are not indicated in the "aniconic" scenes butrather that the time of the activity of the scene is not the time of the Buddha life event and that the sacred site being shown may not be one where a Buddha life event occurred. Thus, I have suggested that, while a scene like King Prasenajit's visit to a shrine (fig. 19) may take place at the very site where a Buddha life event occurred, the time at which King Prasenajit's visit occurs is not the time of the Buddha life event but simply the time at which the king made the visit. Further, since there are many sacred sites throughout the Buddhist world that are revered even though the Buddha may never have visited them, some compositions may be depictions of such places. Dehe-jia's own suggestion that the two scenes from Amarävatí that she illustrates (her figs. 20 and 21) show local sacred sites, not places associated with Buddha life events, seems to argue for my case. What I intended to communicate in my statement was that, unless a scene is clearly a Buddha life event, one cannot presume that the place or the time being depicted have any reference at all to the Buddha or one of his life events.
Summation Regarding Dehejia's Article and the Theory of Aniconism
In the foregoing, I have attempted to clarify some of the points that Dehejia raises in response to ideas I have presented about the problem of aniconism as a general theory for the interpretation of early Buddhist art. It is clear that many of
the objections and counterarguments she presents arise from her misunderstanding of my work and incorrect extrapolations based on statements I have made. The problem is further exacerbated by the fact that she bases her argument in part on what she anticipates I will say in my forthcoming book. Because the topics that she draws from my work are not clearly associated with her main theme of multivalency, it is not always evident how they relate to her general effort to reinterpret the art. For example, her discussion of jätakas is directly aimed at statements I have made but does not address the problem of relating this type of scene to her multivalency theory. The reader is left to wonder how the jätakas, an important narrative subject on the early monuments, can be reconciled with her theory since they do not seem to be multivalent and do not use emblems. The fact that Dehejia does not use a systematic and comprehensive approach to critiquing my ideas, but rather selects only a few themes, creates problems in clarity for the reader, misrepresents my work, and undermines the presentation of her own multivalency theory.
Dehejia's Conclusion
In her conclusion, Dehejia claims that while Alfred Foucher "misstated the nature and extent of aniconism, he was certainly accurate in perceiving its existence" (64). In the first part of this quote, she makes precisely my point: I believe that Foucher (and others who have espoused the aniconic theory) have misstated the nature and extent of aniconism. Whether he was accurate in perceiving its existence is still to be determined. But as I have said already in publication, even if a few aniconic images do exist, the probability that the majority of representations from this corpus are not aniconic means that the theory of aniconism cannot be used as a universal explanation of the art of the period. Dehejia continues the first paragraph of her conclusion by making statements that agree with points I have made and that undermine the evidence for the aniconic theory: 1) what is now known to be the lack of correlation between the old Hínayäna/Mahäyäna model and aniconism and 2) the existence of numerous "site scenes" among the repertoire of early Buddhist art. With slight changes, I could have written this paragraph myself.
Dehejia's final paragraph addresses her theme
ANICONISM AND THE MULTIVALENCE OF EMBLEMS: ANOTHER LOOK
143
of multivalency. She states: "One must accept the multilayered significance of many éarly bas reliefs and recognize that more than one meaning may have been intended by the artist, as well as read by the ancient beholder" (64). She notes thattwentieth-centuryviewersmightfinditstrange that a composition showing the Buddha's enlightenment might include a shrine erected after the event, and that they might find it anachronistic that a panel showing Ašoka's visit to Bodh Gayä would include a shrine built only after his visit. "Yet," she concludes as her final sentence, "artists and devotees of the first century b.c. probably viewed such a scene as a perfectly reasonable way to present a reminder of both the sacred site at Bodh Gayä and an event in the life of the Buddha or of King Ašoka" (64) .152 The interpretations I have offered regarding these two scenes (figs. 4 and 11) suggest that the choice of these two examples is unfortunate as an argument for Dehejia's position.
Position Statement and Conclusions
Readers of this pair of articles by Dehejia and me are probably well aware of the history of the aniconic theory ,jts pivotal role in the interpretation of the early Buddhist art of India, and my challenge of the theory's validity as an overarching explanation for the subject matter of early Buddhist art. The position that Dehejia takes is that aniconism is still a viable theory for the interpretation of early Buddhistart, although she agrees, following my work, that it has been too widely applied. Nonetheless, our positions diverge in that, as I infer from her article, she claims that aniconism should still be considered a dominant aspect, while I propose that aniconism is not a satisfactory general theory by which to interpret the art and may not even be applicable to more than a small fraction of the early Buddhist art. With the recent invalidation of the extemaljustification for the practice of aniconism in art, I believe that it is imperative that the theory be reexamined. The now-recognized fallaciousness of the long-held Hlnayäna/Mahäyäna model and the acknowledgment that Hlnayänists used images of Buddhas force scholars along new paths of inquiry.155 If it can be determined that "aniconism"154 does not exist in the art, then new explanations must be found to interpret the art. Or, if the art is found to contain representations that must truly be characterized as "aniconic," then new explanations—for example, Buddho-
logical, social, cultural, or artistic—must be discovered to provide a rationale for this practice. Further, because "aniconism" might be only one of many other themes in the early Buddhist art of India, its role in the communicative message of the art must be correlated with the many other subjects depicted in the surviving art.
One of the areas of strongest disagreement between Dehejia and me relates to the underlying meanings and the communicative message of early Buddhist art. While she retains the traditional belief that early Buddhist art was primarily concerned with the historical Buddha Šäkyamuni and the events of his life, I believe that a different message, or perhaps several different messages, comprise the more likely content of the art.155 Looking at the artistic compositions not only as individual entities but also as components of an overall message or series of messages, I suggest that the artists and patrons responsible for the art were reinforcing a number of important didactic, spiritual, religious, and social themes within Buddhist culture. These themes might include the idea of generosity (exemplified by Anätha-pindada's gift of the Jetavana grove); religious piety by lay persons (as seen in the many illustrations of lay devotions at sacred sites); the belief that all sentient beings, not only humans, are moving along the religious path (exemplified by the worship of the Rämagräma relics by the nägas, the Bhärhut reliefs showing the devas and apsara-sas, and the many scenes showing devotional activities by elephants, monkeys, and other animals); conversion and reform (exemplified by the inclusion of the story of Ajätašatru); paradigmatic Buddhist kingship (exemplified by illustrations of life events of individuals like Ašoka and Prasenajit); and the perfection of moral and spiritual qualities (exemplified by the popular inclusion of jätaka stories on the early monuments). While I would never suggest that Šäkyamuni Buddha did not play a vitally important function in early Buddhism as a role model and in other capacities, I suggest that the religious message of the art also included a range of instructional and soteriological/methodologi-cal models addressed to the ordinary being. (And here, when I refer to the "ordinary being," I do not want to be misconstrued as referring specifically to a lay person. Rather, I mean "ordinary" in the sense of the vast majority of living creatures, including animals as well as people, who, according to the tenets of Buddhism, are still struggling to develop their spiritual qualities.) Dehejia's abiding belief in the symbolic and
144
SUSAN L. HUNTINGTON
emblematic role played by nonfigurative motifs and subjects in the art has not been supported through either internal evidence within die artistic compositions or by external documentation. I suggest that the expectation—not the actuality— that these elements play a symbolic role is an important reason why the idea persists in the interpretation of early Buddhistart. This is not to say that there are no symbols and emblems in early Buddhist art, for clearly there are many. The very pillars, cakras, and some of the other motifs discussed by Dehejia can serve as emblems, but this is not, I suggest, their principal role when they are depicted in many of the narrative scenes on monuments such as the stüpas at Säňcí and Bhärhut.156
By selecting out certain themes I raised in my publications, Dehejia does not convey the cohe-siveness of the work that I have published thus far—mainly visible in the Art Journal article. Although she addresses a number of points that I make in my Art Journal article and the Art of Ancient India, as enumerated in the section on "Dehejia and the Theory of Aniconism" above, she removes ideas from their contexts and fails to undertake a systematic critique of my work. The most reasonable solution (s) to the aniconic problem must také into account the strengths and weaknesses of both the traditional theory and any new ideas that are presented. Yet her arguments ignore the substantial evidence I have presented about a number of important issues. I invite the interested reader to consult my "Early Buddhist Art" article, and particularly my discussions of the following issues: sartraka, paribhogika, and udde'sa-ka relics and the related questions about the prioritization of relics versus images; Sri Lankan Buddhist practices relating to the sixteen sacred sites, which, I believe, are reflective of and strongly related to some of the practices and religious concerns revealed in early Buddhist art; the role of pilgrimage in early Buddhist religious practice; the problems of varied interpretations of the Bhärhut inscriptions; the evidence of inscribed reliefs, such as those from Bhärhut, showing and labeling the bodhi trees of some of the former Buddhas; the certainty from archaeological evidence that Buddha images were being produced (and have survived from) the very period of production of the so-called aniconic works of art; and, finally, the crucial issue of historiography and how it relates to the theory of aniconism.
