This article, based on a lecture delivered to a symposium held in Leiden on the function and meaning of Buddhist art, appears in a book entitled Function and Meaning in Buddhist Art, edited by K.R. van Kooij and H. van der Veere. Gonda Indological Series, vol. 2 (Groningen: E. Forsten, 1995).

Tibetan art regularly functions as icon, intermediary between man and divinity. It subsists within a highly sophisticated liturgy designed to evoke a deity for purposes of worship and communication. But even iconic art is subject to the laws of social and political life; and every work of art functions and has meaning on many levels.

One facet of Tibetan iconic art is to be found in early portrait paintings. Portraiture figured prominently in Tibetan art between ca. 1000 and 1400 A.D., and yet almost nothing is known about its functions and its significance.¹ This essay addresses two main questions: What aesthetic and theoretical guides did artists observe in painting historical persons? And what social, political, and religious purposes did portraiture serve in pre-fifteenth century Tibet?

Plate 1

Some of Tibet's finest examples of early portraiture are still preserved at Alchi (al-chi), a monastic complex along the Indus River in Ladakh, founded around 1200 A.D.² Alchi's two earliest structures, the Dukhang ('du-khang, "Assembly Hall") and the Sumtsek Lhakhang (gsum-brtsegs lha-khang, "Three-tiered Temple") both contain numerous examples of portraiture. A narrative along the east wall of the Assembly Hall features three figures: a man seated and holding a staff, a woman who faces him and offers a stemmed-cup, and a young man (an observation deduced from his size), presumably their son.³ (pl. 1)

Above the three are ministers and warriors with their shields, as well as female attendants. Below, other attendants pour libations from a jug and see to the needs of the protagonists.

Note that the two main figures are haloed, a phenomenon common in Tibetan depiction of royal, noble, and saintly persons. In this instance, the halos are likely to reflect both a pre-Buddhist Tibetan notion of divine kingship and the imported Indian notion of cakravartin, the divinely inspired ruler.⁴ An inscription below the scene positively identifies these figures as royalty, stating that the king and queen, having given royal sanction for the temple's construction, were refreshing themselves in the region.⁵ Their names and their precise dates are unknown.

Plate 2

In Tibetan portraiture, one does not usually find ordinary mental states (i.e., pride, humility, fear, or anger) depicted. Thus, in the case of this king at Alchi, one has absolutely no sense of his character as a man, although his apparel, his halo, and the composition within which he appears tell us of his existence and of his importance in the social hierarchy of his time. (pl. 2)

Plate 3

In some instances, one cannot know whether images are intended to portray historical individuals or merely to represent social types. The three wonderfully costumed figures in plate 3 appear to the left of the great Tara scene in Alchi's Three-tiered Temple, bearing tribute. The figures' descending size probably suggests relative social status, the two largest figures adorned with flaming halos.⁶

Plate 4

One finds considerable consistency in artists' depictions of garments, hairstyles, and jewelry throughout the many historical scenes at Alchi and in contemporary cloth painting and manuscript illumination. In plate 4, one is able to discern in some detail the extraordinary garment worn by the senior member of this tributary, its design consisting of squares containing elephants and horses with riders. Analyses of artworks and surviving textiles of the period indicate that similar garments were worn throughout Asia between the seventh and the twelfth centuries.⁷

Plate 5

Other scenes of royalty appear in Alchi's Three-tiered Temple. Plate 5 features a king and two queens, found on the temple's ground floor.⁸ Here, the king and his two queens are haloed, as are their offspring (represented as miniature persons) and various associates. Both queens appear within abbreviated architectural settings. The king holds a written document close to his chest, one page of which he extends to the queen on his right, while the queen on his left offers something to her husband.

Plate 6

Attendants carry golden, jewel encrusted caskets towards each queen. The scene probably illustrates royal sanction of this temple's founding. One of these queens appears again in a proximate scene, her dress, coiffure, and physiognomy virtually identical. (pl. 6) Here, she is the central figure in the narrative, flanked by smaller and thus less socially important figures--that on her right probably her son and on her left, a religious figure. The monk is likely to be Tshultrim O (tshul-trim-'od, act. late twelfth century), a member of the powerful Dro ('bro) clan mentioned in a nearby inscription as the temple's founder.⁹

The queen's similar portrayal in both scenes--while not surprising, since the narratives are contemporaneous and produced by the same group of artists--nevertheless raises the question of the relationship between the artist's rendition of this queen and her actual likeness.

