19

Avoiding Eye Contact: The Negative Aspects of Sight in Early South Asian Buddhism

ROBERT DECAROLI

IN THE MID-1960S Richard Gombrich documented Sri Lankan ceremonies performed to consecrate newly enshrined images of the Buddha.¹ Such rituals culminated with rites intended to “open the eyes” of the Buddha. This entailed inviting a specialist to paint in the eyes of the statue, a cause for celebration in the local community. Gombrich’s monastic informants did not fully approve of these rituals because they understood the image as merely an inert reminder of the Buddha. Their attitude might be characterized as resigned tolerance, seeing the ritual as misguided but essentially harmless.² By contrast, the local artists and broader community took the consecration ceremony very seriously. Skilled specialists (sittaru) were called in to paint the eyes. They did so by using a mirror to avoid direct eye contact with the Buddha’s gaze. The power of the image’s vision was considered so potent that, even after taking these precautions, the craftsman was blindfolded and led from the temple so that he would first look upon a waterpot or other prepared vessel that could be symbolically sacrificed, thus dissipating the residual power.³ Such care and such potency certainly imply that the Buddha, or some measure of his identity, was resident in the completed statue. Gombrich characterized the dichotomy between the knowledge of the Buddha’s absence and the efficacious power of his image as a tension between cognitive and affective ways of thinking, both of which were embedded in the Sri Lankan Buddhist worldview of the 1960s.⁴

20

It might be tempting to characterize this dichotomy as one between the educated saṅgha and the parochial laity, or more broadly as traditional practices confronting newer modes of scientific rationalism informed by modernism and the colonial experience. Although such powerful cultural and political forces must have certainly played a role in the participants’ understanding of the event, tensions over the presence or absence of the Buddha are far older than the modern period. Whether ancient or modern, however, questions about vision—both the devotees’ and the Buddha’s—have remained central to the debate over his presence.

There has never been a single Buddhist way to understand the Buddha’s images. Multiple, often contradictory, views of what it means to see an image of the Buddha have been present in Buddhism since the inception of the image tradition. In fact, because of the way the depictions of the Buddha were intentionally avoided prior to the start of the first century CE, one must conclude that this topic was an issue even before the creation of the earliest Buddha images. Artists, donors, and members of the saṅgha justified avoiding his depiction in various ways, some of which I have discussed elsewhere.⁵

Does his depiction imply his presence and accessibility? Might images undercut the finality of nirvāṇa, and if not, then how can image use be an effective mode of religious practice? In these and other ways, the tension over the presence of the Buddha is, in part, connected to the importance of vision in Buddhism. This concern also manifests in textual sources as a desire to regulate how people understand their interactions with figural artwork.⁶ For better or worse, vision, more than any other sense, has the power to impact how we behave. A natural corollary of this is the recognition that some aspects of sight must be regulated.

My primary goal is to explain how the formalized ritual use of vision and eye contact (darśana) led to a change in the way statues of the Buddha were made after centuries of uniformity and consistency. This change was contingent on the increased use of sculpture as a focus for Brahmanical rituals that gained popularity in fourth-century South Asia and was eagerly embraced by the Gupta royalty. Before elaborating, however, an overview of how vision was regulated by the Buddhist community will be helpful.

21

NOT FOR THE EYES OF THE SAṄGHA

It is no exaggeration to state that almost every rule in the Buddhist codes of behavior (Vinaya texts) for the saṅgha is predicated on a monk, nun, layperson, or god witnessing improper behavior and reporting it to the Buddha. Nothing escapes scrutiny. All manner of monastic behavior—ranging from terrible acts of violence to the style of a monk’s undergarments and the length of his nostril hairs—are regulated.⁷ In many cases, blameworthy actions are forbidden because they are inherently contrary to the Buddha’s teachings. In other instances, however, as in the case of the underwear and nose hairs, rules appear to have been set in place exclusively to prevent public misconceptions or disapproval. In these cases, one might easily assert that concerns over public perception of the behavior prompted the rule rather than any inherently immoral quality of the act itself.