While Dehejia and I agree that there are some works that are, or, in my opinion, might be,
aniconic, and that there are some works that clearly are not, we differ vigorously in our respective analyses of specific works of art. Some of this difference in interpretation may arise from what may be our respective propensities for wishing to claim one or another individual specimen for the aniconic or non-aniconic camp. But some of this difference must also be seen as reflecting a transitional point in our knowledge about early Buddhism and its art, when many of the old presuppositions have been stripped away but new generalizations are notyet in place. I suggest that Dehejia's multivalency theory represents an admirable attempt to reconcile the seeming invincibility of the institutionalized aniconic theory with the flaws and weaknesses that theory embodies, some of which I have recently addressed. I suggest that Dehejia's view represents amidway point in a transition that evinces a reluctance to abandon the old while recognizing the validity of the new. Dehejia has essentially proposed a compromise position in which elements of the old theory that still seem to make sense are reconciled with new information that changes the direction of earlier thinking on the subject.
I believe that scholars of Indie art should continue to move forward on the issue of aniconism and seek a new generalization or series of generalizations that accommodate the importance of a variety of themes evident in the art, including but not limited to what I have identified as scenes of pilgrimage and lay devotion (which Dehejia has dubbed the "site" scenes). The new generalizations should be consistent with the patterns of belief and practice of the Buddhism of the corresponding period, as documented by texts, inscriptions, and other sources. Further, the new generalization (s) must move beyond consideration of individual compositions and consider their potential role in a larger context of a series of images or even an entire monument. As I have tried to demonstrate in this article, I do not believe that the multivalency theory offered by Dehejia fulfills these criteria. Whether my own work will help move scholarship further along the path to these goals, of course, remains to be
seen
157
A Closing Thought
Although it is unlikely that any individual can be completely unfettered by the values and concerns of his or her own culture, scholars today
ANICONISM AND THE MULTIVALENCE OF EMBLEMS: ANOTHER LOOK
145
have become aware of the ways in which our personal vantage points bias our interpretations of other contemporary cultures and those of the past. Concomitantly, scholars in the humanistic and social science disciplines recognize the need to understand other cultures in their own right, rather than according to the standards and criteria of the observer. While recognizing the desirability of this ideal does not necessarily liberate us from lapses in objectivity, this awareness at least helps us to confront some of the biases that color our interpretations.
In the historiographic portions of my forthcoming book, I plan to suggest—as I did in the paper I presented at the 1991 conference of the American Committee for South Asian Art—that the origination, perpetuation, and passionate advocacy of the aniconic theory is deeply embedded within a matrix of Western cultural viewpoints that were transferred to, or perhaps imposed upon, the Indie situation. I will suggest, for example, that nineteenth- and early twentieth-century writers may have been predisposed by their knowledge of the Judaeo-Christian-Islamic heritage, with its recurring disputations over whether to represent God in the art, to puzzle over the apparent avoidance of Buddha images in early Buddhist" art. At the same time that the religious heritage of Western scholars may have colored their interpretations of ancient Indie culture, the political relationships between the Asian and European worlds must have played a significant role in the formulation of their ideas, causing Western imperialist viewpoints also to affect the ways in which European scholars per-
ceived early Buddhist art Alfred Foucher, for example, explains the apparent absence of early Buddha images by claiming that the Indians were not clever enough to think of the idea of an image themselves; he credits the Western (Greco-Roman) culture of the Indie northwest for what he considered to be this superior innovation.158 Such views have influenced the study of early Buddhist art for more than a century.
Two terms from the discipline of linguistics that have now become current among disciplines involved in crosscultural studies seem pertinent here, and, I believe, should be at the forefront of every scholar's mind as he or she embarks upon a study of the past.159 The first term, etic, is defined as "of, relating to, or involving description of linguistic or behavioral phenomena considered in isolation from a particular system or in relation to predetermined general concepts."160 The second and opposing term, etnic, is defined as "of, relating to, or involving analysis of linguistic or behavioral phenomena in terms of the internal structural or functional elements of a particular system."1611 suggest that the study of early Buddhist art has been dominated by an etic approach for more than a century, in which a predetermined, general concept—namely, the theory of aniconism—has influenced the way in which the art has been interpreted. Itis time, I propose, that we examine the art for what is there and look beyond the expectations that may have their origin largely in the imaginations of those who encounter and interpret, but did not create, the art.
146
SUSAN L. HUNTINGTON
Notes
I am grateful to Frederick M. Asher, Nancy E. Eder, John C. Huntington, Miranda Shaw, and especially Janice M. Glowski for reading drafts of this manuscript and for their advice and help throughout the writing process.
1.     Vidya Dehejia, "Aniconism and the Multivalence of Emblems," Ars Orientalis2l (1991): 45-66 (subsequent page references given parenthetically in the text).
2.     Dehejia defends her discussion of my work at this stage on the basis that the ideas have been presented in a volume that has "entered the textbook repertoire" (65 n. 4). She refers to my Art of Ancient India. See Susan L. Huntington, with contributions by John C. Huntington, The Art of Ancient India (New York and Tokyo: John Weatherhill, Inc., 1985). I do not understand this rationale, particularly since the ideas are offered in thatbook as tentative new interpretations. Further, in my article "Early Buddhist Art and the Theory of Aniconism," Art Journal 49, no. 4 (Winter 1990): 401-8,1 explain that I will present the full range of my research in a forthcoming book on the early Buddhist art of India.