Plate 7

While one can never ascertain the portrait's verisimilitude, it is likely that the artist--who may never have seen her--conceived of the queen as a physiognomic type; and being a queen, she was obliged to possess a kind of unearthly beauty. It is interesting to observe the close resemblence that our primary queen shares with Green Tara, featured in another scene on the ground floor of the Three-tiered Temple.¹⁰ (pls. 7, 8)

Plate 8

Despite the obvious difference in their complexions, the queen and the goddess share fundamental elements of beauty. The shape of their faces, eyes, noses; their rosebud lips, and their extended, pointed chins are all extremely similar, as are their earrings and necklaces. The coiffures of both figures are adorned with a cloth strip which affixes to the crown of the head and falls down the back. Their close resemblance calls to mind a story from the Sanskrit text Vikramcarita, recounted by Ananda Coomaraswamy.¹¹ An Indian artist, called upon to make a portrait of a queen, recognized in a glance that her beauty was of a type--padmini ("likened to a lotus flower")--and painted her portrait accordingly, primarily guided not by what he observed, but by an ideal.

The theoretical guidelines for early Tibetan portraiture are still obscure. Indigenous literary accounts, which would greatly enhance our understanding of this genre, have not yet come to light--if, indeed, any were written. In their absence, it is instructive to briefly consider portraiture in India and China, whose arts contributed greatly to the early formation of Tibetan art. India is not generally thought to have had a highly developed tradition of Buddhist portraiture. Portraiture is, however, mentioned in Sanskrit texts from the Gupta period (A.D. 320-486) onwards, and there is evidence that some form of religious portraiture was practiced at medieval Buddhist monasteries such as Vikramasila, whose ruins lie near the modern city of Antichak, Bhaglapur district, Bihar.¹²

Tibetan translations of Indian treatises, such as the Citralaksana and the Manjusrimulakalpa, as well as Sanskrit texts including the Visnudharmottara and the Vikramacarita briefly describe portraiture in India. The Visnudharmottara, a work of the seventh century or earlier, describes resemblance (sadrsya) with the subject as one of the five important features of painting.¹³ But, as Coomaraswamy stresses in his commentary, "resemblance" in this instance should not be understood as accurate physical likeness; rather, a reminiscence of the subject's qualities beyond mere physical appearance.¹⁴

The Visnudharmottara discusses portraiture within the context of five types of men: hamsa, bhadra, malavya, ruchaka, and sasaka.¹⁵ Each type possesses a unique physiognomy, its corresponding mentality, and--when rendered in art--specific iconometric proportions. The text describes the five in colorful, if imprecise terms. Of the Sasa it is said:

...[he] will have somewhat projecting, otherwise fine teeth, fine nails...a swift pace; he takes delight in science, mining and trade; has full cheeks, is...a good general; fond of love's sport and partial to other men's wives; restless, valorous, obedient to his mother, and attached to woods, hills, rivers and wildernesses."¹⁶

The implication is that the subject of a portrait would have been cast as one of these five types and painted accordingly, concessions also having been made to the individual's caste.¹⁷

There were, however, instances when the portrait artist's powers of observation were called upon. The Manjusrimulakalpa states that the officiant, who appears in the painting's lower corner, should be drawn after nature.¹⁸

Of particular interest are the pronouncements on portraiture which appear in the Citralaksana, a text no longer surviving in Sanskrit but extant in its Tibetan translation.¹⁹ The text describes the portrayal of various categories of gods and men. It is clear, however, that in portraying the various physiognomic types, an artist was expected to "improve" upon nature. "How should one proceed in the matters of proportions, if a body is not possessed of perfect beauty? It is for this reason that the definition of the proportions...[has] been imparted."²⁰ Elsewhere, the text states that these iconometric rules, based upon "worship of all the Gods", exist "because emphasis must be laid on what appears agreeable to our eyes..."²¹

Among the human types to be represented are sages, siddhas, and the universal monarch (cakravartin). The universal monarch's appearance is described in detail:

"The face should be made squarish in form, sharply delineated, beautifully full and endowed with brilliant and pleasing marks. It shall not be made triangular, nor sloping; it shall not be made angry nor round...the cheeks should be made to the measurement of 5 digits, and the jaw bones to that of 4 digits. One should know that the measurements of the contour of the cheeks amount to 4 digits..."²²

Despite these prescriptions, some leeway was permitted the artist. Regarding the sage (pandita), it is said, "...everyone should have his measurements according to his own digits."²³ Later, regarding proportions of "normal" human beings, the artist is cautioned: "...in the case of those born of woman, one should exercise one's own judgement..."²⁴ Thus, as Coomaraswamy has already noted, Indian portraiture observed two apparently contradictory approaches: on the one hand, informed by observation of the subject and, on the other, purely imaginative, following the prescriptions for ideal types.²⁵

Portraiture has a long history in China. A tradition of hierarch portraits existed from the Tang period (618-907) onwards, often exhibiting a high degree of realism and appearing in a functional context quite similar to that in Tibet.²⁶ Chinese artists placed great emphasis upon close observation of an individual, for accurate depiction of the subject's physiognomy as well as his mind or "spirit" were both considered fundamental to proper portraiture.

