Painters, Patrons and Paintings of Patrons  in Early Tibetan Art

Figure 1.  Acacia Forest Tārā, John Gilmore Ford Collection.  

Photo courtesy of John Gilmore Ford.

Painters, Patrons and Paintings of Patrons

in Early Tibetan Art*

Dan Martin

Jerusalem 

NOTE:  This paper is not identical to the published version.  In some ways it is better.  Some minor textual changes have been made.  More significantly, the illustrations are here in the correct order, with the correct labels, and also some are in color where the published version has them in black-and-white.  The reference to the published version is as follows:  “Painters, Patrons and Paintings of Patrons in Early Tibetan Art,” contained in: Rob Linrothe & Henrik Sørensen, eds., Embodying Wisdom. Art, Text and Interpretation in the History of Esoteric Buddhism, vol. 6 in SBS Monographs series, The Seminar for Buddhist Studies (Copenhagen 2001), pp. 139-184.  

In the time that has gone by, the Acacia Forest Tārā illustrated in Fig. 1 has moved to Baltimore and may be seen at The Walters Art Museum.

Try this link as well as this one.  

For purposes of citation, I recommend using the original publication, and not this one. Naturally, now that it is over a decade later, I feel I ought to modify a few of my conclusions, in particular my position has shifted somewhat on the issue of the meaning of thugs-dam in relation to icon construction, but these reconsiderations will have to wait.

Note (October 27, 2014):  Since this paper has provoked a number of reactions (both published and unpublished) from professionals in the world of Tibetan art, both positive and negative, I would like to clarify that I have not "categorically denied" (as I find in a very recent book) that an Indian hand could have been involved in the production of the Ford Tara.  I simply argue that there isn't a shred of evidence for Indian involvement (apart from the “subjective and therefore fallible” grounds of aesthetic judgement).  And I positively argue that all the available evidence of persons involved with this particular painting tells us they were Tibetans, not Indians.  And finally, I argue that Tibetans were perfectly capable of performing the tasks involved, including most especially the painting. I don't pretend to overturn the possibility that this painting was done by an Indian painter,* just to poke holes in our reasons for entertaining it or (as some do) insisting on it as something more than a possibility.  As for my dating of the painting to somewhere between 1164 and 1175 CE, as yet there have been no counter arguments offered against it.

*I draw attention here to instances of Indian artists active in Tibet in 11th-12th centuries, most especially the one named Krishna who painted Atisha's famous portrait painting.

In a recent book entitled Sacred Visions (Kossak & Singer 1998), a catalogue published in conjunction with the exhibition of the same name at the Metropolitan Museum of Art (Autumn 1998), and in the October 1998 issue of Orientations, a general characterization of early Tibetan art[1] seems to be conveyed, as may be seen particularly in the words of Steven Kossak: 

“Although I think we are justified in concluding that the most aesthetically accomplished and compositionally complex paintings are probably Indian commissions…  The Ford Tara, a painting that we put firmly in the Indian sphere…”[2]

If the painting is of superior artistic quality and bears Indian stylistic traits, it simply must have been painted by an Indian artist.  From this we may easily infer that mediocre paintings are probably by Tibetans.  While Jane Casey Singer’s contributions to the same publications do not adhere to Kossak’s generalizations, in the particular case of a Tārā thang-ka from the John and Berthe Ford Collection, she does believe that it is likely, on primarily stylistic and aesthetic grounds, to be the work of an eastern Indian artist.[3]  The intentions of the following notes are basically two: to present evidence that might support or undermine Kossak's arguably Indocentric stance, and to attempt to illuminate the relationship between the main subject matter of the thang-ka and the small figures depicted in a lower corner of many early thang-kas in what is clearly some kind of ritual scene.  As an example to illustrate and test both issues, we will concentrate on the Ford Tārā, in the process presenting some fresh arguments for its provenance, based in large part on a close rereading and contextual reinterpretation of its inscriptions.[4]  (Fig. 1)

Part of our task will be to add to, but of course not exhaust, the fund of literary sources on early paintings, and especially those produced by Indian artists.  The first story to be brought in evidence was noticed already in English in 1984 (p. 43, note 6) and studied in more detail in 1996 (p. 370) by David Jackson.  Despite its relevance as a particular example of a Tibetan commissioning a work of art from an Indian artist, it seems to have been ignored by other art historians.  It may be worthwhile to return to this story once more, since earlier sources than those used by Jackson have been located.  The following account is from a brief biography of the Nag-tsho Translator, as told in the most detailed of the early histories of the Bka’-gdams-pa school, that by Las-chen (1494/1972: I 197-9).  Nag-tsho is best known for his role in the invitation of the celebrated Bengali teacher Atiśa to Western Tibet.[5]  Born in 1011, Nag-tsho was certainly a western Tibetan, given the evidence of one of the names by which he is known, Gung-thang-pa, ‘the person from [Mang-yul] Gung-thang.’  Mang-yul Gung-thang, a small kingdom which would later initiate a rich artistic history all its own, lies just to the north of western Nepal.  Nag-tsho was his clan name, while his given name was Tshul-khrims-rgyal-ba.  He was also known under a title common during these centuries, ’Dul-ba ’Dzin-pa, ‘Vinaya Holder,’ because of his knowledge of monastic codes.  As a youth he studied Abhidharma literature with Rgya Brtson-seng and spent a few years in India.  In his 27th year, he went to India to invite Atiśa, and continued to study with him faithfully for nineteen years (but note the chronological problem this would seem to cause, since Nag-tsho was born in 1011, and Atiśa died in 1054).[6]  Then Nag-tsho took his leave of Atiśa in order to move to Western Tibet.  As parting gifts he received an image and copies of various texts, including an Indian manuscript of the Guhyasamaja Tantra.  At this point we switch from paraphrase to direct translation:

(Note that the short-hand conventions of cursive manuscripts have been tacitly standardized in the following transcription.)  jo bo nyid kyi sku mthe bong tsam sing lding la bris te ga’ur bcug nas dpung pa la btags / mnyes mnyes rgya cher mdzad / de nas yar byon nas ras khru bcu drug pa gcig la ri mo mkhan mkhas pa kriṣṇa bya ba rgya gar ba gcig yod pa ’drir [i.e., ’brir] bcug nas / stod la jo bo’i yi dam gyi lha rnams bris / de’i ’og na phar la jo bo’i bla ma bcu gnyis po bris de nas bla ma nyid kyi sku tshad khrus gzhal te bris / g.yas g.yon gnyis na nye gnas re ldem skur bris / g.yas g.yon gyi kha ras la jo bo’i mdzad pa rnams dang / ’og na tshur khu rngog ’brom gsum la sogs pa bod ston rnams chos gra’i tshul du bris / de’i mdun du lo tstsha nyid [g]sol ba ’debs pa’i tshul du bris / de’i rgyab la jo bo’i bstod pa brgyad cu pa bris / rab gnas mdzad pa’i tshe / bla ma nyid dga’ ldan nas ’byon pa zhal gyis bzhes pas / ’di bla ma nyid dang khyad med pa yin / mnga’ ris skor gsum gyi rten la byin brlabs che shos yin gsung.[7]

“He drew [or painted] an image of himself, the Reverend [Atiśa], about the size of a thumb, on acacia  wood, and after placing it in a reliquary (ga’u), attached it to his [Nag-tsho’s] shoulder.[8]  This made [Atiśa, evidently] very pleased.

“Then, having gone thither [lit., ‘upward’], he [Nag-tsho] had a skilled painter named Krishna, an Indian who was there, paint on a sixteen-cubit [high] canvas.  At the upper part he painted the aspirational deities of Atiśa.  Under them and to the sides he painted all twelve gurus of the Reverend.  Then he painted the image of none other than the guru, its size to be measured in cubits.  To his right and his left there were [two] attendants, painted as figures inclining [toward him].  On the right and left canvas surfaces were the deeds in Reverend’s life, and on the lower part [of the canvas] he painted the Tibetan teachers, including the triad of Khu, Rngog, and ’Brom facing hither (near him, toward him? tshur) as if they were in the Dharma instruction area (chos-grwa).  In front of them he painted the Translator himself in an attitude of prayer ([g]sol-ba ’debs-pa).  On the back of it [the thang-ka], was written the eighty-verse praise of the Reverend.  At the time of the consecration ritual, the Guru himself came down from Tuṣita Heaven and attested to it,[9] and so it has been stated that this [thang-ka] is indistinguishable from the Guru himself, that it is the greatest in blessing among the receptacles of Western Tibet (Mnga’-ris Skor Gsum).”[10]

With the help of Helmut Eimer’s (1977 & 1979) fine textual studies of the sources for the life of Atiśa, it was possible to locate older witnesses for this story.  The following passage is to be found in the Atiśa biography composed by Mchims Nam-mkha’-grags, who served as abbot of Snar-thang Monastery from 1250 until his death in 1289.

dge bshes lo tsā bas ras khru bcu bzhi pa zhig la rgya gar gyi lha bzo mkhas pa kri sna pa zhes pas bris te / stod na phar jo bo’i thugs dam gyi lha rnams bris / de’i ’og na phar jo bo’i bla ma bcu gnyis po bris / de’i ’og jo bo nyid kyi sku chad [i.e., tshad] du khrus gzhal ba gcig bzhengs / g.yas g.yon gnyis su nye gnas gnyis ldem skur byas / g.yas g.yon gyi kha ras la jo bo’i mdzad pa rnams bris / de’i ’og tu khu rngog ’brom gsum la sogs pa bod ston rnams chos grwa’i tshul du bris / de’i mdun du lo tsā ba gnyis [sic! for nyid] gsol ’debs kyi tshul du bris / de’i rgyab tu jo bo’i bstod pa brgyad cu pa bris nas / de la rab gnas mdzad pa’i dus su dge bshes lo tsā ba’i zhal nas bla ma nyid kyis dga’ ldan nas byon par zhal gyis bzhes pas ’di bla ma nyid dang khyad par med pa yin gsung / de mnga’ ris skor gsum gyi rten la gtsigs [i.e., rtsis] che ba yin skad /  da lta yang thog gi lha khang na bzhugs shing yod gsung.[11]

There are particular readings here that might necessitate a somewhat different interpretation, but rather than translate the passage all over again, I will just point out that the size of the painting is here only fourteen cubits instead of sixteen and that, at the time of writing, it was still to be seen at the temple of Yang-thog.[12]  In still another biography of Atiśa contained in the same section of the Bka’-gdams Glegs-bam, there are interesting references to still other thang-kas (or does it contain a severely condensed and somewhat distorted description of the same giant thang-ka?).

dge bshes gung thang pas jo bo’i sku gcig dang / u pa si ka’i sku gcig zhal sprod kyi ras bris shin tu che ba zhig mdzad / u pa si kas lha btsun pa dang / gung thang pa gnyis ldem du byas pa’i jo bo’i sku ’dra chen po zhig mdzad pa la / bal po’i rgyal pos skyel ba dang / mnga’ ris skor gsum gyis bsu ba la sogs pa ya mtshan can sna tshogs kyang bris la bkod do.[13]

“Geshe Gung-thang-pa [Nag-tsho] made a very large canvas painting of one image of Reverend [Atiśa] and one image of Upasika[14] face to face with each other.  Upasika made a great likeness of Reverend in which Lha-btsun-pa [the Western Tibetan king] and Gung-thang-pa were inclining [toward Atiśa as the central figure], and on it were painted as well all sorts of marvels, including the sendoff by the Nepalese King and the reception by the people of Western Tibet.”