'%
3.     See S. Huiitington, "Early Buddhist Art," 408 n. 43, in which I'state: "At this time, I am unable to predict whether there are indeed some images that require a Buddha figure and must be seen as truly 'aniconic' in the sense that they employ a symbol as a substitute for what should be an anthropomorphic representation. However, even if a few images are truly aniconic, the vast majority are not, and the role of'aniconism' has been vastly overemphasized, ultimately leading to the misinterpretation of most of the extant art." The definition of aniconic that I used in the Art Journal article, that the Buddha might be referred to by symbols, is expanded in the present article to include depictions that may refer to him through his absence. See below.
4.     S. Huntington, Art of Ancient India, 70-71, 72-73, 87, 98-99,100.
5.     I am certain that Dehejia was in the audience for the 1988 paper, since she chaired the session in which it was presented and led the ensuing discussion. I do not know whether she was in the audience for the 1991 presentation.
6.     Although my husband, John C. Huntington, has not been working on the "aniconic problem" perse, he has incorporated some of my ideas into his work and has published a number of articles that are
pertinent to the topic, some of which are referred to by Dehejia. I include mention of John C. Huntington's work here since Dehejia seems to conflate us and our work, as when she refers to my work and then says, "For further understanding of /A«Vargument" (65 n. 4; italics mine). For the most relevant publications by John C. Huntington see his The Origin of the Buddha Image: Early Image Traditions and the Concept of Buddhadarsana-punyä," in Studies in Buddhist Art of South Asia, ed. A. K Narain (New Delhi: Kanak, 1985): 23-58. He later published a Chinese dated Buddha image that he discussed in that article: John C. Hunting-
; ton, "A Note on a Buddha Image from China dated to the year 36 of the Pre-Christian Era (former Han Chien Chao third year)," Laut Kala 22 (1985): 27-31 but retracted it since the piece was found to be a forgery. See his letter to the editor in Laut Kala 23 (1988): 44-45. Although this piece has been shown to be a forgery, the discovery regarding its authenticity does not alter the basic premise of his work or negate the other strong evidence that he presents. See also John C. Huntington, "Sowing
v the Seeds of the Lotus: A Journey to the Great Pilgrimage Sites of Buddhism," pt 1 [LumbinI andBodhgayä], Orientationslß,no. 11 (Nov. 1985): 46-61; pt. 2 [Rsipatana Mrgadäva], OrientationsVl, no. 2 (Feb. 1986): 28-43; pL 3 [Šrävastl and Säiikäsya], Orientations 17, no. 3 (March 1986): 32-46; pt. 4, OrientationsT7, no. 7 (July 1986): 28-40; pt. 5 [Kušinagara, Appendices, and Notes], Orientations 17, no. 9 (Sept. 1986): 46-58.
7.   S. Huntington, Art of Ancient India, 642 n. 18.
8.   The paper is being published in a volume of conference papers by the Rijksmuseum in Amsterdam that is being edited by Pauline Lunsinghe Scheurleer.
9.  John C. Huntington, "Pilgrimage as Image: The Cultof theAs.tamahäprätihärya,"pt. 1, Orientations 18, no. 4 (April 1987): 63.
10.   Dehejia does not seek religious, philosophical, social, and political explanations for what she perceives to be multivalency in the art but relies purely on a literary analogy. I suggest that these avenues of investigation might be equally, if not more, important in attempting to understand multivalency in the art.
11.   There are strict rules for determining where and how to break the words in Sanskrit, but, in spite of the rigidity of the rules, significant play and ambiguity is possible.
12.   The term pun often suggests a humorous double meaning in the English context. I use it here
ANICONISM AND THE MULTIVALENCE OF EMBLEMS: ANOTHER LOOK
147
without any implication that humor is intended. Chinese and Sanskrit are particularly well suited to punning. While in contemporary American English punning is often derided as a low form of humor, in other languages, such as Chinese and Sanskrit, it is a highly respected literary technique.
13.   In many languages, multiple meanings can also be implied by using homonyms, such as dear and deer.
14.   Here Dehejia says: "Equally, a relief may be read both as an event in the life of the Buddha and as the holy site at which that event occurred."
15.   Dehejia credits the artist with the choice of subject matter (57 and passim). However, in ancientlndia the situation was not so clearcut; it is likely that patrons as well as members of the clergy played significant roles in the selection of artistic subject matter.
16.   For example, see S. Huntington, Art of Ancient India, 642 n. 18.
17.   See Susan L. Huntington and John C. Huntington, Leaves from the Bodhi Tree (Dayton and Seattle: Dayton Art Institute and the University of Washington Press, 1990), 104-5. Because the metaphor occurs on imperial copperplates issued by the Päla kings themselves, the comparison is especially compelling, leaving little doubt that the kings themselves were aware of the analogy. While the metaphor is given expression through its literary form in the copperplates and visually in the image type, it represents far more than a mere literary or artistic phenomenon. Rather, revealing as it does the views the Päla kings held about their political power and probably their attempt to legitimize their rule through the authority of the prevailing Buddhist culture, the analogy is deeply embedded within the Päla cultural sphere and can only be fully understood as such.
18.   Although I use the term emblem in some of my writings, I do not use the term in the same situations as Dehejia. While I do believe that there are emblems (and symbols) in Buddhist art (see my "Position Statement and Conclusion" section), I do not agree that the motifs in the compositions she discusses are used emblematically.
19.   Although Dehejia does not use the term nonfigural or nonfigurative in her definition of an emblem, she uses the term nonfigural several times in her text, making it clear that she refers specifically to nonfigurative motifs. See, for example, her emphasis on the nonfigural aspect of the Buddhist wheel (47). The assumed correlation between being nonfigurative and being emblematic is a
cornerstone of traditional aniconic thinking.
20.   She uses the mango tree as one example (45).
21.   For my discussion of Dehejia's phrase "attribute of the faith," see section entided The Emblems of the Tree, the Pillar, and the Stüpa as an Attribute [of the Faith)." See also n. 27.
22.   Corresponding figure numbers for Dehejia's article are cited in the captions so that the reader can correlate our two articles. I am grateful to Vidya Dehejia for supplying photographs of all the works of art published in this article except for figs. 6,7, and 12, which are by John C. Huntington.
23.   Monier Monier-Williams, A Sanskrit-English Dictionary (first ed., 1899; Oxford: At the Clarendon Press, 1960), 919.
24.   See Eugene Watson Burlingame, Buddhist Legends, Translated from the OriginalPali Text oftheDhamma-pada Commentary (Cambridge, Mass.:. Harvard University Press, 1921; rpt. ed., London: Luzac and Co. for the Pali Text Society, 1969), pL 3 [books 13-26], 56-60, for the story. On p. 59, the text specifically names the tree as a iirisa tree and states that the Buddha sat beneath the tree.
25.   It is impossible for me to foreshadow many of the conclusions I will reach in my completed book.
26.   Dehejia herself mentions the empty space above the horse (48, lines 10-11). Rather than recognizing the contradiction the empty space creates for her theory of emblematic representations, she presents the idea of the empty space to argue against my pageantry interpretation.
27.   I have added "of the faith" since when Dehejia first discusses this concept (45), she includes the phrase. The term faith seems antiquated as a description of the complex social, cultural, soteriological, and other aspects of the Buddhist religious complex.
28.   Helmuth von Glasenapp, Buddhism—A Non-theistic Religion, trans, from German by Irmgard Schloegl (New York: George Braziller, 1966).