Along these lines, Chen Shidao (1053-1101) wrote: "A loss of formal likeness and spiritual harmony alike would produce a painting like a shadow silhouette, which would not be a portrait."²⁷ Consider a passage from Wang yi (active ca. 1360):

Whoever paints a portrait must be thoroughly familiar with the rules of physiognomy, for the disposition of the parts of people's faces is like that of the Five Mountains or Four Rivers, each element being different....I begin with the left side of the nose, and then go to its right side. Then follows the tip of the nose....If the bridge of the nose is high, I start from its point of concentration and go downward in one stroke; if it is low, I start from the side of the eye and go downward in one stroke. If it is neither high nor low, then I start from a point eight-tenths or nine-tenths in distance [from the tip] and go downward in one stroke at each side. Then follow the part between nose and mouth, the mouth, the sides of the eyes...It is necessary to proceed like this from one part to another so that not even a hair will be omitted or incorrect.²⁸

The artist's painstaking effort to capture minute details of the subject's particular physiognomy contrasts sharply with Indian portraiture and with aspects of early Tibetan portraiture, as I will argue below. To what extent the traditions of Indian and Chinese portraiture influenced the formation of early Tibetan portraiture remains to be seen. Future studies will put these questions into perspective, once the salient characteristics of early Tibetan portraiture are clear.²⁹

Plate 9

A special genre within Tibetan portraiture is that of the Buddhist hierarch. In Alchi's largest chorten (chos-rten; skt., stupa), south of the monastic compound, one finds wall paintings of four impressive figures.³⁰ Plate 9 features the image on the chorten's western wall: a white robed man with brown skin and white hair. He is surrounded by religious implements--books, a bell, staff, incense, and a waterpot. Without an inscription, his identity is uncertain, but his brown skin probably indicates that he was an Indian, for Tibetans customarily depicted Indians in various shades of brown complexion.³¹ Clearly, the man was a religious figure, probably one of the Indian dharma masters involved in the introduction of Buddhism to Tibet.³² The figure on the chorten's southern wall (pl. 10)

Plate 10

compares closely with a portrait in Alchi's Lotsawa Lhakhang (lo-tsa-ba lha-khang; "Translator's Temple") which is accompanied by an inscription.³³ The inscription mentions Rinchen Sangpo, suggesting the image itself portrays Rinchen Sangpo (rin-chen bzang-po, 958-1055), a pivotal figure in the revival of Buddhism in western Tibet during the late tenth and early eleventh centuries. As emissary of the local ruler Yeshe O, Rinchen Sangpo journeyed several times to India were he studied with many of the great Indian masters of his day and amassed Buddhist texts which he later translated.

All three portraits of Rinchen Sangpo at Alchi share the same physiognomic features: ovoid face, receeding hairline, bulging eyes, pointed chin. All display the teaching gesture and wear monk's robes. Since Rinchen Sangpo died in 1055, his portraits at Alchi were undertaken at least one hundred and fifty years after his death. What guided Tibetan artists when the subject of their portrait was not available?

Plate 11

Artists sketchbooks, such as that seen in plate 11, undoubtedly provided some assistance.³⁴ The manuscript once belonged to the Nepali artist Jivarama, and bears a date of 1435. Inscriptional evidence indicates that Jivarama's sketchs may have been done during a journey to Central Tibet. One can only postulate the sources for his idiosyncratic drawings of various Tibetan figures, each identified by Newari or Tibetan inscription: earlier portraits, oral history?

Plate 12

Among Jivarama's sketches is a depiction of the prominent twelfth century Tibetan Buddhist, Phagmodrupa (phag-mo-gru-pa, 1110-1170), shown in plate 11 as the middle figure in the top row, wearing a hat. Phagmodrupa also appears in the Three-tiered temple at Alchi, where he is again identified by inscription.³⁵ (pl. 12) It is interesting to compare the two portrayals for there is little resemblance between them; probably neither portrait actually resembled the man.