This would seem to describe a different thang-ka made by Nag-tsho along with yet another made by Atiśa’s attendant Upasika, unfortunately without any indication of who the painters might have been.

The verses written on the back of Krishna’s giant (more or less twenty-four feet high) Atiśa thang-ka are significant as the only existing text known to have been composed by Nag-tsho,[15] and this set of verses has formed the subject of studies by Helmut Eimer (1989) who indeed calls it the earliest existing source for the life of Atiśa.[16]  The story of Krishna’s thang-ka would appear to show not only that in the middle of the eleventh century a Tibetan could commission an object of devotion from an Indian artist resident in Tibet (or was he in Nepal?), but also that the patron (a term that will require further qualification) could have himself depicted in an attitude of reverence in the ‘foreground’ at the bottom of the thang-ka, and even further that a thang-ka may be said to be fully identical to the figure depicted, that the impact of viewing the thang-ka would be the same as being in the bodily presence of the holy being, and all of these latter statements could equally apply to the Ford Tārā thang-ka.

Before going on to discuss the Ford Tārā thang-ka, I would like to look at what I believe could be among the prototypes of that thang-ka.  The source is a modern dkar-chag[17] to Reting (Rwa-sgreng) Monastery of Central Tibet.  The author notes that only a few of the things he lists exist today at Reting, and he explicitly based the work as a whole mainly on a previous dkar-chag without any authorship attribution.[18]  The following passage is one of many entries in a list of the main building’s thang-ka holdings, which once numbered over 3,600.[19]  Included in this list is a thang-ka constructed by Nag-tsho, which was then consecrated by Atiśa himself, one based on a print of Atiśa’s feet.[20]  Here is another entry, for our purposes more relevant, from the Reting Monastery list:

jo bo rje’i thugs dam gyi rten sgrol ma ’jigs pa brgyad skyob ma ’di ni / rnal ’byor pa chen pos rgya gar du bzhengs su btang ste / shar phyogs sems dpa’ mkhas pa’i chos sgrol mdzad dang / nā lantra la bzhugs pa’i sgrol ma rang byon dang / ma ga dha na byang chub chen po’i drung du mchod rten brgyad pa bzhengs te tshur byon nas jo bo rje snye thang du bzhugs skabs rab gnas brgya rtsa brgyad mdzad cing gsol ba btab pas / jo bo rje la dam pa’i chos gsungs pas gsung byon ma yin no //  jo bos dge bshes ston pa la gnang nas ’di la gsol ba thob dang khyod kyi ’dod don thams cad ’grub nas ’ong gsungs pa sogs byin rlabs can du grags.

“This Eight Fears Protector Tārā [thang-ka], a receptacle (rten) of the high aspiration (thugs-dam) of the Reverend Lord [Atiśa] — Great Yogi [i.e., Rnal-’byor-pa Chen-po Byang-chub-rin-chen, 1015-1078, abbot of Reting from 1065 to 1078] was sent to erect (bzhengs) it in India.  He erected the deeds of Tārā which is a teaching of the learned beings of Eastern [India?], the self-produced Tārā which was kept in Nälandā, and the eight stūpas near the Mahābodhi in Magadha, and after arriving back [in Tibet], while the Reverend Lord was staying in Netang Monastery, he performed one hundred and eight consecration rites and made petition.  Because it preached the Holy Dharma to the Reverend Lord, it is a Speaking [Icon].  The Reverend granted it to Dge-bshes Ston-pa [i.e., ’Brom-ston, founder of Reting] and said, “Make your prayers to this.  All your desired ends will be achieved.” and so forth. It is widely known to possess blessings.”[21]

Although there are certain difficulties in the reading, which would at least in part explain difficulties encountered in reading the translation, it is clear that this Eight Fears Tārā was, using typical vocabulary of the dkar-chag genre, ‘erected’ by Atiśa’s disciple Great Yogi, the Indian painter[s] not specified, although it was Atiśa’s ‘high aspiration’ (thugs-dam) that initiated and accomplished it.  Some might feel justified in assuming that this particular thang-ka could be identified as being the Ford Tārā.  It would seem to have the same main-subject, the Eight Fears.  It, too, was kept as Reting Monastery.  But there are at least two reasons why the Ford Tārā cannot be this product of Atiśa’s high aspiration, the first being that the Ford Tārā was the high aspiration of one Bya Brtson-’grus-’od.  Secondly, Atiśa’s thang-ka would not have had the small images that may be identifed as Atiśa and ’Brom-ston,[22] and neither (I would argue) would it have had a small Tibetan monk engaged in a ritual scene in its lower right corner.  (Fig. 2)

Figure 4.  The red inscriptions, in the shape of a chorten, in the center of the back of the Ford Tārā.  Only the lower part is visible here.

Photo courtesy of John Gilmore Ford.

On the back of the Ford Tārā are two sets of inscriptions:  those in black running along the top (Fig. 3), and those in vermilion at the center.  (Fig. 4)  The vermilion inscriptions together form, in a very traditional form of ‘concrete poetry,’ the shape of a mchod-rten.  The first six lines are in an Indic script usually referred to as Lañtsa (or Rañjanā).[23]  The first three of these lines are mantras, while the the last three are the ubiquitous ‘Buddhist Creed,’ the Ye Dharmāḥ, ending with the usual Tārā mantra and a few illegible letters.  These six lines ought to be studied by an expert in Indian (and Tibetan Indian) paleography, and no more will be said about them here.  The four vermilion lines which form the ‘throne’ of the mchod-rten are in cursive Tibetan script.  These four lines are not completely legible in my photograph, but the already published version (Kossak & Singer 1998: 59, n. 8) is basically sound.[24]  The four-line verse is a very famous one from the Dhammapada, and corresponds fairly well to the following English translation from the Pāli:

Patience that is enduring is the best discipline.

Nirvana, say the Buddhas, is the highest goal.

Whoever hurts another is not a monk.

Whoever insults another is not a renunciate.[25]

or, to freshly translate the Tibetan,

Forbearance is the supreme ascetic practice, forbearance is

supreme nirvāṇa, said the Buddha[s].

The renunciate who harms another and who

injures another is no śramaṇa.

In the original context in the Dhammapada, the message is a purely moral one.  The Buddhist virtue of ‘patience’ (Tibetan bzod-pa; Sanskrit kṣānti) includes within its semantic range the additional concepts of longsuffering, forbearance and tolerance.  In its use in inscriptions on receptacles, however, since it begins with the word ‘forbearance,’ it would seem to have the additional petitionary function of reminding the Buddhas of their transcendent ‘forbearance’ in case of unwitting mistakes by the artist, scribe or consecrator.[26]  Requests for forbearance are extremely common in author, printer and scribal colophons to literary works.

Two distinct inscriptions in black ink at the top of the back of the Ford Tārā provide crucial historical information.[27]  (Fig. 3)  The first, written in a different hand from the other, simply says that it is a ‘divinity of [the] Reting [community].’[28]  This I believe provides us with no further information than that this thang-ka once belonged to Reting Monastery; it hardly provides much if any clue for dating, since it could have become a part of that impressive thang-ka collection any time since the monastery’s founding in 1056.[29]  It is a mark of ownership.[30]  The other inscription provides very important information.  It tells us three basic things:  1. who had the high aspiration to have it erected, 2. who consecrated it, and 3. the Dharma Protectors to whom it was entrusted.

bya brtson ’grus ’od kyi thugs dam / sa’ [se’] spyil phu ba’i rab gnas gzhugs / mchad kha ba’i / spyil phu ba’i chos skyong la gtad do /  (reading as found in Kossak & Singer 1998: 59, note 10).

“[This is] the high aspiration of Bya [B]rtson-’grus-’od.[31]  [In it] abides the consecration of Se Spyil-phu-ba.  [It] was entrusted to the Dharma Protector/s of Mchad-kha-ba [the resident/s of Mchad-kha, or Bya ’Chad-kha-ba himself?], of Spyil-phu-ba [the resident/s of Spyil-phu, or Se Spyil-phu-ba himself?].”

The photographs of this inscription allow a few improved readings, one of them quite significant for the interpretation.  In the interest of exactitude, I have here represented the intersyllabic dot, the tsheg, with a dot (·).[32]  The three-syllable insert beneath the line, which might possibly be a correction or an ‘update,’ is enclosed in brackets {} at the point which seems to be marked with a short downward stroke to the left.[33]  The underlinings point to ligatures that are unclear or invisible in the photographs (due to fading, blotching or the glare of the camera flash).

bya·brtson·’grus·’od·kyi·thugs·dam·  /  sê·spyil·phu·ba’i·{mchad·kha·ba’i·} rab·gnas·bzhugs·  /  spyil·phu·ba’i·chos·skyong·la·gtad·do· /

“[This is] the high aspiration of Bya Brtson-’grus-’od.  [In it] abides [did abide, will abide?] the consecration of Se Spyil-phu-ba {[and/or] of Mchad-kha-ba}.  [It] was [will be?] entrusted to the Dharma Protector[s?] of Spyil-phu-ba [of the resident/s of Spyil-phu or, less likely, of Se Spyil-phu-ba himself?].”

Figure 5.  Retracing of a woodcut of Bya ’Chad-kha-ba Ye shes-rdo-rje (1101-1175 ce).  

Bya Brtson-’grus-’od has been given, by Singer (in Kossak & Singer 1998: 57), the death date of 1175.  This is based on an unarticulated argument that he ought to be identified with Bya Mchad-kha-ba, who did die in 1175.  (Fig. 5)  As reasonable as this identification might seem, given that “Mchad-kha-ba” does occur later on in the inscription, there is no basis for it in the available biographical material on Mchad-kha-ba, and this unsubstantiated — I would even say, to the best of my knowledge, unwarranted — identification vitiates several of the historical statements that are in turned based on it.  There were a number of early Bka’-gdams-pas with the clan name Bya, and there is no special reason to assume that Bya Brtson-’grus-’od is identifiable with any one of them.  All we can do is state that he is, for the time being at least, unknown, as is very often the case with the patrons of art works in Tibet.  It might be possible to argue that he was a monk, since the element ’od, ‘light,’ as the final element of monastic names is quite common in these centuries.[34]  Little more may be said about him, unless, as will be argued in due course, he is the subject of the portrait in the left lower corner of the thang-ka.

Figure 6.  Retracing of a woodcut of Se Spyil-phu-ba Chos-kyi-rgyal-mtshan (1121-1189 ce) from the same work as Figure 5.