29.   Her claim that the artist added the label may be debatable; it is unknown whether the artists were also the scribes. Regarding the term vamdate, which Dehejia translates as "bows," see n. 23 and the corresponding text.
30.   S. Huntington, "Early Buddhist Art," 403-4. Dehejia does not state her indebtedness to my work.
31.   S. Huntington, "Early Buddhist Art," 403,408 nn.
148
SUSAN L. HUNTINGTON
20, 21. Dehejia vastly oversimplifies the problems relating to the terms bodhoand bodhi(65-66 n. 20). The issue is discussed in my forthcoming book. Also see Heinrich Lüders, ed., BharhutInscriptions, rev. and supplemented by Ernst Waldschmidt and Madhukar Anant Mehendale, CorpusInscriptionum Indicarum,\o\. 2,pt. 2 (Ootacamund: Government Epigraphist for India, 1963), 94-96.
32.   See also Lüders, Bharhut Inscriptions, 97-100.
33.   Actually, Dehejia's statement is contradictory. She says that the gods came to praise the enlightened Buddha, implying that he was already enlightened, and that the event occurred simultaneously with the enlightenment. Dehejia cites Rajendrala-la Mitra, TheLalitavistara (Calcutta: Asiatic Society of Bengal, 1877), 457 (end of ch. 22 and beginning of ch. 23) as the source of her information.
34.   Gwendolyn Bays, trans., The Voice of the Buddha: The Beauty of Compassion,2voh. (Berkeley, Calif.: Dharma, 1983), 2:499ff.
35.  J.J.Jones, trans.,The Mahavastu, 3 vols. (London: Luzac and Co., 1949-56), 2:318.
36.   The discrepancies between textual accounts remain to be worked out in relation to this scene.
37.   The Lalitavistara enumerates sixteen ways in which the gods of Šuddhaväsä "spoke words to weaken Mära," and sixteen ways in which other gods attempted the same. See Bays, The Voice of the Buddha, 2:499-502.
38.   She does not clarify whether the figures arrive before or after the enlightenment.
39.   Dehejia states: "Since early Buddhist inscriptions [does she mean all early Buddhist inscriptions or only those at Bharhut?] are so easily readable, I have used my own translations to avoid burdening the text unduly with footnotes [sj'cfor endnotes]. In the single instance where variant readings are possible, a note has been added" (65 n. 9). Although Dehejia claims that she has translated all of the inscriptions herself, her wording in the case cited here is curiously like that of Lüders, Bharhut Inscriptions, 100, no. B27. In particular, since Lüders infers the word mimic from the sculpture and an understanding of particulardance forms although the term is not directly used in the inscription, it is strange that Dehejia includes the term mimic in what she implies is her own literal translation. See Lüders, Bharhut Inscriptions, 101, for his reasoning regarding the dance. Dehejia's statement that a variant reading is possible only in one instance [of the Bharhut inscriptions?] oversimplifies the case.
First, there are discrepancies among authors regarding the accurate transliteration from the ancient script; second, while the gist of the inscriptions might be easily discerned, there are points of grammar that can be disputed and that have relevance to the meanings of the inscriptions.
40.   S. Huntington, "Early Buddhist Art," 404.
41.   Although Dehejia does not specify whom she refers to with the words Those who deny the existence of aniconism," I assume she is referring to my work. However, as I have explained above, I have not denied the existence of aniconism but only proposed that aniconism is not an all-inclusive explanation for the art. Throughout Dehejia's article, it is clear that she has confused these two issues.
42.   She does not specify how she is able to predict what another individual might say.
43.   But it might not; see n. 46.
44.   S. Huntington, "Early Buddhist Art," 404.
45.   In spite of this similarity, "aniconic" phases for Hinduism andjainism have never been proposed.
46.   Conversely, the presence of Indra and Brahma is not proof that the scene represents the Buddha life event, if the writings of the seventh-century Chinese pilgrim Xuanzang can be trusted. Xuan-zang records that the early Buddhist kings had constructed ladders at the site of the descent. Atop the ladders, they built a vihära to house a stone image of a Buddha, and flanking the ladders (apparently near the bottom) were figures of Indra and Brahma. The relevant passage of text is cited in S. Huntington, "Early Buddhist Art," 404. See also Samuel Beal, trans., Si-yu-ki. Buddhist Records of the Western World. Translated from the Chinese ofHiuen Tsiang (a.D. 629), 2 vols. (London: Trubner and Co., 1884; rpt., Delhi: Oriental Books Reprint Corporation, 1969), 1:203. Therefore, it is possible that some representations that include depictions of Indra and Brahma might show the sacred site, not the Buddha life event. In the case of fig. 5, because the scene contains a figure that is likely to represent the nun Utpalä (Utpalavarnä), who, according to some sources including Xuanzang, greeted the Buddha on his descent, the composition may show the life event. However, something puzzles me about this particular composition, which I have never seen or examined in person. From the photograph kindly supplied by Martha Carter, it appears as if an upper register might have been broken off, in which case a representation of a Buddha at the top of the ladder may
ANICONISM AND THE MULTIVALENCE OF EMBLEMS: ANOTHER LOOK
149
have been originally included. Such a composition would correlate with Xuanzang's description of the monument that had been erected at the site, which included a vihära with a Buddha image at the top. Thus, it is possible that this would not be an "aniconic" representation of a Buddha life scene but a damaged piece that had originally included a Buddha figure. In other words, I see at least three possible interpretations of this sculpture: it could depict a Buddha life event without a representation of the Buddha, which would make it an "aniconic" scene; it could be an "iconic" representation of a Buddha life scene with the Buddha figure now missing; or it could be a representation of the sacred site with the figures of Indra and Br ahm ä intended to be images that had been installed at the site (possibly, with a now-lost Buddha image at the top). In the latter case, the kneeling figure would need explanation. The argument that images of the Buddha, Indra, and Brahma had been installed at the sacred site may not affect the interpretation I have offered regarding the Bhärhut composition. The textual account that refers to the image mentions all three; if representations of Indra and Brahma were shown, but not the Buddha, then this might be evidence for "aniconism." If none of the figures are depicted, as in the Bhärhut example, this might only indicate the existence of alternative textual traditions or that the statues and ladders reported by Xuanzang in theř seventh century might not have existed as early as the date of the Bhärhut carving.
47.   She refers to John C. Huntington, "Pilgrimage as Image," pt. 1, 56.
48.   It is unclear whether by "earlier" Dehejia is referring to the earlier (that is, pre-Kusäna) artistic tradition that preceded the creation of this carving or to a practice of homage to a wheel-topped pillar that existed prior to the lifetime of the Buddha. In other words, it is not clear whether her use of the term earlier refers to the date of the carving or to the history of the subject matter it portrays.
49.   This situation might, therefore, be parallel to the case of the Prasenajit scene that Dehejia discusses (fig. 19), where she argues for this viewpoint. See my discussion below.
50.   Dehejia does not specify whether she is the first to coin this phrase or whether the idea has been proposed by an earlier author.
51.   Though not acknowledged by Dehejia, this section is heavily dependent upon my work. See S. Huntington, "Early Buddhist Art."
52.   The term pithais more appropriate in the Buddhist
context. I did not know this at the time I wrote Art of Ancient India but corrected the terminology in my Art Journal article.