This comparison and other similar examples lead one to conclude that accurate physiognomic likeness was not crucial to Tibetan portraiture of this period. If strict rules of verisimilitude did not guide portrait artists, then what aesthetic and theoretical guides did artists observe in making hierarch portraits?

Plate 13

In some cases, portraits of religious hierarchs are hardly distinguishable from divine images. Plate 13 shows the frontispiece from a woodblock edition of the Chinese Qi sha Tripitaka, dated 1301.³⁶ On the left is a seated Tibetan monk attended by two standing monks, on the right is Sakyamuni Buddha, flanked by an Indian monk and an arhat.

It is interesting to observe the many iconographic parallels between these representations of the most sacred founder of the faith and, one might say, a mere monk. The Tibetan monk and Sakyamuni offer the same teaching gesture, and they are similarly dressed, although the Tibetan monk wears the peculiarly Tibetan sleeveless undershirt beneath his robe, while the Buddha is bare-chested, wearing only the traditional Indian robe.

Both figures are seated upon identical thrones, the sides of which are constructed from an elephant, vyalaka (leogriff), makara (mythological crocodile), and kirttimukha ("face of glory"), familiar to anyone acquainted with South Asian art. This throne originated in India where, as Madame Auboyer has shown, it signifies the assemblage of natural and supernatural forces and their obeisance to him (or her) who sits upon it.³⁷ In Indian art, it is reserved for the rare universal monarch (cakravartin) and for Buddhas, bodhisattvas and other deities. In short, this is not merely a seat, but a setting of great symbolic significance.

The figures differ chiefly in that the Buddha, following traditional iconography for his representation, exhibits peculiar physiognomic characterisitcs known in India as laksana or the physical signs of his full enlightenment: elongated earlobes, the urna (mark between the brows), and the usnisa (cranial protruberance).³⁸

Plate 14

There is considerable evidence that Tibetan artists borrowed iconographic conventions originally developed for depictions of Buddhas and bodhisattvas in their renditions of Buddhist hierarchs. This can be seen in the mid-thirteenth century portrait of Shangton Chogyi Lama (zhang-ston chos-kyi bla-ma, d. 1244), now in a private collection. (pl. 14) This painting depicts the fifth abbot of Narthang (snar-thang) monastery, south central Tibet.³⁹

Plate 15

The artist has observed iconographic rules for depicting an enlightened being: the temple-throne setting, laksana (here, wheels on the soles of his feet), and the gesture of religious instruction (dharmacakrapravartana mudra). And yet, one is struck by the sensitive portrayal of his features. (pl. 15) The artist has taken care to show lines around his mouth and between his brows, the stubble of his chin, a faint mustache, and a receding hairline--all suggesting a unique individual.

Plate 16

The painting's apparent realism is put into perspective when one considers a remarkable kesi now in the Potala, Lhasa. (pl. 16) This portrait, described as that of Shang Lama (1123-1194), bears strong resemblance to the painted portrait of Shangton Chogyi Lama.⁴⁰ They are extremely similar, not only in basic physiognomy (e.g., body type, eyes, nose, and mouth), but also in details such as hairline and facial lines. The Shang Lama (pl. 16) wears a small goatee, but little else distinguishes them. One cannot know how closely these portraits resembled their subjects. But the close resemblance which these portraits share suggests the artists relied on physiognomic types, perhaps adapted to reflect physiognomic peculiarities of the subject, but sometimes possibly of a purely imaginative nature.

Plate 17

A thirteenth century painting portraying two unidentified religious figures, now in the Cleveland Museum, underscores this point. (pl. 17) The figures are identically presented, their thrones differ only in that the waters issuing from the makaras on one throne lift upwards; on the other, downwards. The two were probably contemporaries and possibly leaders of rival groups. The monks are distinctive in their physiognomies: that on the left possesses a broader face, more bulbous nose, larger teeth, and a different pattern of facial hair. (pls. 18, 19) But the regularity of their features, the near-perfect pattern of their beards and mustaches suggest that their portrayal, even if informed by the subjects' actual physiognomy, is nevertheless idealized.

Plate 18	Plate 19

Plate 20

A somewhat later double portrait also presents idiosyncratic renditions of the subjects, in this case, very possibly based on the subjects' actual physiognomic traits.⁴¹ (pl. 20) This silk painting, now in a private collection, features two Sakya hierarchs, probably Sonam Tsemo (bsod-nams-rtse-mo, 1142-1182) and his younger brother Drakpa Gyaltsen (grangs-pa rgyal-mtshan, 1147-1217).⁴² Sons of Kunga Nyingpo (kun-dga snying-po, 1092-1158), third abbot of Sakya monastery, they were distinguished theologians and revered teachers of their day.