This is particularly apparent in rules related to monks undertaking menial chores and physical labor. Gregory Schopen has noted how the Mūlasarvāstivāda Vinaya requires monks to avoid being seen by the public when performing activities that might be associated with pollution or people of low social status.⁸ Manual labor, smithing, leather working, and cutting hair were all permissible, and at times necessary, activities. But in each case monks were allowed to perform them only in private, away from the gaze of the public. Such activities were to be done privately in order to preserve the community’s positive image and not generate negative public opinion. In other words, some activities that were permissible for the saṅgha were not suitable viewing for the public.

Just as there were things the public should not see monks or nuns doing, certain sights acceptable for the lay community were forbidden to the saṅgha. There is a rule, for example, that limits the subject matter used for adorning the monastery. This rule is found in the Mūlasarvāstivāda Vinaya as well as in the Chinese version of the Sarvāstivāda Vinaya.⁹ As is standard for Vinaya texts, the restriction is predicated by a story that explains and justifies it. In both versions, a monk is feeling wistful and begins to draw an image of the young bride he left in order to join the saṅgha. At this inopportune moment a senior monk sees the image and scolds the amateur 22artist—ultimately bringing the matter to the Buddha. This leads to the pronouncement of a rule forbidding monks from creating images of living beings. This passage seems primarily intended to counter the titillating aspects of figural art that may be contrary to celibate monastic life; however, it is also likely that it was intended to prevent public disapproval.

For example, the Sarvāstivāda Vinaya contains rules explicitly preventing the decoration of stūpas with images of “men and women coupling.”¹⁰ In the Cullavagga of the Pāli Vinaya, similar concerns are expressed. It contains a general restriction on any representations of men or women used as decoration in monastic dwellings. In both of these cases there is good reason to think that public perception played a role in the formalization of the rule. The Pāli Vinaya makes this apparent by framing the rule with a story about guests in the monastery who are shocked to see depictions of people on the walls of a monk’s residence and begin to question the legitimacy of the entire Buddhist monastic community.¹¹ Presented in this fashion, the Buddha’s restriction would seem to have as much to do with external perceptions as internal discipline. Either way, these rules identify visual imagery that was suitable for the laity but deemed inappropriate for members of the monastic community.

This is certainly not the only rule in the Vinaya requiring saṅgha members to control their gaze. As is well known, members of the monastic community were always expected to conduct themselves with proper decorum, and this was especially true when traveling outside the monastery. At these times they were required to direct their eyes downward.¹² This restriction is emphasized in the Mahāparinibbāna Sutta. When the disciple Ānanda asks the Buddha how monks should act toward the women they encounter while on pilgrimage, he instructs monks to simply not see them.¹³ The self-control required to regulate their vision was seen as conducive to both monastic practice and the positive public perception of the saṅgha. In short, appearances mattered. What members of the monastic community saw and what they allowed others to see them doing were of deep concern to the writers of the monastic codes. It is also clear that a great deal of this concern goes beyond the moral implications of actions and aims to forestall any criticism by witnesses. What an individual sees can impact their thinking. The visual, therefore, holds great potential for misunder23standings that can place both the individual and the Buddhist institution at risk.

SEEING THE BUDDHA

The Buddhist relationship with sight is a complex one. In many South Asian contexts, vision and sight were (and continue to be) synonymous with understanding.¹⁴ For example, in the Divine Stories (Divyāvadāna), a monk wishes to see the Buddha’s physical body (rūpakāya) to complement his understanding of the Buddha’s teachings, or Dharma body (dharmakāya). For the monk in the story, seeing both aspects of the Buddha’s nature ultimately benefits his spiritual advancement.¹⁵ Being in the presence of a living buddha is frequently described as one of the greatest possible karmic rewards, because seeing and hearing a buddha preach provides an unparalleled opportunity for insight and awakening. For those not fortunate enough to live in the time of a buddha, systems of internal visualization were created to replicate the experience.