53.   Rather than saying that the bodhi tree represents Bodh Gayä, or the wheel-crowned pillar represents Sarnäth, I would say that these motifs are key identifying elements in compositions in which sacred sites are depicted.
54.   In spite of my general concurrence regarding the subject matter, I feel that Dehejia overstates the case when she appeals to universal agreement as a source of authority for the identification of these scenes. For example, of her fig. 14 (my fig. 11) Dehejia says that "There is universal agreement that" (57) and with regard to the second scene, which she does not illustrate, she says: "The second Ašokan episode ... as everyone agrees" (57). Even if it could be ascertained that "everyone agrees" about the identification of these scenes, "everyone" could still be wrong. The correctness of an interpretation does not depend upon how many people agree with it or the assertiveness with which it is stated.
55.   Other than those that are, of course, inherent in the tree itself.
56.  John S. Strong, The Legend of King Ašoka: A Study and Translation of the Ašokävadäna [hereafter Aíokävadäna] (Princeton: Princeton University Press, 1983), 266.
57.   Strong, Ašokävadäna, 250.
58.   It is not clear whether the term caitya refers specifically to the type of building we commonly call a caitya (i.e., a caitya hall).
59.   Strong suggests that the statement that the tree returned to its normal state in the Quinquennial story might be an interpolation from the Tisyaraksitä story. See Strong, Ašokävadäna, 266 n. 48. But this cannot explain Dehejia's confusion since the Ašokävadänais very clear regarding which incident is being described. In the Tisyaraksitä story, Ašoka's Queen Tisyaraksitä becomesjealous of "bodhi," whom she mistakes for another woman that she assumes Ašoka loves. Wishing to destroy "bodhi," Tisyaraksitä asks another woman to put a thread on the tree to cause it to wither. When the queen learns that bodhiis not another woman, she regrets what she has done and requests that the thread be removed. The woman who had assisted the queen previously carried out this newer instruction and also watered the tree with a thousand pitchers of milk a day until the tree was restored. See Strong, Asokävadäna, 257-
150
SUSAN L. HUNTINCTON
58. According to this version of the Tisyaraksitä story, the tree was lustra ted by the woman who had caused the tree to wither, and, therefore, the Säňcl relief, which shows Ašoka, not a woman, could not be the same. Based on the Asokävadäna, the Säňcl relief is more likely to represent the Quinquennial Festival.
60.   See Alfred Foucher's writings in John Marshall and Alfred Foucher, The Monuments of Säňchí, 3 vols. (Calcutta: Manager of Publications, 1940; rpt., Delhi: Swati Publications, 1982), 1:212.
61.   The inscription on the stüpa depicted in the center of the relief does notidentify the subject of the composition but is dedicative in nature. It reads: "Aya-Cudasa dhamakathikasa ateväsino Balamitrasa dänarň" (The gift of Balamitra, a pupil of the Preacher of the Law Aya-Cuda [Ärya-Cüda or Ärya Kshudra]). See Marshall and Foucher, The Monuments of Säňchi, 1:342, inscr. 399; vol. 3, pi. 134.
62.   Strong, Asokävadäna, 111-12.
63.   This does not mean that there is no richness of meanings or layers of meanings or metaphoric meanings but rather that there is no direct and overt use of symbols or emblems as equivalents of something else in this context. Further, I do not intend to\ imply that there is not a "higher" or paradigmatic meaning.
64.   This theme and others that I mention briefly in this article will be developed and substantiated in my forthcoming book. I mention them here not with the idea of offering my proof but to enable the reader to understand the different direction my work takes from previous interpretations of the artistic remains.
65.   Strong, Asokävadäna, 113. Strong also suggests that the Rämagräma story demonstrates Ašoka's imperfection since he was unable to obtain the relics from the nägas, therefore rendering his collection of the relics incomplete (p. 113).
66.   Strong, Asokävadäna, 109-10. The number 84,000 is a conventional Buddhist number. The actual number of stüpas constructed by Ašoka is unknown. This actprobably also consecrated Ašoka's empire with the ultimate symbols of Buddhist authority, a feat that later kings throughout the Buddhist world attempted to duplicate, though never so extensively.
67.   Strong, Asokävadäna, 221 and 110, especially for the significance to Ašoka regarding the relics.
68.   S. Huntington, "Early Buddhist Art," 405.
69.   Dehejia says that some reliefs show the stüpa and the pillar as memorials, but she illustrates not only a stúpa and a pillar but also a bodhi tree as examples of the type. The reader is left not knowing whether the stüpa and pillar are used as examples of types or whether Dehejia had neglected to list other possibilities (57—58).
70.  I would modify this statement to say that these sacred objects might indeed indicate the presence of the Buddha at the site, but they are not substitutes for his physical form.
71.   S. Huntington, Art of Ancient India, 100, and fig. 6.11. Dehejia does not cite my interpretation of the relief; however, she criticizes an interpretation offered by John C. Huntington (66 n. 35).
72.   S. Huntington, "Early Buddhist Art," 407.
73.   I have cited the passage in the Mahäparinibbäna suttanta wherein the Buddha entrusts the relics to the laity, not the clergy. S. Huntington, "Early Buddhist Art," 408 n. 34. See also T. W. Rhys Davids, trans., "The Mahä-Parinibbäna Suttanta," in Buddhist Sullas, Sacred Books of the East, vol. 11 (Oxford: Oxford University Press, 1881; rpt., Delhi: Motilal Banarsidass, 1968), 131-36. Dehejia acknowledges the same passage (59) but states that Buddhological research shows that the clergy were also involved in relic worship. As I explain in the text, I do not argue with this evidence but refer only to what is portrayed in the art.
74.   S. Huntington, "Early Buddhist Art," 403.
75.    See S. Huntington, "Early Buddhist Art," 407.
76.   A major form of documentation for the extensive-ness of pilgrimage and emphasis on sacred ptthas comes from the many pilgrims' accounts and records that survive. In S. Huntington, "Early Buddhist Art," 407, for example, I quote the famous thirteenth-century monk Dharmasvämin's poignant account of his visit to Bodh Gayä.
77.   This is a theme I discuss in detail in my book.
78.   See, for example, S. Huntington, Art of Ancient India, figs. 8.24 (one figure to each side of the lotus pedestal), 10.18, 10.19, and 12.18.
79.   Dehejia is counting according to the images she reads that way; I am counting others into the group. She also does not explain whether, for
ANICONISM AND THE MULTIVALENCE OF EMBLEMS: ANOTHER LOOK
151
example, at Bhärhut she means there were six to eight total or six to eight in the approximately one-fourth of the monument that has survived.
80.   The meaning of this portion of Dehejia's sentence is unclear to me. Does she mean "emphasis on life of the Buddha"?
81.   Huntington, "Early Buddhist Art," 405.
82.   The examples that Dehejia cites in her text to support her viewpoint do not seem to lend credence to her statement. That is, Ašoka's interviews with beings at the site who had seen the Buddha during his lifetime only suggest Ašoka's desire to learn more about the Buddha but do not prove that Ašoka's pilgrimage emphasized the Buddha's physical form.
83.   Strong, A'sokävadäna, 244. Ašoka was not under the illusion that he would see the Buddha.
84.   What does it mean to "experience the Buddha himself in all his glory"?
85.   S. Huntington, "Early Buddhist Art," 402-3. See also below regarding "time and place." Dehejia does not refer'to my work regarding this image.