Se Spyil-phu-ba is quite a well known figure.  (Fig. 6)  A brief biography of him is to be found in Las-chen’s history.[35]  He was born in Upper Gnyal, at a place called Dar-ma Sgang in the year 1121, when Bya ’Chad[-kha-ba] was in his seventeenth year.  Se was his clan name.  He took full ordination under the celebrated logician Phya-pa, receiving the name Chos-kyi-rgyal-mtshan.  He studied Mind Training (Blo-sbyong) for twenty-four years with ’Chad-kha-ba.[36]  In 1164, as prophesied by ’Chad-kha-ba, he founded New ’Chad-kha Monastery.  Later on he founded the monastery of Spyil-phu, and still later the tradition began in which the summer sessions would be spent at Spyil-phu, and the winter sessions would be held at ’Chad-kha.  After the death of ’Chad-kha-ba, from 1176 until 1189, he stayed in the upper and lower monasteries for a total of fourteen years.  He died at Spyil-phu in the summer of 1189, his 69th year.[37]

This historical information certainly helps us in interpreting the last part of the inscription, and not incidentally, in dating when this thang-ka must have been made.  It could not have been made before the founding of Spyil-phu monastery, which occurred soon after the year 1164 (because Se Spyil-phu-ba absolutely could not have been called by this name prior to his founding of the monastery of Spyil-phu).  Although it depends on our reading of the ‘insertion,’ it ought to have been made before ’Chad-kha-ba’s death in 1175.  This is about as close as we may hope to come to a precise dating of such an early painting, in the absence of an inscriptional date.[38]  Our new reading of the inscription does inject a note of ambiguity as to whether the consecrator was Se Spyil-phu-ba and/or [Bya] Mchad-kha-ba.  It seems to me rather likely at the moment that Mchad-kha-ba’s name was inserted at this point in order to replace the name of his disciple Se Spyil-phu-ba.[39]  This may mean, although it is nothing more than a guess, that the thang-ka was originally supposed to have been consecrated by Se Spyil-phu-ba, but in the end it was actually (also?) consecrated by his teacher Mchad-kha-ba, and so the inscription had to be adjusted to that fact.  Variant story lines might be invented, but it would be wiser to be satisfied with the relatively secure information we do have.

The last part of the inscription tells us that the thang-ka was entrusted to [the protection of] the Dharma Protectors of the monastic community of Spyil-phu which gave Se Spyil-phu-ba his name.[40]  Anticipating possible objections it may be important to emphasize that receptacles (rten) of the Buddha-body, Buddha-speech and Buddha-mind (and temples housing them) are supposed to be, and almost invariably are in traditional Tibetan culture, consecrated as soon as possible after they are completed, the fear being that, if left for any length of time unconsecrated, ghosts or mischievous spirits might take up domicile in them.[41]  Hence, there would have been very little lapse of time between the completion of the thang-ka and its consecration.[42]  Similarly, the thang-ka would have been entrusted to the protectors simultaneous with the consecration.[43]

For contrast, the translation against which I have proposed my alternative reading (Singer 1998: 67; almost identical to Singer in Kossak & Singer 1998: 57, cf. p. 16), reads as follows:

“The personal meditational image [or meditational commitment] of Chason Dru-o. Consecrated by Sechilpuwa of Chekhawa. Placed [opposite] the Protector Deity at Chilbu [monastery].”

I will not try to justify every point with which I differ.  But I would at least like to point out that particular words such as thugs-dam and gtad do indeed have what would seem to be broad ranges of meanings, but I am quite sure nevertheless that when used in these particular contexts they do not mean ‘personal meditational image’ or ‘placed [opposite]’ as the following examples should make clear.  When Emperor Khri-srong-lde-btsan planned to build the temple complex of Bsam-yas, Bsam-yas was his ‘high aspiration’; it was not his ‘personal meditational image.’  For dozens of examples of gtad being used in the sense of entrusting the protection of a holy object or temple to Dharma Protectors, see Sba-bzhed (44-51), and for examples of thugs-dam being used to refer to a ‘receptacle’[44] that is being brought into existence, see Sba-bzhed (37-38, 42, 44, 48, 53, 54, etc.), and for inscriptional examples, the transcription and translation by Hugh Richardson (contained in Pal 1983: 259) and, still more chronologically relevant, an image inscribed as being a thugs-dam of the Western Tibetan King Nāgarāja, active in the first decades of the 11th century.[45]  Another word frequently encountered elsewhere in inscriptions and in dkar-chags, bzhengs, ‘to [have] erect[ed],’ does not mean that the person performing this virtuous activity actually lifted a finger in the physical act of making anything, the contrary being most likely the case.

One sentence is provided here as an example, selected because it would have been composed close to the year 1200 (and therefore reasonably close to the time of our inscription), found in the historical work by Nyang-ral Nyi-ma-’od-zer.  This is one among many sentences that could be cited from that same work, which demonstrate the meaning of the word thugs-dam, particularly when it is, as in the inscription, preceded by a substantive in genitive case.  The sentence refers to the activities of Emperor Sad-na-legs, who ruled the Tibetan empire in the first decade of the ninth century.  It reads, yab-mes-kyi thugs-dam kun zhig-ral gso-ba, which translates, ‘[he performed many deeds, among them] restoring all the thugs-dam of [his] predecessors [lit., ‘father-grandfather’ or ‘ancestors’].’[46]  Here the word thugs-dam stands in place of the temples built by his father Khri-srong-lde-brtsan and earlier members of the dynasty.  To those who want to argue that thugs-dam and yi-dam are the same word, a simple response should be sufficient.  However similar their meanings might be (and I think they are quite similar), the contexts of their usage are different.  The word yi-dam is never in my experience used in place of thugs-dam in the context of describing the meritorious construction of an icon.  However, that an meritorious construction that is a thugs-dam of a certain person might include a depiction of a yi-dam that forms a focus for that person's religious practice is by no means excluded, quite the contrary.

In the published catalogue it is speculated that the Ford Tārā probably predates the main black inscription by seven decades.  Since this is largely, if not quite entirely, based on the understanding of what thugs-dam is believed to mean, we might look rather briefly at one other inscribed thang-ka, one of Vajravārāhī, that is likewise predated on the same premise.[47]  There are two lines of reasoning that can prove that this thang-ka, also, could not predate the one whose thugs-dam it was. One is based on the reading of the inscriptional evidence (for which we must rely on the published transcriptions), the other on the subject matter of the thang-ka itself.  Centered below the central field of the thang-ka is a gold inscription: bla-ma rin-po-che Dbon-po Dpal-gyi thugs-dam lags //, ‘This is the high aspiration of the precious lama Dbon-po Dpal.’[48]  Dbon-po Dpal (‘Glorious Nephew’) is quite correctly identified as the one who served briefly as abbot of Stag-lung, after which he founded Khams Ri-bo-che.  He is mostly known by the fuller monastic ordination name (not a birth name as stated in the catalogue entry) of Grags-pa-dpal-’od-zer-bzang-po, or more simply as Sangs-rgyas Dbon (1251-1296).[49]  Another gold inscription is on the back of the painting.  My translation of this inscription (transforming the Sanskritized Tibetan names back into their ‘natural’ Tibetan forms) follows:

“The incomparable holy Lama Shes-rab-bla-ma and

I, Grags-pa-dpal-’od-zer-bzang-po, may we not [ever] be separated.  While carrying out the words he has expressed

may the mistaken notions of my own mind be purified and

may I prove able to guide animate beings.”[50]

Lama Shes-rab-bla-ma is, in the catalogue, rightly identified as the 3rd abbot of Stag-lung with the fuller name Sangs-rgyas-yar-byon Shes-rab-bla-ma (1203-1272; he ascended the abbatial throne in 1236).  In my interpretation of this inscription, ‘not being separated’ implicates a wish to meet again and again in future lives; this makes it seem probable that the thang-ka was made after the 3rd abbot’s passing in 1272.  Two further inscriptions, closely paralleled in inscriptions found on a number of other early paintings, as noted in the catalogue, are found along the top of the back side, one in gold the other in red.  As argued in the catalogue, the red one is probably a recopying for one purpose or another of the gold inscription.  My difference with the translation contained in the catalogue[51] turns around the understanding of the word yan-chod.  I will first suggest my own translation, and then argue for and against it.  I translate the complete red inscription, which I think might have been meant to replace the earlier (?) incomplete (partly effaced?) gold inscription (I cannot feel sure of that point), as follows:  ‘[Herein] abide the innumerable consecrations of four [generations of spiritual] fathers and sons from[52]chos-rje rin-po-che[53] upward to [i.e., as we will argue, moving back in time to] bla-ma rin-po-che Dbon-po Dpal.’  Now yan-chod (more frequently spelled yan-chad) and its opposite man-chod (more frequently man-chad) belong to an interesting set of postpositions that share similar morphological patterns as charted out in detail in Michael Hahn’s Classical Tibetan grammar as well as in an article especially devoted to the subject by Zuiho Yamaguchi.[54]  When used in a spatial context, the usage of yan-chad is quite clear and uncontroversial.  Take any arbitrary dividing point in vertical space (social hierarchies included), and whatever lies upward of that point is yan-chad in relation to that point, while everything downward from that point is man-chad.  When used in a temporal context, however, we are liable to run into problems.  The example used in the ‘standard’ Tibetan-English dictionaries of Heinrich August Jäschke (1972: 506) and Sarat Chandra Das (1973: 1128) is misleadingly defined.[55]  The confusion created by their misdefinition may be most easily circumvented by going directly to a Tibetan-Tibetan dictionary and to some actual examples from Tibetan literature.  The dictionary entry for yan reads, in translation:  ‘the upper part or the preceding part of an arbitrarily (lit., ‘as one likes’) chosen boundary in something.’[56]  The literary examples may prove tedious for most readers, so they will be relegated to a footnote.[57]  While this body of evidence may not be entirely persuasive, I believe that, at the very least, it does inject a significant note of uncertainty.

The second argument is based on the series of teachers depicted at the upper register and on two further teachers on the thang-ka’s right and left sides.  Both logically and iconographically, these two further teachers may be identified as the 2nd and 3rd abbots of Stag-lung, respectively.[58]  This argument, if acceptable, leaves the place for the ‘patron’ in the lower corner of the thang-ka open for the next member of the lineage — the one whose high aspiration the thang-ka was, as we know from the front inscription — the 4th (but only very temporarily so) abbot of Stag-lung, Dbon-po Dpal.  Hence the thang-ka cannot be dated to circa 1200, as it is in the catalogue, since the 3rd abbot only became abbot in 1236, and Dbon-po Dpal had not met the 3rd abbot until 1273.  Even if we were to disregard our understanding of yan-chod, from this argument based on subject matter alone we may know that the series of repeat consecrations referred to in the black and red inscriptions on the back occurred after, not before, Dbon-po Dpal.