Plate 21

The painting is remarkable for its idiosyncratic treatment of the subjects. Their emphatically long eyes, distinctive features, and detailed, lavish robes show a degree of realism unusual in early Tibetan portraiture. (pl. 21) But is it realism or, like most portraits considered thus far, merely realistic technique? Drakpa Gyaltsen's curly red hair bears particular mention.⁴³ To my knowledge, this physiognomic feature does not appear in later portraits, where he is usually shown with short white hair. However, one other ca. fifteenth century painting may also depict Drakpa Gyaltsen with red hair.⁴⁴ While extremely unusual, red hair is not genetically impossible for a Tibetan. Tibet's early medieval Central Asian neighbors are known to have had red hair, as do some Afghans today. It is unlikely that an artist would have fabricated this physiognomic feature, which must have been familiar to Sakya historians of the fifteenth century.

Plate 22

If Tibetan artists borrowed iconographic conventions developed for the depiction of Buddhas and bodhisattvas in their renditions of cherished hierarchs, they did so because they often saw their hierarchs as divinities. This is evident in a painting now in the Cleveland Museum of Art dating to the late twelfth or early thirteenth century and featuring Vajrasattva in his form as the sixth or Adi-Buddha.⁴⁵ (pl. 22) Most interesting for the present discussion is the diminuitive seated monk who appears in Vajrasattva's tiara. (pl. 23)

Plate 23

While it is not uncommon for an Esoteric Buddhist deity to bear an image of his spiritual superior in his crown, it is rare for a deity to bear a human image in this superior position. The Tibetan cultural historian R.A. Stein notes the particular reverence with which Tibetans imbued their religious masters, stating that "it is typical of [Tibetan Buddhism]...that one's Lama (guru) is superior to all `deities,' even the most prominent."⁴⁶ Moreover, "The Lama's dominant position, higher even than the buddhas, is common to all orders [of Tibetan Buddhism]."⁴⁷ Stein offers insight into the possible iconologic significance of this image:

...the disciple conjures up the tutelary deity (yi-dam) chosen for him by his Lama, with whom the deity is closely fused; then he himself merges with the Lama, who has absorbed, as it were, the deity. From him he draws the desired state of purification.⁴⁸

Go Lotsawa's ('gos lo-tsa-ba, 1392-1481) early fifteenth century The Blue Annals (deb-ther sngon-po) frequently describes religious figures in terms of the deities whom they are said to embody. The venerable Ling (gling, 1128-1188), is attributed with the statement, "My body, speech, and mind do not differ from the Body, Speech, and Mind of all the Tathagatas...Whoever will pray to me with devotion, will realize all his wishes..."⁴⁹ One of Drigungpa's disciples "...did not abandon even for a single moment the notion that the Dharmasvamin [Drigungpa] was a Buddha."⁵⁰

Go Lotsawa also describes a disciple who chastised himself for thinking of his master as merely a "tenth bhumi" bodhisattva: "From then on, he did not differentiate his Teacher from the notion of Buddhahood..."⁵¹ Near death, Taglung Thangpa Chenpo (stag-lung thang-pa chen-po, 1142-1210) is said to have uttered: "I am the Sugata Himself."⁵² Of Taglung Thangpa Chenpo's successor Kuyal Rinchengon (sku-yal rin-chen-mgon, 1191-1236), it was said: "On many occasions...he exhibited different manifestations of his physical form. Further, many saw him as Sakyamuni, Samvara and other divine beings."⁵³

Who was the diminuitive monk pictured here? Analysis of the figures along the painting's top register provides some clues. They ilustrate a now well-documented early Kagyu lineage, seen in dozens of paintings from Central Tibet, and at Alchi.⁵⁴ Thus, from left to right are: Dorje Chang (rdo-rje 'chang; skt., Vajradhara), the Indian yogin Tilopa (act. late tenth, early eleventh century), his Indian disciple Naropa (956-1040), followed by his Tibetan disciple Marpa (mar-pa, 1012-1096). Next to Marpa is his Tibetan disciple Milarepa (mi-la-ras-pa, 1040-1123), followed by Gampopa (sgam-po-pa, 1079-1153), the latter's chief disciple. Gampopa's greatest disciple was Phagmodrupa (1110-1170, whose portrait has already been noted at Alchi), perhaps the diminuitive figure in Vajrasattva's crown.