86.   Dehejia suggests (60) that the Buddha image in the second sculpture is made of metal, but it is unclear how thiican be inferred from the drawing.
87.   How can the fact that a stüpa shows a circular medallion provide evidence that the scene represents a local site? It should also be noted that the AmarävatI remains have notyielded medallions as separate roundels but rather that depictions of roundels appear in other sculptural compositions.
88.   Because Dehejia uses the phrase "of the faith" in her text (45), I include it here. Also see n. 27 above for my comments regarding the use of the term faith as a description of the Buddhist religious complex.
89.   She does not specify which tree, but it may be presumed that she means the bodhi tree here.
90.   At this point in her text, she does not suggest any meanings for the stüpa. Because she has used the triad of the tree, the pillar, and the stüpa throughout her article, it is unclear whether Dehejia has deliberately avoided telling what the stüpa represents here or if this was just an inadvertent omission. As mentioned previously, it is also unclear whether she intended to omit the possibility of other motifs, such as the cakra, the throne, and the
parasol. Further, she assumes thatSärnäth and the first sermon would always be shown by a pillar, not a cakra, but this is not documented.
91.   See S. Huntington, Art of Ancient India, figs. 6.4 and 6.10 for illustrations. Some of the compositions on these blocks are clearly scenes of lay worship at sacred sites. See Marshall andFouch-er, The Monuments ofSäňchl, vol. 2, pis. 26 (scene showing worship of a wheel installed on an altar); 27 (scene at top that shows worship of a pillar topped by a wheel and addorsed lion capital); and 32 (lower scene, which shows worship of a stüpa), to cite just a few.
92.   It might represent a site in the Bactro-Gandhära region itself. Chinese pilgrim accounts reveal that there were numerous places throughout the Bactro-Gandhära region where visitors could revere sacred objects that were reputed to have been used by the Buddha and places he was said to have visited. See, for example, Xuanzang's account of the Gandhära region (Beal, Buddhist Records of the Western World, 1:99). Some of the sacred sites are reputed to be places where the Buddha had lived during his past lives. (See, for example, Beal, Buddhist Records of the Western World, 1:110.)
93.   She further states that there are two hundred or more dome slabs about eleven by three feet, the compositions of which were divided into three registers. During the final phase of embellishment of the stüpa, she claims, their design followed a standard repetitive scheme. In an endnote (n. 42) following her mention of the final phase of embellishment of the stüpa, she states that the earlier slabs are decorated with a variety of subjects but that the later slabs all conform to the scheme she discusses and illustrates with her fig. 24. However, it is not explained how many of the two hundred or more slabs belong to the earlier versus later phases (that is, how many slabs actually have the fig. 24 design) nor her criteria for determining which examples belong to which phase. From what I have seen of the AmarävatI remains in the three main collections housing these materials (at the Madras Museum, the AmarävatI Site Museum, and the British Museum), I am doubtful that there are almost two hundred nearly identical eleven-foot-high slabs, and, therefore, the impactof the design she cites is not as great as she implies.
94.   Here it becomes clear that in this corpus of two hundred slabs, Dehejia is not really talking about one type but about a number of variations. The fact that Dehejia believes that the three so-called "Descent" scenes (the Swat example, the Bhärhut relief, and the one on the Mathurä carving; figs. 5,
152
SUSAN L. HUNTINGTON
6, and 7) portray the same subject suggests that she may not be taking into account variations among the compositions.
95. Dehejia also claims that the precise interpretation of these slabs "is crucial to an appreciation of the Buddhist message of the site" (64). (Does she mean the message of the monument?) How can the message of the site (or monument) be determined without studying other aspects, such as the main sculptures at the four entrances of the monument, the sculptures at the gateways, identifying the relic contents of the stüpa, studying other structures at the site, and so on? In other words, are these images alone enough to make such avast interpretation?
96. My assumption is that Dehejia is not attempting to make a distinction here but has simply used different wording. I infer that her intention is to suggest that the prerequisites are for the interpretation of early Buddhist art, not the emblems, and that she intended to say that the second prerequisite for the interpretation of early Buddhist art is to acknowledge the multilayered significance of the emblems^
97. Dehejia states, "There are two critical and complementary prerequisites for the accurate interpretation of early Buddhist art" (45, para. 2). The implication is that these prerequisites are applicable to all of the art However, Dehejia's use of the term frequently in the description of her second prerequisite ("the artists working at the early Buddhist sites frequently seem to have intended a conflation of meanings," 45) suggests that she does not believe that the two prerequisites are requisite in every instance.
p                          98. Dehejia's discussion is sometimes unclear as to
whether she is talking about two or three levels of it                               meaning. In her explanation of her second pre-
, j                               requisite, each of the possibilities she cites has only
two levels of meaning (46). She introduces the Bhärhut relief showing King Prasenajit (fig. 19; her fig. 1) by saying: "Significantly, a double layer of meaning appears to inform the greater number of narrative reliefs at Bhärhut, Sänchi, and other early Buddhist sites" (46). But her explanation of this panel provides three levels of meaning (47).
99. Lüders, Bharhut Inscriptions, 113, no. B39. Dehejia does not provide a transliteration of the inscription. Lüders translates the inscription "King Pasena-ji, the Kosala," rather than "of Kosala," as given by Dehejia.
lOO.Lüders, Bharhut Inscriptions,  113, no.  B38.
Dehejia's transliteration dhammachakodiifers from Lüders' dhamachakam.
101. To be accurate, there is no such thing as completely internal evidence; that is, we read into the composition things we know from the world outside the composition, for example, that the structure is a building, that the trees are trees, that the wheel is the type of Buddhist wheel we know from other experiences, and so on. What I mean by internal evidence, though, is the components that are actually in the composition and the physical properties of the object.
102. It may be inferred from literary and historical sources that the king might ride in a chariot, and, therefore, it is likely that the central figure in the chariot represents Prasenajit. Additionally, the two large figures in the hall are likely both to represent the king, who is thus shown twice as he performs his circumambulatory devotions to the wheel.
103. In light of the virtual absence of textual, inscrip-tional, literary, or other evidence in Dehejia's article, one of the most perplexing interpretations she offers is her identification of three male figures at the bottom right of the composition she illustrates as her fig. 8 as representing "three gods hovering anxiously over the weakened Buddha" (53). Because the figures are not clearly identifiable as gods, because they do not look anxious, because they do not appear to be hovering, and because there is no weakened Buddha in the scene, it is necessary that external evidence be employed to support such an identification.
104. She says that the scene shows the actual historical event in which the monarch visited the Buddha at Prasenajit's capital of Šrävasťi and listened to his sermon.
105. It is unclear to me why the idea of a nonfigural emblem is linked with emphasis on the Law. There is no inherent reason why the emphasis on the Buddha's teachings cannot be communicated through figural means as well. This may be a holdover from thinking within the aniconic framework.