Turning now to the question of the identity of the person (or persons) depicted in the lower corner of many an early thang-ka in a ritual scene, I believe it is possible to be more precise about the role of this figure.  He and/or she is the practitioner (sādhaka/ā) who aspired to have it made as part of their sādhana practice (and most likely also the [strictly speaking] patron who brought its construction to completion), the one whose thugs-dam it is.[59]  In the case of the Ford Tārā, it is the otherwise unknown Bya Brtson-’grus-’od.  (Fig. 2)  Like almost all such figures, he and/or she is located at the most remote lower corner, not depicted frontally, but rather inclined toward the center of the thang-ka, also here as in so many cases seeming to face upward toward the central figure, and even as in somewhat fewer other cases in a special kneeling posture.[60]  Most of these figures hold a censer with a long branch- or tendril-shaped handle, although in the present case it is not in the hand, but standing on the ‘ground’ below and to the side of the figure.[61]  The general picture of the patron corresponds quite closely with a description in Chapter Four of the Mañjuśrīmūlakalpa:

“Below the whole great lake [of the nāgas] the sādhaka should be painted in the extreme corner at the cloth painting’s right side, with accoutrements and color[s] as previously described, looking at the maṇḍala of Saintly Mañjuśrī’s face, holding the incense vessel in the hands, bowing the head and the knee.”[62]

Somewhat similar information about the making of a patron painting may be gleaned from this quotation taken from the Acalakalpatantrarāja, Chapter Three, a chapter on painted images:

“After painting [Acala] like that, having drawn with various pigments a portrait of oneself and having well produced an understanding in the painter, one offers beauty as a gift, pronounces words of auspiciousness.  Then, keeping it in a separate part of one’s own dwelling, one must visibly practice being Acala.  Having finely made offerings of flowers and so forth, the total knowledge wrath king will dissolve into that same [image].”[63]

In front of most of the painted patron figures is an upright stand with a pair of fly-whisks (sometimes referred to as yak tail fans, Indians have long esteemed this Tibetan export item) crossed across its single leg, and, on the top of the stand, a (“Kadampa style”) mchod-rten and a flat object probably representing a book.[64]  This seems to represent the ‘three receptacles,’ the receptacles of Buddha-body (not figured here, since the thang-ka itself represents this), the receptacles of Buddha-speech (the book), and the receptacles of Buddha-mind (the mchod-rten).  Although the origins of the practice are obscure, the three receptacles are still used as offerings made by the patron of a ritual to the Vajra Master at the beginning of the ritual, and in old Tibet when granted audience with a very high spiritual dignitary like H.H. the Dalai Lama.  The fly-whisk is well known as a royal symbol, and seems to connote the bestowal of royal honors and tribute.[65]  In most examples, as in this case, there is a set of offerings in a rather abstract array, rather than resting on an altar as they would be in actual practice (in some examples, they are in a row, each on its individual stand, and in at least one example[66] they are arranged on four pairs of limbs attached to a single vertical stand).  In the majority they seem very clearly to represent the five upacāras, which might be increased to eight.  The five main ones are flowers (rice usually substituted), incense, light, fragrance and food (water usually substituted[67]). With the addition of one or more water offerings (which I believe may be symbolized in the paintings by the multiple conch shells, used in rituals as pouring implements in Tibet as well as ancient Mesopotamia), a torma (gtor-ma) in many cases, or an offering of music (not usually represented as such), there may be eight altogether (special further offerings may be required in the cult of a particular focus of high aspirations [yi-dam], such as Vajrayoginī, but augmenting, not replacing, the upacāras).  This altar setup is found not only in nearly every ritual, but also in every Tibetan Buddhist altar, whether it be in a temple, a household or a monastic cell.  The suggestion has sometimes been made that this scene depicts a consecration (rab-gnas), which would imply that the figure depicted is the ritual officiant, but since none of the special implements necessary for consecration are present (perhaps a spouted pitcher and a mirror[68] would have been appropriate in this case), there is really no reason to think that this is what is happening.  It is probable that these figures are meant to represent patrons in one of the closing rites of the consecration ritual in which patrons are exhorted to make offerings to the receptacles they have brought into being.  In fact, this rite, called ‘exhorting [or commanding] the patron[s]’ (yon-bdag bsgo-ba or sbyin-bdag bsgo-ba) constitutes the very first offering made to the newly consecrated receptacle (see Bentor 1996: 334 ff.).  This is not to say that the ritual officiant of the consecration might not also be depicted.  This seems clear in one early thang-ka at least (Kossak & Singer 1998: plate 36c), but the one identified as the ‘patron’ in the prose description, located closer to the center of the thang-ka holding a (scarcely visible) vajra and a bell, is rather the Vajra Master.  The one in the far corner with the ‘ritual’ setup and holding the censer is, again, the patron, not the “consecrating monk.”  In the one corner-scene in which there is clearly a hierarch on a high throne with an elaborate altar setup including pitchers, the term sbyin-pa’i bdag-po (the long form of sbyin-bdag), ‘patron,’ is inscribed immediately below him, as if to dissolve any possible misunderstanding (plate 55, a thang-ka tentatively dated to the early 15th century). After all, there is nothing to prevent a hierarch from being a patron, too, no reason why a hierarch should not be both patron and consecrator.[69]

A number of art-historical writers have cast their votes for the ‘Indian’ nature of this thang-ka, some going even further in saying it ‘must’ have been done by an Indian painter.  Regardless of any currently or soon-to-be recognized Indian stylistic traits in this particular thang-ka, there is not the least shred of evidence, apart from the “subjective and therefore fallible”[70] grounds of aesthetic judgement, that an Indian hand was involved in its making.  It may not be, therefore, very surprising to find no unanimity among the art historians.  Recently, one art historian, Eva Allinger (1998: 113), has concluded, after weighing the several perceptible Indian elements, that it nevertheless “possesses a distinctive Tibetan character.”  Thirteenth century and still earlier Indian palm leaf manuscripts have survived in Tibet, and of course there are many examples of Indian metalworks to be found in Tibetan monasteries even today.  So here there is no intention to dampen the Indological hope that cloth paintings by Pāla period Indian artists might have been preserved in Tibet.  There is just no compelling reason why the Ford Tārā ought to be placed within this so-far empty set (empty, that is, if we agree to temporarily disregard the literary sources on paṭas).

Did Tibet suffer from a lack of qualified native artists?  In the history by Las-chen (I 28, 500-502), there are several mentions of a Bka’-gdams-pa teacher active in the 1110’s who was so famous as an artist that he is simply called Dge-bshes Lha-bzo-ba, ‘Geshe Artist,’ and he was known as such ‘because he knew painting skilfully’ (ri-mo mkhas-par mkhyen-pas). In Phag-mo-gru-pa’s biography, it says that as a child he could draw deities without ever being taught (for this and still other testimonies on 12th-century Tibetan artists, see Jackson 1996: 69-70).  His biography specifically mentions ‘likenesses’ (sku-’bag), presumably painted or at least drawn by him, of his teachers Sgam-po-pa and Sa-skya-pa, as well as of his teacher in visions only, Dam-pa Rgya-gar (an early way of referring to Pha-dam-pa Sangs-rgyas, active in Tibet in the late 10th and the first decades of the 11th centuries).  Among the disciples of Zhang Rin-po-che (1123-1193) were several known for their more practical skills, including at least two artists (lha-bzo).[71]  In an autobiographical passage of a circa 1180 work by Zhang Rin-po-che, in the course of relating events that occurred in the late 1140’s, he tells of an encounter, in Khams province, with an artist (lha-bzo) who was native to Stod-lung in Central Tibet in the act of painting a Wheel of Life (Zhang, Samdo A: III 120).  An early 13th-century history of the Zhi-byed school by Zhig-po Rin-chen-shes-rab (1171-1245), in his recounting of the life of his own teacher Rten-ne (1127-1217?), tells how Rten-ne, in about 1149, decided to visit a painter to borrow a book from him.  The painter was a native of the lower ’Phan-yul valley by the name of Bran-ka Lha-bzo Klu-btsan-grags.  When he found him, he was painting a thang-ka (the word used is thang-sku) of Tārā according to the system of Reverend Lord (Atiśa) for a young woman patron.[72]  In short, there were Tibetans practicing the art of painting in these centuries.  It is certain that they would have admired and patterned their own works after those by Indian artists, and they would have had ample opportunity to see them even without ever leaving Tibetan soil.  These more and less contemporary sources show that Tibetan artists were well appreciated for their work, that they were being patronized by other Tibetans.  The idea that high quality paintings must be of Indian origin is, at least, brought into serious question.

For those who are still not convinced that the Ford Tārā is a late twelfth-century masterpiece by a Tibetan artist, and for those who are more convinced by art historical than by literary historical arguments, I would advise them to place a reproduction of the Ford Tārā side by side with a thang-ka of Zhang Rin-po-che (also known as Bla-ma Zhang, or Zhang G.yu-brag-pa Brtson-’grus-grags-pa) from the Potala collection that has been reproduced a few times (the best being in Bod-kyi Thang-ka, plate 62).  It is one of a very few stitched or woven thang-kas known to survive in Tibet from the vicinity of the 12th century.[73]  Given the difficulties of the medium, it is nevertheless striking just how portrait-like Zhang Rin-po-che (note that, although clearly balding, he has black, not white, hair) appears in the central ground.[74]  (See Dunnell, this volume, Fig. X)  Agreeing to ignore for this purpose the central figures, which are after all quite different, the general composition (the positioning and proportions of the 26 [25?] niches surrounding the central figure), style and many of the details — including most of the aspects that have impressed previous writers with the ‘Indianness’ of the Ford Tārā — are strikingly similar.  The Zhang portrait-thang-ka ought to post-date his founding of Mtshal Gung-thang Monastery (located just up river from Lhasa) in 1175 (reason: that monastery’s special crow-headed protector, Bya-rog-gdong-can, is depicted in the lower right-hand corner[75]).  Just how much later it might be it is difficult to say with certainty, but the patron figure in the lower left-hand corner is so like the patron figures in other early thang-kas that, also taking stylistic features into account,[76] it seems it probably should not date from much later than the mid-thirteenth.  I am confident that those who perform the exercise will feel certain that the Zhang thang-ka was not lovingly woven by an Indian seamster for the Tibet export trade, and will find every reason to see the Tibetan hand in it regardless of any number of perceptible Indian influences.[77]  But then what are we to make of the statement in the preface of Bod-kyi Thang-ka that at first most fabric thang-kas were manufactured in China?  Was Sung China producing Pāla tapestries with portraits of Phyi-dar[78] Tibetans?  Even if it is common wisdom that earlier fabric thang-kas were all constructed in China (or possibly by Tanguts, or Chinese in Central Asia),[79] we are still brought to a cul-de-sac of confusion, practically an example of reductio ad absurdam (Did Indians also work from Tibetan-made exemplars in order to place such convincingly Tibetan and sometimes truly portrait-like patron figures in their paintings?  Were these Indian made paintings then taken to China to be made into tapestries before being delivered to their patrons in  Tibet?), and the only reasonable way to extricate ourselves is to go along with the solid information we do have, abandoning any previous suppositions about the Indian manufacture of the Ford Tārā until the time we may stumble upon a single convincing clue pointing in that direction.  No matter how many instances of paintings commissioned by Tibetans from Indian painters might be uncovered, they cannot be used as proof that this one was so commissioned.  Styles, which are practically by definition a result of artists copying what they most admire (the love and technique that always lay together so uncomfortably in the hearts and minds of artists everywhere), may not be taken as decisive evidence in the absence of other indications.[80]

In conclusion, we could say that Tibetan art historians might do well to consult more often with their more literarily and historically oriented fellow Tibetologists.  It may be true that the ‘connoisseurship’ demanded of the art historian would require a lifetime of devotion to visual interpretation alone.  (Being able to approximate the provenance of a painting based on a fold in a garment or the curl of a lotus petal is surely no mean accomplishment.)  Still, ‘art historians’ quite clearly do not fit into a single mold, demonstrating in their writings varying degrees of reliance on ‘outside’ sources of verification.  In fact, the distinctions between these two areas of expertise have been, and still are, blurred by a number of Tibet-specialists who fall on both sides of the (imaginary?) line (Doran 1998).  Cooperation and sharing, not disciplinary entrenchment and pride, will help us to move along the paths of Tibetological progress.  Historians themselves ought to stop privileging the canonical text over the artefact and accept these two different kinds of evidence in a more realistic way.  Here there is just a modest, hardly revolutionary, suggestion that the dedicatory, donatory, patronage, artist and dhāraṇī inscriptions found on artworks need to be read not only in light of each other, but also in light of contextually appropriate meanings, and with as much help as possible from readings in dkar-chags and other literature relevant to the arts, including scriptures, histories and consecration ritual texts.[81]  The same could be said for contextually appropriate readings of the paintings themselves.  The Ford Tārā is, for all we know, a very fine Tibetan painting, one with religious, historical, cultural and even, finally, stylistic depths that have yet to be very well appreciated.