Despite Go Lotsawa's typically hagiographic prose, Phagmodrupa stands out as a particularly remarkable teacher. Go describes him as the "second Buddha."⁵⁵ He elaborated by saying that Phagmodrupa was a Buddha "to those possessed of excellent understanding, a siddha-purusa to those possessed of medium understanding, and a fortunate human being (prthag-jana)...to those possessed of inferior understanding."⁵⁶ Phagmodrupa apparently shared this view when he "openly proclaimed that he was the Buddha of the Past and Future, as well as the Sakyendra of the Present Age."⁵⁷

The notion of a teacher's pivotal role in the spiritual development of his disciples was familiar to Phagmodrupa and his predecessors. Naropa, Indian progenitor of Phagmodrupa's spiritual authority, is attributed with the view that, "[a]s long as there is no Lama, there is not even the word buddha; even the buddhas of a thousand kalpas appear on the basis of Lamas...these gods are only forms of myself."⁵⁸ With its privileged placement of the spiritual teacher, might this painting have been made by one of Phagmodrupa's disciples, for whom this master was greater than all the Buddhas? The full iconological significance of this remarkable work awaits further study.

It seems that the most critical factor in religious portraiture was to present the subject as an accomplished Buddhist. The apotheosized portrait need not be reminiscent of the man's physiognomy, so long as it conveyed his spiritual accomplishments. These were conveyed by iconographic conventions such as posture, gesture, associated ritual symbols and by other iconographic codes which suggest the subject's inner life. And yet, not all portraits were purely imaginary; elements of an individual's particular physiognomy such as facial shape and features, hair color, hair line, and facial hair may have informed some portraits. Even in these cases, however, the portrayed physiognomy appears to have been based on the notion of physiognomic types.

Why were hierarch portraits made? What purposes did they serve? The Blue Annals states that portrait paintings were sometimes made when a great religious figure died and were then distributed to monasteries and chapels frequented by the hierarch.⁵⁹ Such portraits may have served both public and private commemorations.

Go Lotsawa also describes the portrait's function as icon. That portraits could transmit a spiritual presence is suggested by Go Lotsawa's description of Taglung Thangpa Chenpo's introduction to Phagmodrupa through his portrait. Upon seeing the portrait, Taglung Thangpa Chenpo "felt in himself that he must go and meet his teacher."⁶⁰ Some portraits acted as the focus of rituals in which an individual "absorbed" teachings from the painted image.⁶¹ Traditionally, a consecrated painting is the reflection (gzugs-brnyan; skt., pratima or pratibimba) of transcendent spheres, the "physical support" (rten) of a deity. In the final stages of consecration, the deity is invited to "inhabit" the work of art. Typical of early Tibetan paintings, most portraits include reverse inscriptions of mantras which mark the forehead, throat, and heart centers (skt., cakra) of the deities and religious figures on the painting's obverse. This aspect of portraiture has yet to be fully examined.⁶²

As we have already seen with Rinchen Sangpo's images at Alchi, portraits were also created long after an individual's death. When Ngor (ngor) monastery was established in 1429, its founder Kunga Sangpo (kun-dga' bzang-po) ordered painted portraits of his predecessors in the Sakya school.⁶³ Go Lotsawa reported that a portrait of Taglung Thangpa Chenpo (1142-1210) was made around 1370.⁶⁴ Herbert Franke notes that eleven painted portraits were made of the great Sakya hierarch Phakpa ('phags-pa, 1235-1280) in 1324, and statues subsequently made from these, so that he could be worshipped in the halls.⁶⁵

Approximately one-fourth of the surviving pre-fifteenth century cloth paintings are hierarch portraits.⁶⁶ One wonders why they were so popular. To explore this question, one must look to the social and political factors surrounding the creation of hierarch portraits in pre-fifteenth century Tibet.

Religious hierarchs were always highly revered in Tibetan society. Lamas (bla-ma), or "superior persons" were important in Tibet even before the introduction of Buddhism, a phenomemon which helps to explain why the Indian notion of religious master (skt., guru) was so easily adapted by Tibetans.⁶⁷ King Trisong Detsen (khri-song lde('u)-btsan, ca. 756-797), had leading religious figures sit in a position superior to that of his political ministers, on the same level as the king himself.⁶⁸ And the Sakya hierarch Phakpa, First Imperial Preceptor (ch., dishi) to the Yuan rulers, is said to have insisted that his seat be even higher than that of Kubilai Khan.⁶⁹

After the demise of the Tibetan monarchy in 846, Tibet reverted to an essential feudal social structure, with local aristocrats controlling the land and its resources. During the second introduction of Buddhism beginning around mid-tenth century, no central ruling authority monitored the growth of Buddhist institutions. Religious communities were evolving rapidly, but somewhat haphazardly.