106. Alexander Cunningham, The Stüpa of Bharhut: A Buddhist Monument Ornamented withNumerous Sculptures Illustrated of Buddhist Legend and History in the Third Century b.c. (London, 1879; rpt, Varanasi: Indological Book House, 1962), 90-91. Cunningham does not cite the textual source he used to confirm that King Prasenajit did indeed build a shrine'containing a wheel at Šrävasti. From other
ANICONISM AND THE MULTIVALENCE OF EMBLEMS: ANOTHER LOOK
153
information he provides, the reader may infer that he based his information on Xuanzang's seventh-century account. See Cunningham, The Stúpa of Bharhut, 90.
107. Although an early believer in aniconism, Cunningham was not the first to articulate the theory fully.
108.The Buddha did not meditate for six years under the bodhi tree. Cunningham may be confusing the period under the bodhiiree with the period of asceticism following the great renunciation.
109. Cunningham, The Stüpa of Bharhut, 90.
110. See, for example, Burlingame, Buddhist Legends, 363-64, for index listings; I. B. Horner, The Collection of the Middle Length Sayings (Majjhima-Nikäya), 3 vols. (1954—59; rpt., London: Luzac and Co., for the Pali Text Society, 1967-70), esp. vol. 2, passim; Carolyn Rhys Davids, TheBooh of the Kindred Sayings (Sanyutta-Nikäya) or Grouped Suttas, pt. 1 (rpt. London: Luzac & Co., for the Pali Text Society, 1950), 93-127; and W. Woodville Rockhill, The Life of the Buddha and the Early History of His OrderDcrived from Tibetan Works in the Bkah-hgyur and Bstan-hgyur (London: Kegan Paul, Trench, Trübner & Co., 1907), 111-1-6.
111. C. Rhys Davids,5ooA of Kindred Sayings, pt. 1,93-96. Note that the text Rockhill used says that the Jetavana was at Räjgir. See Rockhill, Life of the Buddha, 49.
112.C. Rhys Davids, Book of Kindred Sayings, pt. 1, 93-127.
113. The Jetavana was the principal location for numerous conversations. See C. Rhys Davids, Book of Kindred Sayings, pt. 1, 96 n. 1. However, other venues are also named, such as a market town called Medajumpa. See Horner, Collection of the Middle Length Sayings, 2:302.
114.At the time of the Buddha (and of Prasenajit), the Kosala kingdom included Varanasi (and, therefore, Särnäth). See Nundo Lai Dey, The Geograph-icalDictionary of Ancient and Mediaeval India (1927; 3rd ed., Delhi: Oriental Books Reprint Corporation, 1971), 103; D. C. Sircar, Studies in the Geography of Ancient and Medieval India (2nd ed., rev. and enl., Delhi: Motilal Banarsidass, 1971), 254-55.
115. It is irrelevant to my position whether the relief shows Särnäth or Šrävastl, since my point would remain the same: it is a scene showing the devotions of the king (at a sacred site) and not a life
event of Šäkyamuni Buddha.
116.  Beal, Buddhist Records of the Western World, 2:2. Also see Burlingame, Buddhist Legends, 40, where King Prasenajit offers to build a pavilion for the Buddha, but the Buddha refuses the offer.
117. It might even be argued that, because of the importance of Särnäth, where the Buddha's first sermon was performed, it would have been the first teaching site to have had a monument erected to commemorate a teaching event.
118. Cunningham, The Stúpa of Bharhut, 90.
119.1 develop andjustify this theme in my forthcoming book.
120. The purpose of including the story of Prasenajit in the Middle Length Sayings is to show his good works and spiritual evolution. See Homer, Collection of the Middle Length Sayings, 2:xvi n. 1.
121. Rockhill, Life of the Buddha, 203, says that the first Tibetan monarch claims descent from Prasenajit, King of Kosala, "one of the early converts and the lifelong friend of the Buddha Gautama."
122. It is unfortunate that a text comparable to the A'sokävadána is not known for King Prasenajit.
123. Illustrated in S. Huntington, Art of Ancient India, fig. 5.16.
124.Dehejia refers to S. Huntington, Art of Ancient India, 99, fig. 6.10, upper relief; and J. Huntington, "Sowing the Seeds of the Lotus," pt. 4, 30. The term pageantry theory has been coined by Dehejia. Neither John nor I have presented the idea in this way.
125. Since passion plays are specifically connected with the passion (that is, the suffering of Christ between the time of the Last Supper and his death) and the crucifixion of Christ, not the life of Christ in general, the analogy that Dehejia infers is particularly inappropriate. A Buddhist equivalent, if there is one, would have to be concerned with the death and events associated with the death of the Buddha.
126.S. Huntington, Art of Ancient India, 99. Regarding John Huntington's reference that Dehejia mentions, he says that a scene at Säňcí may be a "re-enactment of the event at the site before a group of pilgrims. "J. Huntington, "Sowing the Seeds of the Lotus," pt. 4, 30. Like me, he does not refer to anything as formalized as a passion
154
SUSAN L. HUNTINGTON
play tradition.
127. Webster's Ninth New Collegiate Dictionary (Springfield, Mass.: Merriam-Webster, 1988), 846.
128. For a typical account of the Great Departure, see Jones, Mahävastu, 2:155-61; also see Patricia Eichenbaum Karetzky, The Life of the Buddha, Ancient Scriptural and Pictorial Traditions (Lanham, New York, London: University Press of America, 1992), 70-77.
129.Jones, Mahävastu, 2:156, records that the Bodhi-sattva's groom, Chandaka, "cried out at the top of his voice so that the king and all the people in Kapilavastu might be awakened. But no one woke up." This passage emphasizes the soundness with which the people of the city slept.
130. Karetzky, Life of the Buddha, 70.
131. Bactro-Gandhäran represen ta tions generally show devas uplifting the horse, although often two, rather than four, figures are depicted. For examples, see S. Huntington, Art of Ancient India, fig. 8.19; Islay Lyons and Harald Ingholt, Gandhäran Art in Pakistan (New York: Pantheon Books, 1957), fig. 45.
132. If this scene indeed represents the Great Departure, this figure might be the Bodhisattva's groom, Chandaka. However, it would be peculiar for a groom to be carrying a ritual water vessel. If this is a reenactment, this figure might represent Chandaka but, assuming that he has a ceremonial role, might appropriately show him carrying a ritual object.
133.Foucher identifies this as a jambu tree. See Marshall and Foucher, Monuments ofSaňchx, vol. 2, text opposite pi. 40.
134.Edward J. Thomas, The Life of the Buddha as Legend and History (London: Roudedge and Kegan Paul, 1927; rpt., 3rd ed., 1969), 55. The Lalitavislara records a turning-around place of the Bodhisattva's groom, Chandaka. See Bays, The Voice of the Buddha, 1:339.
135. Thomas, Life of the Buddha, 55. Thomas also claims that according to one version of the story, Kanthaka did not turn around (p. 56), but the earth did (p. 55). Pali texts record this story as relating to the Buddha Vipassin. (Thomas, Life of the Buddha, 55.) The lives of each of the mänusi Buddhas are held to be identical in Buddhist theory, though not every text recounts each of the life stories in detail for each Buddha.
136. My suggestion that this may be a reenactment occurrring at the very place of the departure, Kapilavastu, implies that the celebration might be related to activities at important Buddhist sites. Dehejia's perception of the "site theory" and the "pageantry theory" as distinct, unrelated subjects does not take the full implications of my proposal into account.