Bibliography:

Allinger 1997:  Eva Allinger, ‘The Green Tārā in the Ford Collection: Some Iconographical Remarks.’  Contained in:  South Asian Archaeology 1995: Proceedings of the 13th Conference of the European Association of South Asian Archaeologists, Cambridge, 5-9 July, 1995, Oxford & IBH Publishing (N. Delhi), pp. 665-671.

Allinger 1998:  Eva Allinger, ‘The Green Tārā in the Ford-Collection: Some Stylistic Remarks.’  Contained in:  Deborah Klimburg-Salter & Eva Allinger, eds., The Inner Asian International Style 12th-14th Centuries, Verlag der Österreichischen Akademie der Wissenschaften (Vienna), pp. 107-119.

Allinger 1999:  Eva Allinger, ‘The Green Tara as Saviouress from the Eight Dangers in the Sumtsek at Alchi.’  Orientations, vol. 30, no. 1 (January), pp. 40-44.

Beer 1999:  Robert Beer, The Encyclopedia of Tibetan Symbols and Motifs, Serindia Publications (London).

Béguin 1993:  G[illes] B[éguin], ‘Nepal, Tibet.’  Arts Asiatiques, vol. 48, p. 141, and illustration on p. 93.

Bentor 1993:  Yael Bentor, ‘Tibetan Tourist Thangkas in the Kathmandu Valley.’  Annals of Tourism Research, vol. 20, pp. 107-137.

Bentor 1995:  Yael Bentor, ‘On the Indian Origins of the Tibetan Practice of Depositing Relics and Dhāraṇīs in Stūpas and Images.’  Journal of the American Oriental Society, vol. 15, no. 2, pp. 248-261.

Bentor 1995a:  Yael Bentor, ‘On the Symbolism of the Mirror in Indo-Tibetan Consecration Rituals.’  Journal of Indian Philosophy, vol. 23, pp. 57-71.

Bentor 1996:  Yael Bentor, Consecration of Images and Stūpas in Indo-Tibetan Tantric Buddhism, E.J. Brill (Leiden).

Blue Annals:  ’Gos Lo-tsā-ba Gzhon-nu-dpal (1392-1481), The Blue Annals, tr. by Dge-’dun-chos-’phel and George Roerich, Motilal Banarsidass (Delhi 1949/1976).

Bod-kyi Thang-ka:  Bod Rang-skyong Ljongs Rig-dngos Do-dam U-yon Lhan-khang, Bod-kyi Thang-ka, Rig-dngos Dpe-skrun-khang (Lhasa 1984?).

Brown, forthcoming:  Kathryn Selig Brown, ‘Representations of Handprints and Footprints on Thangkas.’  A paper given at the Eighth International Association of Tibetan Studies meeting, Bloomington, Indiana, 1997, forthcoming in the proceedings.

Bsod-nams-lha’i-dbang-po 1484/1977:  Lo-dgon-pa Bsod-nams-lha’i-dbang-po (1423-1496), Bka’-gdams Rin-po-che’i Chos-’byung Rnam-thar Nyin-mor Byed-pa’i ’Od Stong, contained in: Two Histories of the Bka’-gdams-pa Tradition from the Library of Burmiok Athing, Gonpo Tseten (Gangtok 1977), pp. 207-393.

Bsod-nams-'od-zer 1997:  Bsod-nams-'od-zer, Grub-chen U-rgyan-pa'i Rnam-thar, Bod-ljongs Bod-yig Dpe-rnying Dpe-skrun-khang (Lhasa).  Gangs-can Rig-mdzod series no. 32.

Bsod-nams-rtse-mo 1968:  Bsod-nams-rtse-mo (1142-1182), Chos-la ’Jug-pa’i Sgo zhes bya-ba’i Bstan-bcos, contained in:  Bsod-nams-rtse-mo’i Bka’-’bum as contained in:  Sa-skya-pa’i Bka’-’bum, The Toyo Bunko (Tokyo), vol. 2, pp. 318-345.

Bsod-nams-rtse-mo 1968a:  Bsod-nams-rtse-mo (1142-1182), Mi-g.yo-ba’i Chos-skor Rgya-gzhung Bstod-pa dang bcas-pa, contained in: Bsod-nams-rtse-mo’i Bka’-’bum as contained in:  Sa-skya-pa’i Bka’-’bum, The Toyo Bunko (Tokyo), vol. 2, pp. 449-457.

Bu-ston 1971:  Bu-ston Rin-chen-grub 1290-1364), Bla-ma Dam-pa-rnams-kyis Rjes-su Bzung-ba'i Tshul Bka'-drin Rjes-su Dran-par Byed-pa, contained in:The Collected Works of Bu-ston, ed. by Lokesh Chandra, International Academy of Indian Culture (New Delhi), vol. 26, pp. 1-142.

Carnahan 1995:  Sumner Carnahan and Lama Kunga Rinpoche, trs., In the Presence of My Enemies:  Memoirs of Tibetan Nobleman Tsipon Shuguba, Clear Light Publishers (Santa Fe).

Clifford 1984:  Terry Clifford, Tibetan Buddhist Medicine and Psychiatry: The Diamond Healing, Samuel Weiser (York Beach, Maine).

Dag-yig 1979:  n.a., Dag-yig Gsar-bsgrigs, Mtsho-sngon Mi-rigs Dpe-skrun-khang (Xining).

Dalai Lama V 1970-71:  Dalai Lama V Ngag-dbang-blo-bzang-rgya-mtsho, Thob-yig Gangga’i Chu Rgyun: The Gsan‑yig of the Fifth Dalai Lama, Nechung and Lhakhar (Delhi), in 4 volumes.

Das 1973:  Sarat Chandra Das, A Tibetan-English Dictionary with Sanskrit Synonyms, Motilal Banarsidass (Delhi), reprint of the 1902 Calcutta edition.

Dbang-’dus 1983:  Dbang-’dus, Gso-ba Rig-pa’i Tshig-mdzod G.yu-thog Dgongs-rgyan, Mi-rigs Dpe-skrun-khang (Beijing).

Decleer 1996:  Hubert Decleer, ‘Master Atiśa in Nepal: The Tham Bahīl and Five Stūpas’ Foundations accordimg to the ’Brom-ston Itinerary.’  Journal of the Nepal Research Centre, vol. 10, pp. 27-54.

Decleer 1997:  Hubert Decleer, ‘Atiśa’s Journey to Tibet.’  Contained in:  Donald S. Lopez, Jr. , ed., Religions of Tibet in Practice, Princeton University Press (Princeton), pp. 157-177.

Dhammapada, 1988:  The Venerable Balangoda Ananda Maitreya, tr., and Rose Kramer, reviser, The Dhammapada: The Path of Truth, Lotsawa (Novato, California).

Dkon-mchog-bstan-’dzin 1994:  Dkon-mchog-bstan-’dzin (b. 1949), Bzo-gnas Skra Rtse’i Chu Thigs, Krung-go’i Bod-kyi Shes-rig Dpe-skrun-khang (Beijing).

Doran 1998:  Valerie C. Doran, ‘Recent Developments in the Tibetan Art Field.’  Orientations, vol. 29, no. 9 (October), pp. 99-107.

Dpa’-bo 1564/1986:  Dpa’-bo II Gtsug-lag-phreng-ba (1504-1566), Chos-’byung Mkhas-pa’i Dga’-ston, ed. by Rdo-rje-rgyal-po, Mi-rigs Dpe-skrun-khang (Beijing 1986), in 2 volumes.

Dza-ya 1981:  Dza-ya Pandi-ta Blo-bzang-'phrin-las (1642-1708), Collected Works of Jaya-Pandita Blo-bzan-hphrin-las, International Academy of Indian Culture (New Delhi), in 4 volumes.  Composed in 1702.

Eimer 1977:  Helmut Eimer, Berichte über das Leben des Atiśa (Dīpamkaraśrījñāna): Eine Untersuchung der Quellen, Otto Harrassowitz (Wiesbaden).

Eimer 1979:  Helmut Eimer, Rnam thar rgyas pa: Materialien zu einer Biographie des Atiśa (Dīpamkaraśrījñāna), Asiatische Forschungen vol. 67 (Wiesbaden).

Eimer 1989: Helmut Eimer, ‘Nag tsho Tshul khrims rgyal ba’s Bstod pa brgyad cu pa in Its Extant Version.’  Bulletin of Tibetology (Gangtok, Sikkim), new series no. 1, pp. 21-38.

Eimer 1997: Helmut Eimer, ‘Hymns and Stanzas Praising Dīpaṃkaraśrījñāna.’  Contained in: Kameshwar Nath Mishra, ed., Glimpses of the Sanskrit Buddhist Literature, Central Institute of Higher Tibetan Studies (Sarnath), pp. 9-32.

Erickson 1996/7:  Susan N. Erickson, ‘The Freer Gallery of Art Boshanlu: Answers to A.G. Wenley’s Questions.’  Oriental Art, vol. 42, no. 4 (Winter), pp. 27-38.

First Dalai Lama 1980:  Gyalwa Gendun Drub [Rgyal-ba Dge-’dun-grub], the First Dalai Lama, Six Texts Related to the Tara Tantra, tr. by Glenn H. Mullin, Doboom Tulku, and Chomdze Tashi Wangyal, Tibet House (New Delhi).

Goldstein 1989:  Melvyn C. Goldstein, A History of Modern Tibet, 1913-1951, University of California Press (Berkeley).

Gtsang Mkhan-chen 1985:  Gtsang Mkhan-chen ’Jam-dbyangs-dpal-ldan-rgya-mtsho (1610-1684), Zab Khrid Kun-las Btus-pa’i Man-ngag Byin-rlabs-kyi Chu-gter, National Library of Bhutan (Thimphu).

Gyatso 1988:  Geshe Kelsang Gyatso, Universal Compassion: A Commentary to Bodhisattva Chekhawa’s Training the Mind in Seven Points, Tharpa Publications (London).

Hahn 1974:  Michael Hahn, Lehrbuch der klassischen tibetischen Schriftsprache, Michael Hahn (Bonn).

Heller 1997:  Amy Heller, ‘A Set of Thirteenth Century Tsakali,’ Orientations, vol. 28, no. 10 (November), pp. 48-52.

Heller 1999:  Amy Heller, Tibetan Art: Tracing the Development of Spiritual Ideals and Art in Tibet 600-2000 A.D., Jaca Book (Milan).

Henss 1997:  Michael Henss, ‘The Woven Image: Chinese Textile Thangkas of the Yuan and Early Ming Dynasties.’  Orientations, vol. 28, no. 10 (November), pp. 26-39.