Aspirants journeyed great distances to meet anyone who might claim to possess knowledge of Buddhism. Such individuals sometimes met dozens of teachers before they met one with whom they could commit themselves for prolonged study. Some masters charged high fees for their initiations and teachings. When the would-be disciple Ngok Chodor (rngog chos-rdor, 1036-1102) brought Marpa a horse, he was told: "If this is your offering...for instruction in the Doctrine, it is too small, but if this is your offering ...for an interview, it is too large!"⁷⁰ Ngok returned to present his master with seventy black female yaks, a black tent, a dog, a butterchurn, and a pitcher. He later remarked that it seemed that the master's teachings were as plentiful as his disciples' material resources.⁷¹

During the eleventh and twelfth centuries, there arose many small religious communities, each led by a charismatic figure. Indeed, such individuals were the driving force behind Buddhism and its institutions. They founded monasteries, attracted disciples, and propagated their own version of the faith. The future of their religious institutions depended upon the charismatic strength of the founder, the extent of the material resources at his disposal, and the manifold skills of his successors. There arose what might be described as a "cult of personalism," great emphasis placed not upon schools or institutions, but upon individual teachers as repositories of the faith. Perhaps the relatively large number of hierarch portraits from this period reflects the special prominence of Buddhist leaders in early Tibetan society.

An important element in a hierarch's portrayal, already touched upon but not yet analyzed, is that of his spiritual lineage. Many early portraits prominently feature the hierarch's spiritual lineage, typically along the top and side borders. The purpose of this series of teachers was to demonstrate the founder's association with a reliable, respected lineage of dharma masters, extending from his own master to the much revered Indian masters. The validation of dharma transmission preoccupied Tibetan religious communities, particularly before the reforms of Tsong Khapa (tsong-kha-pa, 1357-1419) and other leading theologians in the late fourteenth century. Although the notion of spiritual lineage is inherent within Buddhism, I believe that the prominence with which these lineages were depicted, particularly in thirteenth century Tibetan portraits, reflects the struggles for spiritual authority which characterised early religious communities.

Personal jealousies and rivalries between groups were common. An elderly monk in Central Tibet admonished a young novice around the turn of the thirteenth century: "The followers of the Mahamudra [i.e., Drigungpas, a sub-sect of the Kagyupas] are great liars. It is better...to visit the seat of the dKa'-gdams-pas [Kadampas]."⁷²

Plate 24

There were also rivalries within groups. Antagonisms within groups often arose during the aftermath of a teacher's death. Most groups had no clear pattern for succession of religious authority. When a charismatic leader died, his disciples usually dispersed, many subsequently founding their own institutions. This pattern of bifurcation, quite innocent at first, became so acute in the thirteenth century that it threatened the future of Buddhism in Tibet. The pitfalls of struggles within groups for the right to succeed a common spiritual teacher is well illustrated by the painting illustrated in plate 24.⁷³

This thirteeth century portrait features Taglung Thangpa Chenpo (stag-lung thang-pa chen-po, 1142-1210), who in 1180 founded Taglung monastery just north of Lhasa, the seat of his sub-branch of the Kagyu school. Many of the surviving early portraits seem to be associated with the Kagyu school, some specifically with Taglung monastery, a phenomenon which may reflect this monastery's privileged position until very recent times. Around 1476, Go Lotsawa wrote that Taglung was the only major monastery to have survived the turmoil of war.⁷⁴ Mongol incursions during the thirteenth century destroyed several Kadampa monasteries, and Phagmodrupa's seat--Densatil monastery--was burned by Mongol and rival Tibetan troops.⁷⁵ Taglung's preservation during this period may be due in part to the intervention of Phakpa who, at Sangye Yarjonpa's (sangs-rgyas yar-byon-pa, 1203-1272) request, agreed to try to protect Taglung from Mongol attacks, a promise he apparently was able to keep.⁷⁶

Taglung's great wealth and power were widely known, inspiring such remarks as that reported by Go Lotsawa: "one was unable to rival even a dog of stag-lung-pa [Taglungpa]."⁷⁷ Taglung Thangpa Chenpo is credited with having patronized the arts, adding buildings to the monastery to house images; and Taglung's third abbot, Sangye Yarjonpa, is said to have commissioned many paintings (ras-bris).⁷⁸

The traditional Kagyu lineage appears in the painting's top register: Vajradhara, progenitor of Kagyu teachings, is flanked by the Indian Buddhist master Tilopa, and his disciple Naropa. Skipping the figure directly above Taglung Thangpa Chenpo, one finds the Tibetans who inherited the Indian teachings: Marpa, his disciple Milarepa, and his disciple Gampopa. Taglung Thangpa Chenpo's teacher, Phagmodrupa--in a gesture of dharma instruction--appears just above his head.