The Pāli tradition introduced a meditative process known as buddhānussati (Skt. buddhānusmṛti). In this form of meditation one “recollects the Buddha”—that is, one envisions the image of the Buddha in one’s mind so as to benefit from being in the Buddha’s presence as he preaches. In some cases, this practice has been linked to the use of images, but typically physical images are not required.¹⁶ In fact, the meditative practice reveals some mistrust of physical forms and the attachments they may engender. Sessions of buddhānusmṛti conclude with the practitioner reversing the process of embodiment and systematically dismissing the Buddha’s form. This exercise may confirm the value in seeing the Buddha’s form but also provides a firm reminder of his absence.

The viewer’s mindset when looking at an image is often presented as the key to positive or negative outcomes. For example, Andy Rotman has written at length about a concept known as prasāda and its role in the avadāna literature. He describes it as a state in which external stimuli—typically seeing the Buddha’s image—inspire acts of generosity and kindness. This compassionate mindset, which Rotman describes as a kind of arousal, automatically places one in the proper mental state for performing meritorious 24acts.¹⁷ In other words, what one sees affects one’s attitudes and one’s actions. This potentially opens the way to great karmic rewards, or might just as easily inspire wicked deeds and the negative karma they engender. Vision is a two-edged sword and was therefore subject to great deal of (occasionally contradictory) regulation and commentary. Although prasāda might be understood as an essentially positive form of vision, I discuss it here because it is one of several rationales that allowed for the use of images without necessitating the Buddha’s presence or attention. In other words, the statue could be inert, the Buddha could remain absent in nirvāṇa, and the use of images could still bear results for the devout because the simple act of looking at it inspires (or actively produces) acts of good karma.

While vision could occasionally be conducive to attaining religious goals, it was more often understood as an impediment or potential source of attachments. The difference between a negative or positive outcome was largely contingent on the viewer’s frame of mind. Therefore, defining how viewers should understand an image of the Buddha was of particular concern to Buddhist authors, but they provide little consensus on the correct approach. The views of Buddhist writers were often at odds with one another. Some Buddhists strongly rejected the idea of the Buddha acting through an image, others conditionally accepted it, and others were not troubled by these implications at all.¹⁸ In many cases the critics saw the use of images as introducing potential risk, and a few even characterize it as dangerously misguided.¹⁹

The development of the Buddhist image tradition at the start of the first century CE introduced new categories of visual imagery and new problems for Buddhist authors. One of the most impactful challenges faced by the Buddhist community involved the common understanding of images as more than just passive objects. Images, including those of the Śākyamuni, were credited with a substantial degree of agency in early South Asian religions.²⁰ This is seen most clearly in the ways Buddhist legal codes grappled with how to manage the presence of images in monasteries. Images of the Buddha were treated as independent entities with several legal rights, including the ability to receive gifts and own property.²¹ This agency ascribed to images meant that there was not only great concern over how devotees understood Śākyamuni’s images and what images might imply 25about impermanence, but also concern over what, if anything, the statues themselves might see.

In the early years of the Common Era, the brahman community also struggled with this topic. Many brahmins were hostile to those who attempted to depict the Vedic gods in physical form, and these arguments often paralleled those found in Buddhist sources. The Pūrva Mīmāṃsā school, for example, was strongly averse to the use of any images in ritual practice. They rejected the possibility of divine embodiment and argued that all references to gods’ “bodies” were entirely metaphorical. A less extreme position was held by the Vedānta thinkers, however. This philosophical school, particularly the Advaita Vedānta branch, took a more moderate approach and allowed for the possibility that gods might embody themselves if they desired to do so.²²

Many early Brahmanical sources held exceptionally negative views of those who used images in worship. This is exemplified by the Manusmṛti, or Laws of Manu, a Brahmanical legal text often attributed to the second century CE, which marginalizes and demeans devalakas, temple priests who attend images. They, along with doctors and butchers, are to be excluded from funerary rituals and from all rites directed to the Vedic gods or ancestors. This disdain for people who would normally hold high status is explicitly connected to their central role in devotional acts involving figural representations of the gods.²³ However, these passages also reveal that enough brahmins were participating in image-based devotion to warrant commentary by concerned traditionalists.