Henss 1997a:  Michael Henss, ‘The Eleventh-Century Murals of Drathang Gonpa.’  Contained in:  Jane Casey Singer & Philip Denwood, eds., Tibetan Art: Towards a Definition of Style, Lawrence King Publishing (London), pp. 160-169.

Jackson 1993:  David Jackson, ‘Apropos a Recent Tibetan Art Catalogue.’  Wiener Zeitschrift für die Kunde des Morgenlandes, vol. 37, pp. 109-130.

Jackson 1996:  David P. Jackson, A History of Tibetan Painting, Verlag der Österreichischen Akademie der Wissenschaften (Vienna).

Jackson & Jackson 1984:  David P. Jackson & Janice A. Jackson, Tibetan Thangka Painting: Methods and Materials, Shambhala (Boulder).

Jäschke 1972:  Heinrich August Jäschke, A Tibetan-English Dictionary, Routledge & Kegan Paul (London), reprint of 1881 edition.

’Jig-rten-mgon-po 1969-71:  ’Jig-rten-mgon-po (1143-1217),The Collected Writings (Gsung-’bum) of ’Bri-gung Chos-rje ’Jig-rten-mgon-po Rin-chen-dpal, Khangsar Tulku (New Delhi), in 5 volumes.

Kapstein 1995:  Matthew Kapstein, ‘Weaving the World: The Ritual Art of the Paṭa in Pāla Buddhism and Its Legacy in Tibet.’  History of Religions, vol. 34, no. 3, pp. 241-262.

Klimburg-Salter 1998:  Deborah E. Klimburg-Salter et al., Tabo: A Lamp for the Kingdom.  Early Indo-Tibetan Buddhist Art of the Western Himalaya, Thames & Hudson (New York & London).

Klimkeit 1990:  Hans-Joachim Klimkeit, ‘The Donor at Turfan.’  Silk Road Art and Archaeology, vol. 1, pp. 177-201.

Ko-zhul 1992:  Ko-zhul Grags-pa-’byung-gnas & Rgyal-ba-blo-bzang-mkhas-grub, Gangs-can Mkhas-grub Rim-byon Ming-mdzod, Kan-su’u Mi-rigs Dpe-skrun-khang (Lanzhou).

Kossak 1998:  Steven Kossak, 'Early Paintings from Central Tibet in The Metropolitan Museum of Art.'  Orientations, vol. 29, no. 9 (October), pp. 50-64.

Kossak & Singer 1998:  Steven M. Kossak & Jane Casey Singer, Sacred Visions: Early Paintings from Central Tibet, The Metropolitan Museum of Art (New York).

Lalou 1930:  Marcelle Lalou, Iconographie des étoffes peintes (paṭa) dans le Mañjuśrīmūlakalpa, Librarie Orientaliste Paul Geuthner (Paris).

Lalou 1935:  Marcelle Lalou, ‘Trois aspects de la peinture bouddhique.’  Annuaire de l’Institut de Philologie et d’Histoire Orientales (Bruxelles), vol. 3, pp. 245-261.

Las-chen 1494/1972:  Las-chen Kun-dga’-rgyal-mtshan, Bka’-gdams-kyi Rnam-par Thar-pa Bka’-gdams Chos-’byung Gsal-ba’i Sgron-me, B. Jamyang Norbu (New Delhi 1972), in 2 volumes.

Lhun-grub-chos-’phel 1994:  Lhun-grub-chos-’phel, Rwa-sgreng Dgon-pa’i Dkar-chag, Si-khron Mi-rigs Dpe-skrun-khang (Chengdu).

Linrothe 1998:  Rob Linrothe, ‘Xia Renzong and the Patronage of Tangut Buddhist Art: The Stūpa and Ushnīshavijayā Cult.’  Journal of Sung-Yuan Studies, vol. 28, pp. 91-121.

Mang-thos 1988:  Mang-thos Klu-sgrub-rgya-mtsho (1523-1596), Bstan-rtsis Gsal-ba’i Nyin-byed, Bod-ljongs Mi-dmangs Dpe-skrun-khang (Lhasa), composed in 1566-1587.

Martin 1994:  ‘Pearls from Bones: Relics, Chortens, Tertons and the Signs of Saintly Death in Tibet.’  Numen, vol. 41, pp. 273-324.

Martin 1996:  Dan Martin, ‘Tables of Contents (Dkar-chag).’  Contained in:  J. Cabezón & R. Jackson, eds., Tibetan Literature: Studies in Genre, Snow Lion (Ithaca), pp. 500-514.

McGuckin 1996:  Eric McGuckin, ‘Thangkas and Tourism in Dharamsala: Preservation through Change.’  Tibet Journal (Dharamsala), vol. 21, no. 2 (Summer), pp. 31-52.

Mon-ban 1985:  Mon-ban Dpal-ldan-bzang-po, Mon-ban Dpal-ldan-bzang-po Bdag-gi Thob-yig Thos-pa Rgya-mtsho: Record of Teachings Received (gsan-yig) of Gnyos-ston Dpal-ldan-bzang-po (1447-1507),  Dorji Namgyal (Thimphu).

Nyang-ral 1988:  Nyang[-ral] Nyi-ma-’od-zer (1124-1192?), Chos-’byung Me-tog Snying-po Sbrang-rtsi’i Bcud, Bod-ljongs Mi-dmangs Dpe-skrun-khang (Lhasa).  Gangs-can Rig-mdzod series, no. 5.

Pabongka 1990:  Pabongka Rinpoche Jampa Tenzin Trinley Gyatso [Pha-bong-kha-pa Byams-pa-bstan-’dzin-phrin-las-rgya-mtsho, 1878-1941], Liberation in Our Hands, Part One: The Preliminaries, tr. by Sera Mey Geshe Lobsang Tharchin and Artemus B. Engle, Mahayana Sutra and Tantra Press (Howell, New Jersey).

Pal 1983:  Pratapaditya Pal, Art of Tibet, University of California Press (Berkeley).

Pal 1984:  Pratapaditya Pal, Tibetan Paintings: A Study of Tibetan Thankas Eleventh to Nineteenth Centuries, Sotheby Publications (London).

Pal 1997:  Pratapaditya Pal, ‘Preface.’  Contained in:  Jane Casey Singer & Philip Denwood, eds., Tibetan Art: Towards a Definition of Style, Lawrence King Publishing (London), pp. 8-13.

Phag-mo-gru-pa 1507:  Phag-mo-gru-pa Rdo-rje-rgyal-po (1110-1170), Bka’-’bum [‘Collected Works’], a photocopied version of a ‘golden manuscript’ (written in gold ink) in four volumes, privately acquired.  This manuscript was constructed under the patronage of ’Bri-gung-pa Kun-dga’-rin-chen (1475-1527).

Pho-brang Po-ta-la:  Bod Rang-skyong Ljongs Rig-gnas Dngos-rdzas Do-dam U-yon Lhan-khang, Pho-brang Po-ta-la, Rig-gnas Dngos-rdzas Dpe-skrun-khang (Lhasa 1988).

Ramble 1996:  Charles Ramble, ‘Patterns of Places.’  Contained in:  Anne-Marie Blondeau & Ernst Steinkellner, Reflections of the Moutain: Essays on the History and Social Meaning of the Mountain Cult in Tibet and the Himalaya, Verlag der Österreichischen Akademie der Wissenschaften (Vienna), pp. 141-153.

Reynolds et al. 1986:  Valrae Reynolds, Amy Heller & Janet Gyatso, Catalogue of the Newark Museum Tibetan Collection.  Volume III: Sculpture and Painting, The Newark Museum (Newark).

Reynolds 1996:  Valrae Reynolds, ‘Fabric Images and their Special Role in Tibet.’  Contained in: P. Pal, ed., On the Path to Void: Buddhist Art of the Tibetan Realm, Marg Publications (Bombay), pp. 244-257.

Reynolds 1997:  ‘Luxury Textiles in Tibet.’  Contained in:  Jane Casey Singer & Philip Denwood, eds., Tibetan Art: Towards a Definition of Style, Lawrence King Publishing (London), pp. 118-131.

Rhie & Thurman 1991:  Marylin M. Rhie & Robert A.F. Thurman, Wisdom and Compassion: The Sacred Art of Tibet, Harry N. Abrams (New York).

Rhie & Thurman 1999:  Marylin M. Rhie & Robert A.F. Thurman, Worlds of Transformation: Tibetan Art of Wisdom and Compassion, Tibet House (New York).

Richardson 1985:  Hugh E. Richardson, A Corpus of Early Tibetan Inscriptions, Royal Asiatic Society, Stephen Austin & Sons (Hertford).

Rossabi 1998:  Morris Rossabi, ‘Behind the Silk Screen: Movements of Weavers in Asia, Seventh to Fourteenth Centuries.’  Orientations, vol. 29, no. 3 (March), pp. 84-89.

Rta-tshag 1994:  Rta-tshag Tshe-dbang-rgyal, Lho-rong Chos-’byung, Bod-ljongs Bod-yig Dpe-rnying Dpe-skrun-khang (Lhasa), composed during the years 1446 to 1451.

Sba-bzhed:  Sba Gsal-snang, Sba-bzhed (= Sba-bzhed ces bya-ba-las Sba Gsal-snang-gi Bzhed-pa Bzhugs), ed. by Mgon-po-rgyal-mtshan, Mi-rigs Dpe-skrun-khang (Beijing 1980/1982).

Singer 1986:  Jane Casey Singer, ‘An Early Painting from Tibet.’  Orientations, vol. 17 (July), pp. 21-25.

Singer 1994:  Jane Casey Singer, ‘Painting in Central Tibet, ca. 950-1400.’  Artibus Asiae, vol. 54, nos. 1-2, pp. 87-136.

Singer 1996:  Jane Casey Singer, ‘Early Thankas: Eleventh-Thirteenth Centuries.’  Contained in:  P. Pal, ed., On the Path to Void: Buddhist Art of the Tibetan Realm, Marg Publications (Bombay), pp. 180-195.

Singer 1997:  Jane Casey Singer, ‘Taklung Painting.’  Contained in:  Jane Casey Singer & Philip Denwood, eds., Tibetan Art: Towards a Definition of Style, Lawrence King Publishing (London), pp. 52-67, 293-295.

Singer 1998:  Jane Casey Singer, ‘An Early Tibetan Painting Revisited: The Ashtamahabhaya Tara in the Ford Collection.’  Orientations, vol. 29, no. 9 (October), pp. 65-73.

Srivastava 1989:  Balram Srivastava, ‘Buddhist Goddesses in Orissan Imagery.’  Contained in:  Buddhist Iconography [Sambhota Series II], Tibet House, New Delhi, pp. 129-137.

Stag-lung 1992:  Stag-lung Ngag-dbang-rnam-rgyal, Brgyud-pa Yid-bzhin Nor-bu’i Rtogs-pa Brjod-pa Ngo-mtshar Rgya-mtsho (= Stag-lung Chos-’byung), Bod-ljongs Bod-yig Dpe-rnying Dpe-skrun-khang (Lhasa).  Composed in 1609.

Stearns 1999:  Cyrus Stearns, The Buddha from Dolpo: A Study of the Life and Thought of the Tibetan Master Dolpopa Sherab Gyaltsen, State University of New York Press (Albany).