137. S. Huntington, "Early Buddhist Art," 405.
138.1 suggest that Dehejia's use of the word whole overstates the case. Most devotees probably had more than one purpose in going to a stúpa— perhaps the principal and "authentic" one was to be in the presence of Buddhist relics, butthere are many others, including social and political, that must be taken into account when understanding the phenomenon and its popularity throughout the centuries.
139. S. Huntington, "Early Buddhist Art," 405.
140. Indeed, while it might be a bias of today's art-collecting, art-appreciating world that a golden image of a Buddha might have more "value" than a portion of his ashes, to practicing Buddhists throughout the Buddhist world and throughout Buddhist history, the relic would be the greater treasure.
141. This phenomenon is discussed in my forthcoming book.
142.1 specifically add "architecture" here. It is sometimes forgotten in the analysis of individual compositions, such as those illustrated by Dehejia in her article or by me in "Early Buddhist Art" and here, that these reliefs adorned structures that were built for purposes (often, to house relics) and that these purposes must be taken into account when analyzing the art.
143. Dehejia here again addresses my work. See S. Huntington, Art of Ancient India, 70.
144.1 believe she means early Buddhist art of India.
145. In this case Dehejia cites my work, referring to a statement I made in "Early Buddhist Art," 405, which she also quotes in her n. 6. What I say there is: "Essentially, I suggest that the early Buddhist art of India was not primarily concerned with the biography of Šäkyamuni Buddha, as has been assumed for so many decades."
146. Also see discussion above.
h*-
ANICONISM AND THE MULTIVALENCE OF EMBLEMS: ANOTHER LOOK
155
147. It is unclear here whether she is using Bhärhut as an example or whether she is implying that this situation is applicable only at Bhärhut.
148. The source of the "time and place" idea is S. Huntington, "Early Buddhist Art," 402-3.
149. S. Huntington, "Early Buddhist Art," 402. This statement is then further clarified by the discussion that follows.
150. The pluralization of the word proponents puzzling since I do not know of anyone beside myself who has discussed the "time and place" concept.
151. Dehejia's statement is unclear. While she (incorrectly) infers that I have said that both time and place are absent from "aniconic" scenes, she assumes that I would interpret such "aniconic" renderings as "site scenes," thereby, I suggest, implying that "place" is not absent. (After all, a site is a place.)
152. Here she shows that she does not apparendy believe that events in the life of Ašoka are Buddha life events. However, the significance of a representation of an event in the life of Ašoka is not made clear here since the idea was never discussed in her text, and it is riot specified how this subject relates to the notions of multivalency or of aniconism.
153. Early proponents of the aniconic theory related the practice to Hlnayäna Buddhism, assuming thatHlnayänists had a doctrinal prohibition against creating Buddha images. Concomitantly, it was believed that Mahäyäna Buddhists were responsible for the introduction of the Buddha image into the art. It is now known that Hínayänists used images as well and that Hlnayäna literature does not contain evidence of widespread (if any) prohibition against Buddha imagery. Therefore, one of the cornerstones of the aniconic theory—the presupposition of a widespread prohibition against creating images—is insupportable. I discussed the fallaciousness of the long-held Hinayäna prohibition model in my "Early Buddhist Art," 401-2. Dehejia agrees that the old Hlnayäna-Mahäyäna model must be abandoned (47).
154.1 use the term in quotes since, if the phenomenon exists, I nonetheless propose that the name it has been given is inappropriate. First, I do not believe that a phenomenon should be defined according to what it is not ("not an icon"); second, since the term is associated with the concept of deification, itis inappropriate for the Buddha. See n. 28 above.
155.Dehejia's presupposition that I see only two
alternatives to what have been traditionally identified as aniconic scenes, namely a "site" interpretation and the "pageantry" interpretation, is simply not correct. My forthcoming book, and hopefully this article to some extent, will make it clear that I see a number of alternative interpretations. That I have allowed for—and am thoroughly concerned with—other types of subjects is clear in the Art Journal article, where I say: "This article presents some of my findings in a preliminary fashion by focusing on one type of representation. Specifically, I will examine a type of relief that is among those that are usually said to illustrate scenes from the life of the Buddha, with the Buddha, however, not depicted. It is possible that most, if not all, of these compositions do not represent events in the life of the Buddha at all, but rather portray worship and adoration at sacred Buddhist sites" (S. Huntington, "Early Buddhist Art," 402; italics added here). When I present what Dehejia calls "the pageantry idea" in Art of Ancient India, I do not state that I conceive of pageantry scenes as one of only two alternative explanations for the so-called aniconic scenes. Further, although Dehejia recognizes what she calls the "pageantry" theory in her article, the introduction to her article implies a polarity between what she calls the "totally aniconic interpretation of the early 1900s [and] the somewhat restrictive site-oriented interpretation of this last decade" (45). Herstatementimpliesthatthe "site" interpretation was offered as a sole alternative to aniconism.
156. When Ašoka or other early kings erected pillars, surmounted by lions, cakras, and other motifs, these probably represent emblematic configurations. However, when these pillars are later depicted in artistic renderings at Säňci, Bhärhut, and other sites, often with worshipers adoring them and circumambulating them, they serve a narrative, rather than emblematic, purpose.
157. The role of my research in the formulation of Dehejia's ideas is not clearly defined in her text and notes. I am aware that Dehejia is publishing a book in which much of what is published in her article serves as a chapter. I hope that acknowledgment of Dehejia's derivations from my work are therein sufficiently cited. In addition to the site theory, which Dehejia incorporates as a major aspectof herwork, her borrowing of the term slesa, her reference to "time and place," and other themes from my work should be cited. Aside from the issue of crediting scholar(s) for their contributions, itis essential to future generations that the derivation of scholarly ideas be clearly presented and acknowledged so that the evolution of scholarly
156
SUSAN L. HUNTINGTON
thinking may be traced.
158. In his essay entitled "The Beginnings of Buddhist Art," Foucher states: "It is no longer a secret to anyone that the regular sweep of this evolution [the Indie art of SäňcI and related sites] was brusquely interrupted by a veritable artistic cataclysm. The Hellenized sculptors of the northwest, strangers to the native tradition of Central India, satisfied to the full; and even outwent, the wishes of their Buddhist patrons by creating for their use the Indo-Greek type of Buddha. Immediately their colleagues of the low country, seduced by this wonderful innovation, greeted with no less enthusiasm than the laity the rupture of the magic charm which had weighed so heavily and so long upon the ancient Buddhist school." Alfred Foucher, The Beginnings of Buddhist Art," in his The Beginnings ofBuddhistArtand OtherEssays inlndianand Central-Asian Archaeology (Paris: Paul Geuthner; London:
Humphrey Milford, 1917), 24. See also Foucher, "The Greek Origin of the Image of the Buddha," in Beginnings of Buddhist Art and OtherEssays, 111-37. For brief discussion of the historiography of the theory of aniconism, see S. Huntington, "Early BuddhistArt,"401,406.
159.  I am grateful to Janice M. Glowski for introducing me to these terms as used in the fields of Comparative Religions and Comparative Studies.
160. Webster's Ninth New Collegiate Dictionary, 427.
161. Webster's Ninth Nexu Collegiate Dictionary, 407. I am grateful to Arnold M. Zwicky for his clarification of these terms in relation to linguistic origin and use and the information he provided me about the originator of these terms, Kenneth Pike of the University of Michigan.