By the fourth century the brahman priests’ prevailing attitude toward images began to change. The Gupta period saw the rise of royally supported, temple-based Hinduism and the increased prominence of both bhakti (devotional worship) and darśana (eye contact with the divine, often mediated through an image). This shift in religious practice posed challenges to both Vedic traditionalists and to Buddhists. It solidified ideas about images and created expectations of an immediate and personal encounter with the object of devotion mediated though the image. As is probably apparent, this approach was directly contrary to longstanding Buddhist views on images that insisted on the Buddha’s absence and the fundamental nature of impermanence.

26

A Buddhist response to this increasingly dominant Hindu mindset can be seen in buddha images created in the Gupta territories. This change starts in the fourth century and is first apparent in North India, centered on the Mathurā region. Specifically, artists introduced a new manner of representing the Buddha’s eyes that broke eye contact with the viewer and, in so doing, also broke with standard visual forms that had remained consistent over prior centuries.

THE BUDDHA’S EYES: ORIGINS AND ALTERATIONS

The earliest images of the Buddha were created primarily in two regions of South Asia, Mathurā (in north-central India) and Gandhāra (in what is now mostly Pakistan). Although the images of Śākyamuni from both regions share similar iconography, they are stylistically quite distinct. The most important of these differences for the present discussion is the way artists represented his eyes. Gandhāran buddhas typically exhibit partially-closed eyes with heavy lids.²⁴ By contrast, the standard Mathurā types had large, wide-open eyes, which were derived from the eyes depicted on statues of yakṣas and other terrestrial deities. The Mathurān artists drew on familiar, large-scale figural forms when sculpting the earliest buddhas, so the similarities to images of yakṣas are most likely a byproduct of their production (figs. 1 and 2). In comparing this yakṣā from Vidiśā (located near Mathurā) with a second century CE image of Śākyamuni that was produced in Mathurā and installed in a monastery at Sarnath, we can note the large, staring eyes, the frontal stance that rests equally on both feet, the broad shoulders, and the fist balled at the hip that grasps either a flask or the outer monastic robe.²⁵ These so-called kapardin-style figures from Mathurā (of which the Sarnath image is a late example) are arguably the earliest Śākyamuni images (ca. first century CE) and are invariably sculpted with prominent, open eyes and clearly defined upper and lower eyelids (fig. 3). I have intentionally referred to these figures as Śākyamuni rather than as buddha images because when they bear inscriptions, they are identified as bodhisattvas. Ju-Hyung Rhi has argued convincingly that they are intended to represent Śākyamuni before his enlightenment and thus before attaining buddhahood.²⁶

27

By contrast, the earliest inscribed Gandhāran Śākyamuni images are typically identified as buddhas despite being iconographically identical to the Mathurān bodhisattvas. This regional difference is made particularly apparent in one image that bears inscriptions identifying it as both a buddha and a bodhisattva, each in a different language. Both Bishwa Nath Mukherjee and Gregory Schopen have discussed this bi-scriptural inscription from the base of a Mathurān sculpture.²⁷ The Brāhmī portion (typical of Mathurā inscriptions) identifies the figure as a “bodhisattva,” whereas the Kharoṣṭhī portion (used in Gandhāran inscriptions) uses the term “buddha.” The implication is that two Buddhists might see the same image and understand it in different ways. There is substantial evidence of contact and exchange between the regions, however, and these differences in nomenclature appear to have eroded over time, with the term “buddha” becoming applied more universally.