Stoddard 1994:  Heather Stoddard, ‘Restoration in the Lhasa Tsuglagkhang and the Fate of Its Early Wall Paintings.’  Orientations, vol. 25, no. 6 (June), pp. 169-173 (also in reprint volume).

Surkhang 1983:  Surkhang Wangchen Gelek [Zur-khang Dbang-chen-dge-legs], ‘Tibet, the Critical Years (Part 3): The Regent Reting Rinpoche.’  Tibet Journal (Dharamsala), vol. 8, no. 2 (Summer), pp. 33-39.

Tanaka 1994:  Yuko Tanaka, ‘A Note on the History, Materials and Techniques of Tibetan Appliqué Thangkas.’  Contained in:  Per Kværne, ed., Tibetan Studies, The Institute for Comparative Research in Human Culture (Oslo), vol. 2, pp. 873-876.

Tanaka 1997:  Yuko Tanaka, ‘Art as Spiritual Practice: A Study of Three Mongolian Works on the Practice and Sponsorship of Buddhist Thangka Painting.’  Contained in:  Helmut Krasser, et al., eds., Tibetan Studies, Verlag der Österreichischen Akademie der Wissenschaften (Vienna), vol. 2, pp. 947-953.

Templeman 1989:  David Templeman, Tāranātha's Life of Kṛṣṇācārya/Kāṇha, Library of Tibetan Works & Archives (Dharamsala).

Tseten Zhabdrung 1986:  Tseten Zhabdrung [Tshe-brtan Zhabs-drung, 1910-1985], ‘Research on the Nomenclature of the Buddhist Schools in Tibet.’  Translated by Tenzin Dorjee. Tibet Journal, vol. 11, no. 3 (Autumn), pp. 48-56.

Uray 1984:  Géza Uray, ‘The Earliest Evidence of the Use of the Chinese Sexagenary Cycle in Tibetan.’  Contained in:  Louis Ligeti, ed., Tibetan and Buddhist Studies Commemorating the 200th Anniversary of the Birth of Alexander Csoma de Körös, Akadémiai Kiadó (Budapest), pp. 341-360.

van der Kuijp 1986:  Leonard W.J. van der Kuijp, ‘Review’ of Per Kværne, Tibet Bon Religion contained in: Acta Orientalia, vol. 47, pp. 202-208.

Vitali 1990:  Roberto Vitali, Early Temples of Central Tibet, Serindia Publications (London).

Vitali 1997:  Roberto Vitali, ‘Nomads of Byang and Mnga’-ris-smad.’  Contained in:  Helmut Krasser, et al., eds., Tibetan Studies, Verlag der Österreichischen Akademie der Wissenschaften (Vienna), vol. 2, pp. 1023-1036.

Vitali 1999:  Roberto Vitali, Records of Tho.ling: A Literary and Visual Reconstruction of the 'Mother' Monastery in Gu.ge, High Asia (McLeod Ganj).

Waddell 1895/1972:  L. Austine Waddell, Tibetan Buddhism with its Mystic Cults, Symbolism and Mythology, Dover Publications (New York 1972), reprint of work originally titled, The Buddhism of Tibet, or Lamaism, W.H. Allen (London 1895).

Whitfield & Farrer 1990:  Roderick Whitfield & Anne Farrer, Caves of the Thousand Buddhas: Chinese Art from the Silk Route, George Braziller (New York).

Wiercimok 1990:  Edith Wiercimok, ‘The Donor Figure in the Buddhist Painting of Dunhuang.’  Silk Road Art and Archaeology, vol. 1, pp. 203-226.

Willson 1996:  Martin Willson, In Praise of Tara: Songs to the Saviouress, Wisdom Publications (Boston).

Yamaguchi 1997:  Zuiho Yamaguchi, ‘The Usage of Yan-chad/Man-chad: The Subjective and Objective Aspects of Tibetan Predicates’ [in Japanese],  Journal of the Naritasan Institute for Buddhist Studies, vol. 20, pp. 1-26.

Yar-lung Jo-bo 1988:  Yar-lung Jo-bo Shākya-rin-chen-sde, Yar-lung Chos-’byung, Si-khron Mi-rigs Dpe-skrun-khang, Chengdu, ed. by Dbyangs-can.  Composed in 1376.

Yongs-’dzin 1990:  Yongs-’dzin Ye-shes-rgyal-mtshan (1713-1793), Lam-rim Bla-ma Brgyud-pa’i Rnam-thar, Bod-ljongs Mi-dmangs Dpe-skrun-khang (Lhasa).  Composed in 1787.

Yule & Burnell 1886/1996:  Henry Yule & A.C. Burnell, Hobson-Jobson: The Anglo-Indian Dictionary, Wordsworth Reference (Hertfordshire).

Zhang, Samdo A:  Zhang G.yu-brag-pa Brtson-’grus-grags-pa (1123-1193) Bka’-’bum [‘Collected Works’].  Microfilm of a six-volume manuscript kept at Samdo Monastery in Nepal.  This nearly complete six-volume set is found in the Nepalese National Archives microfilm collection from running number L4686 through L4697, which is equivalent to the reel numbers L448/2 through L450/4.

Zongtse & Dietz 1990:  Champa Thubten Zontse & Siglinde Dietz, eds., Udānavarga, Vandenhoeck & Ruprecht (Göttingen).

List of Illustrations with long captions.

Figure 1.  Acacia Forest Tārā, John Gilmore Ford Collection.  Photo courtesy of John Gilmore Ford.

Figure 2.  Acacia Forest Tārā, detail of the patron figure who ought, in this case, to be Bya Brtson-’grus-’od.  Photo courtesy of John Gilmore Ford.

Figure 3.  The two black ink inscriptions in the upper register of the back of the Ford Tārā.  Photos courtesy of John Gilmore Ford.

Figure 4.  The red inscriptions, in the shape of a chorten, in the center of the back of the Ford Tārā.  Photo courtesy of John Gilmore Ford.

Figure 5.  A recent woodcut of Bya ’Chad-kha-ba Ye shes-rdo-rje (1101-1175 ce).  It is reproduced from the famous illustrated Lhasa print of the Eight Thousand Prajñāpāramitā scripture done in 1954 by the master woodblock carver Ding-ri-ba Chos-kyi-rgyal-mtshan (1897-1959), containing over a thousand woodcut icons of deities and saints.  This is a retraced version of the original woodcut located at folio 81 recto, lefthand side.  The picture caption reads byang-chub thugs-kyi bka’-babs ’Chad-kha-pa, ‘[Who received the] Word-transmission of Awakened Heart, ’Chad-kha-ba.’  Note that, apart from the facial features, beard and hairline, the iconography is virtually identical to that of Spyil-phu-ba; both hold manuscript leaves in their left hands, with right hands in the ‘earth-touching’ gesture.  The wrapped volume depicted on his right may be a reference to ’Chad-kha-ba’s writings, including the Seven Point Mind Training.  Clearly based on this woodblock are the painting by the contemporary artist Jamyang on the cover of Gyatso (1988), and the drawing by Andy Weber in Gyatso (1988: 92).  That a portrait of Mchad-kha-ba had been made during his own lifetime is proven by the statement “This likeness of ’Chad-kha-ba; [in it] abides the consecration he himself performed” (Lhun-grub-chos-’phel 1994: 131).

Figure 6.  Retracing of a woodcut of Se Spyil-phu-ba Chos-kyi-rgyal-mtshan (1121-1189 ce) from the same work as Figure 5 (folio 82 recto, left side).  He (and/or his teacher Bya Mchad-kha-ba) may have been the first consecrator of the Ford Tārā thang-ka.  The picture caption reads:  lung rtogs mnga’-bdag Byang-sems Sbyil-phu-ba, ‘Master of scriptures and realization, the Bodhisattva Spyil-phu-ba.’  Compare the line drawing by Andy Weber, clearly based on this woodcut, in Gyatso (1988: 104).

* Dedicated to Yael Bentor, on her 15th wedding anniversary, and to my parents, Luren J. and Ardath R. Martin, on their 50th.  I would like to thank Lilian Handlin, Amy Heller, Louis Herlands and Rob Linrothe for directly and indirectly encouraging and inspiring me to write this paper, with gratitude also to Valrae Reynolds from whom I learned a great deal more than I thought I needed to know about the various types of fabric thang-kas.  I am thankful for the generosity of John Gilmore Ford in supplying clear photographs of the inscriptions on the back of the famous Tārā thang-ka that was in his collection.  Helmut Eimer kindly and patiently responded to my bothersome inquiries, in the process eliminating several substantial errors (I hope none remain, but if they do, they are mine, not his), while David Germano supplied helpful criticisms and rare bibliographical items.  Jonathan Silk took the time to answer almost all of my questions, including some that I did not know enough to ask, about the ‘patience creed’ in which, it would appear, he not only believes but practices.  Thanks also to Kathryn Selig Brown, David Jackson and Christian Luczanits for comments on various draft versions.  Comments by Jane Casey Singer, in particular, led to a number of additions and changes.  I am, naturally, the one who ought to be held responsible, ultimately, for whatever is good or bad in the form and substance of this essay.  There were a few relevant published pieces that could not be consulted in time to include in this article.  These include an article by Jane Casey Singer entitled ‘Early Portrait Painting in Tibet’ contained in a recent issue of Asian Arts, and John C. & Susan L. Huntington’s book Leaves from the Bodhi Tree. 

[1]By ‘early Tibetan art’ we here intend a period roughly corresponding to the eleventh through early thirteenth centuries, a period framed by the revival of monasticism at the end of the tenth century, and the political reunification of Tibet under Mongol domination in 1260.  This has sometimes been called the Kadampa (Bka’-gdams-pa) period, although other sects, particularly the Bka’-brgyud-pa, Zhi-byed-pa and Sa-skya-pa, took root and began to flourish during the same time period, while the old Rnying-ma-pa and Bon-po lineages were all along continuing and developing.  These other sects were also art patrons, and should not to be slighted.  (Similar sentiments are expressed in Singer 1994: 89, n. 7 and Pal 1997: 12-13.)  Tibetan art prior to the last decades of the tenth century should then be referred to as ‘old Tibetan art.’  Although little will be said about old Tibetan art here, it is crucial to recognize its existence.  Furthermore, it could be noted that the emphasis here is on cloth images (thang-kas) whether painted, woven or stitched, with only a few incidental references to mural paintings and to paintings on paper.

[2]Kossak in Kossak & Singer (1998: 27-28). 

[3]Following are some samples from the conclusions of Singer (1998: 73), which ought to be read in their entirety: 

“Might the artist have been eastern Indian?  In the author’s view, yes.  The style is not simply inspired by eastern Indian aesthetic and iconographic canons, but … seems to epitomize them in a manner that leads one to conclude that only someone trained and immersed in an eastern Indian painting tradition could have produced it.

“If the Ford Tara painting were created for a Tibetan patron by an eastern Indian artist, one wonders whether others of the five hundred or so early paintings to have come out of Tibet in recent decades might also, in some sense, be eastern Indian. …” 

The scholarly genealogy of the belief that the Ford Tārā was or might have been painted by an Indian artist may be traced back to Pratapaditya Pal (references will be given in a footnote that follows).  Jane Casey Singer is, by the way, the single person most responsible for opening the paths of research into early Tibetan art.  I hope I will be forgiven if I do not shoulder the responsibility of portraying the full wealth of her expert contributions to this (largely thanks to her) growing field and instead concentrate on a few problematic aspects.