While this lineage is not incorrect, it is somewhat misleading because its apparent linear transmission of religious instruction masks a more complex historical truth. Other teachers had played crucial roles in the religious education of Taglung Thangpa Chenpo. Moreover, Taglung Thangpa Chenpo was among hundreds of disciples who received Phagmodrupa's teachings. The painting's simplified lineage underscored an historical perspective which most served Taglung Thangpa Chenpo and his disciples: it meant to signify that they and their community were rightful inheritors of the spiritual mantel from the great Phagmodrupa.

Plate 25

In fact, Phagmodrupa did not identify a successor, and when he died, his lineage splintered: competing disciples fought to legitimize themselves as Phagmodrupa's prime successor. One of Taglung Thangpa Chenpo's greatest rivals was Drigungpa, seen in a mural at Alchi monastery which was executed about the same time as the Taglung Thangpa Chenpo portrait in plate 24. (pl. 25) This illustration of religious transmission shows essentially the same lineage, except that Drigungpa (the last figure in the lineage, identified by inscription) is featured as Phagmodrupa's successor.⁷⁹ One imagines that portrait paintings commissioned by other disciples of Phagmodrupa would have shown the same lineage, all however presenting themselves as Phagmodrupa's successor.

Although Phagmodrupa had not appointed a successor, Drigungpa assumed the abbot's chair at Densatil, Phagmodrupa's monastic center where both Drigungpa and Taglung Thangpa Chenpo had studied. Drigungpa left under duress in 1179, having served as abbot for only two years. Subsequently, the monastery was often without an abbot. In 1198, Drigungpa and Taglung Thangpa Chenpo built a temple surrounding their former master's famous grass hut at Densatil, where he had lived and taught.⁸⁰ The scene in this painting's lower register may depict this early construction at Densatil. The hut, a cone-roofed dwelling enclosed by monastic structures, can be seen in the painting's lower right corner.

When local people threatened to raid Densatil for its valuables just after the turn of the thirteenth century, Drigungpa took much of the wealth and its famed library to his own monastery in Gampo (sgam-po). Taglung Thangpa Chenpo experienced anguish over the desecration of his master's establishment. Early historians suggest that the incident, which brought their rivalry to a head, led to Taglung Thangpa Chenpo's premature death.⁸¹

The inherently unstable charismatic authority which shaped Tibet's monastic establishments from the eleventh through the fourteenth centuries gradually gave way in the fifteenth and sixteenth centuries to a more stable, institutionalized process of spiritual succession.⁸² This, in turn, made different iconographic demands on painting. Post-fourteenth century portraiture is more likely to depict the founder of a school or monastery, rather than the current abbot--whose role in a spiritual lineage would have been relatively well defined.

The thirteenth century Tibetan pilgrim Chojepel (chos-rje-dpal, 1197-1264) acknowledged the faith required by all iconic art when he observed:

When one prays to [an image]...and the notion arises that it is a real Buddha, then the image truly expounds the Doctrine...But if one [considers] it...a material thing, a fashioned image, then the blessing decrease[s]... [W]hen one beholds sacred images, they [must] be considered to be the very Tathagata...⁸³

Chojepel's observation holds true for portraits of religious hierarchs as well as for the more familiar Tibetan iconography of Buddhas, bodhisattvas, and other divine beings.

As for the aesthetic and theoretical guides to portraiture of the period, it is clear that some degree of observation may have informed portraits done close to the lifetime of its subject, but an artist's primary guide seems to have been the notion of physiognomic types. Sectarian tensions also found their way into paintings: as competing groups sought to legitimize their own authority, their paintings featured spiritual lineages and abbot portraits, designed to show their own contemporary leader as the rightful successor of acknowledged spiritual authorities.

In rendering Buddhist hierarchs, artists frequently relied on iconography already established for the portrayal of Buddhas, bodhisattvas, and other deities, an observance justified by the Tibetan perception of their religious leaders as living Buddhas and bodhisattvas. Tibetan painting has never aspired to accurately record the visual world, choosing instead to depict unseen forces and transcendental worlds. It should be no surprise, then, that portraiture as well aspired to represent not the surface of a man, but his profound qualities as a enlightened being.