[4]On the importance of publishing photographs or at least supplying accurate transcriptions of thang-ka inscriptions, see especially Jackson (1993: 118-9, 126).  Note also the statement in Jackson (1996: 38):

It is true that the interpretation of such inscriptions can entail difficulties for the researcher.  But the main problem, it seems to me, is simply that not enough of them have yet been gathered and carefully investigated.  One thing is sure: since inscriptions are the main means for identifying the individual patrons, lamas and painters involved in the production and completion of a painting, no preliminary difficulties entailed in their study and interpretation will diminish in the long run their importance for Tibetan art history.

[5]Since this story has been very well told in English by Decleer (1996, 1997) and Pabongka (1990: 46-57), as well as H. Eimer’s studies to be cited presently, we will not go into it here.

[6]See also Pabongka (1990: 57), which mentions the opinion of Tsong-kha-pa that two of these years were spent serving Atiśa in India, and seventeen in Tibet.  According to this, since Atiśa arrived in Tibet in 1042, Nag-tsho would have served him until 1059, i.e., five years after Atiśa’s death.  The ultimate source of the statement that Nag-tsho accompanied and relied on (bsten-cing ’grogs) Atiśa for 19 years is in the Hymn of Praise in Eighty Verses by Nag-tsho himself (Eimer 1989: 36, lines 333-336; translated in Eimer 1997: 19).  If Nag-tsho met Atiśa as a younger man during his prior travels in India, there is of course no chronological problem at all.  The problem would lie rather in Las-chen’s history.  Dza-ya (1981: II 550-552) says that Atiśa lived 13 years in Tibet, that Nag-tsho served him 19 years, and that Nag-tsho was in his 44th year when Atiśa died.

[7]Compare the transcription in Eimer (1979: II 366, para. 424), and the German summary in Eimer (1979: I 297, para. 424).  This passage is often cited briefly or alluded to in other historical works.  It is quoted in its entirety in Dza-ya (1981: II 553).

[8]If a slight emendation to the text were permitted, we could translate:  “Reverend himself painted an image, about the size of a thumb, on acacia wood, and after placing it in a reliquary, attached it to the shoulder (or upper arm [of Nag-tsho]).”  Knowing the close association of Tārā with acacia wood, we could then assume that the image he painted was one of Tārā.  Atiśa seems to have had a certain amount of artistic talent.  In any case, he would often make sketches (skya-ris) and have others complete the work (for examples of this, see Lhun-grub-chos-’phel 1994: 124, 131; in the first instance, he sketched using his own nasal blood).  For a brief reference to a wall painting made of Atiśa during his own lifetime at the Ke-ru temple, see Blue Annals (257).  There is also a brief reference to Nag-tsho’s painting of Atiśa in Blue Annals (260).

[9]Helmut Eimer has indicated to me that Nag-tsho was away in Nepal (Atiśa had sent him to meet the Pundit Jñānakāra) when Atiśa passed away.  Before Nag-tsho’s departure, Atiśa, prescient about his own approaching death, advised Nag-tsho to construct his likeness, and promised he would then come down personally from Tuṣita Heaven.  The sources are to be located in Eimer’s (1979: I 282-3; II 329-334) edition of the Rnam-thar Rgyas-pa at paras. 393-6 and 424.

[10]A more abbreviated version of this story is located in Dpa’-bo (1564/1986: 706).  Despite some effort, I could learn nothing about the subsequent fate of this particular thang-ka, except a 13th-century testimony (mentioned later on) that it was preserved at Yang-thog.  It is interesting to note that, in the year 1926, the Arhat Temple at Tho-ling in western Tibet came into possession of a sku-thang depicting the triad of Atiśa, Rngog and 'Brom-ston-pa (Vitali 1999: 150, 176), although nothing is said about its previous owner.

[11]This passage is based on Bka’-gdams Glegs-bam, Pha Chos section kha, fol. 91, of the Lhasa woodblock print in the biography of Atiśa known briefly as the Rnam-thar Yongs-grags, but compared with a more recent printing with a few preferable readings.  The text given here is very close, but not identical, to the edited text from the Rnam-thar Rgyas-pa (probably older than the work by Mchims Nam-mkha’-grags) supplied in Eimer (1979: II 365).

[12]Yang-thog is a place mainly known for its association with Nag-tsho and, or so it would seem, forgotten subsequently, although ‘people of Yang-thog’ [Yang-thog-pa] appear in a context which would at least indicate their close proximity to Nag-tsho’s native area, Gung-thang; see Vitali (1997: 1027, n. 14 and 1029, n. 17).

[13]This text is from the Bka’-gdams Glegs-bam, Pha Chos section kha, fol. 121, of the Lhasa woodblock print, in the biography attributed to ’Brom-ston known as the Rnam-thar Lam-yig, or simply as the Lam-yig; compare the translation of this passage in Decleer (1996: 48).

[14]The correct spelling of the Sanskrit word for ordained layman is upāsaka, feminine upāsikā. Tibetan sources tend to use both u-pa-sa-ka and u-pa-si-ka indifferently, both with and without the length mark on the second syllable.  Here it might seem to be a name by which Atiśa liked to address his disciple ’Brom-ston, since the latter never took the vows of monkhood.  But among Atiśa’s Indian attendants during his travels were one Upāsika Sa’i-sang-ga and one Upāsika Sa’i-snying-po.  According to Decleer (1996: 45, 48) the one in question here should be Sa’i-sang-ga, and the thang-kas described here were presented to the Kathmandu monastery of Tham Vihāra at its consecration (although as Decleer notes this episode of Atiśa’s return to Nepal from Western Tibet nine months after his departure might be considered doubtful since it is lacking in other early sources).  If only because of the different historical scenarios, the giant thang-ka made by Krishna should not be the one mentioned here.  But still, there is room for doubt, in part, too, because the Lam-yig is generally confused in its chronology of events.  Helmut Eimer has written to me that he believes this Lam-yig passage may simply contain a further account of the same incident of Nag-tsho’s thang-ka making as told in a quite different way in our previous citations.

[15]His translations were many, although one must be careful not to confuse him with still another Nag-tsho Translator with the given name Shes-rab-rgyal-mtshan who also seems to have lived in the eleventh century.  Our Nag-tsho Tshul-khrims-rgyal-ba translated a Vajrapāṇi text and about a dozen others, mainly Highest Yoga Tantra commentaries, together with an Indian master [mkhan-po] named Nag-po-pa, which would be the Tibetan corresponding to the Indian name Krishna (who cannot be the more famous Indian tantric writer Nag-po-pa or Nag-po-spyod-pa [on whom, see Templeman 1989], since the latter must have lived some centuries earlier) and it seems very likely that he could be identical with our painter.  But then the name Krishna is and has long been a very common one in India.

[16]Apart from Eimer (1989 and 1997), a few other articles devoted to these verses by Helmut Eimer have been published in India, but since they are not at the moment available to me, they have not been listed here.  I would like to thank him for sending some of these articles to me. There are important discussions of this work in Eimer (1977: 138-145, 305-325).  Eimer (1977: 4, n. 15) has also looked into several existing thang-kas which depict scenes from Atiśa’s life. One might note still another thang-ka of Atiśa and two of his disciples based on a vision beheld by Sangs-rgyas-rin-po-che, mentioned in Las-chen (1494/1972: II 356).  A surviving thang-ka of Atiśa (Singer 1994: 112-113; Kossak 1998: fig. 2) bears the inscription “Atiśa.  Offered [as a gift] by Ston-pa Dar-blo to Rin-chen-sgang-pa.  [In it] many consecrations abide.  Divinity of the residence quarters.”  (A-ti-sha / Rin-chen-sgang-pa-la Ston-pa Dar-blos phul / rab-gnas mang-du bzhugs / gzims-kyi lha //; based on Singer 1994: 113, n. 79).  A Dge-ba’i-bshes-gnyen Ri[n-ch]en-sgang-pa and a Stod-lung-pa Dar-blo are both listed among the disciples of Sne’u-zur-pa (Las-chen 1495/1972: I 361; Blue Annals, p. 314).  Sne’u-zur-pa Ye-shes-’bar’s dates are 1042-1118.  Rin-chen-sgang was founded by Dgyer Sgom-pa Chen-po (1090-1171) in 1119  (Las-chen 1495/1972: I 376), and no person could have been named after this place before its founding (it is quite likely, even, that Rin-chen-sgang-pa is here just a way of referring to the founder of Rin-chen-sgang, Dgyer Sgom-pa Chen-po).  Hence this thang-ka cannot predate 1119.

[17]Dkar-chag means a guidebook, catalog or register.  For more on this significant and too often overlooked genre of Tibetan literature, see Martin (1996).

[18]One author of a Reting dkar-chag, a mysterious ’Brom Shes-rab-me-lce, is mentioned in Lhun-grub-chos-’phel (1994: 90), although it is not stated that his was the one on which the modern author based himself.

[19]Lhun-grub-chos-’phel (1994: 121-138).  The author possibly took the number 3,600 from Dpa’-bo (1564/1986: 708), or they may have had a common source, perhaps an early guidebook to Reting.

[20]On this sort of thang-ka, see Brown (forthcoming).  Phag-mo-gru-pa (1507: I 323v-326v) has a collection of brief texts on the ritual of making and consecrating footprints of the teacher made on cloth, a practice explicitly connected with Atiśa.

[21]Lhun-grub-chos-’phel (1994: 130-1).  There are some odd echoes of this story in Dpa’-bo (1564/1986: 697), a passage studied in Singer (1994: 108, note 60) and some subsequent publications.  On the previous page of Dpa’-bo (p. 696) is a story involving Atiśa and a thang-ka of Tārā which is said to exist (in Dpa’-bo’s times) in Reting (compare Singer 1994: 108, note 61; Eimer 1979: 280).  Another passage in the Reting guidebook (Lhun-grub-chos-’phel 1994: 164-5) tells of a thang-ka, once kept in an outlying chapel, of an Acacia Forest Tārā that was a receptacle of the high aspiration (thugs-dam-gyi rten) of Nāgārjuna (see the prayer he composed, translated in Willson 1996: 282-5), who transmitted it to Atiśa, and from him to Dgon-pa-ba (whose monastic name was Dbang-phyug-rgyal-mtshan, dates 1016 to 1082), then to Sne’u-zur-pa, after whose death, in 1118, it was given to Reting Monastery.  This passage also deserves close study.  Still another passage (p. 125) tells of two thang-kas, including one of Acacia Forest Tārā, that was constructed as a receptacle of the high aspiration of Atiśa’s parents.  Then (pp. 134-5) there was an Acacia Forest Tārā that was the receptacle of the high aspiration of Smyug-rum-pa (monastic name Brtson-’grus-rgyal-mtshan and dates 1042 to 1109), consecrated by Dgon-pa-ba.

[22]They are pictured above the Ford thang-ka’s large central Tārā’s right and left shoulders.  Their images ought to be compared to those in corresponding locations in an early Tārā thang-ka, one with certain other iconographic and stylistic similarities to the Ford Tārā, illustrated and described in Rhie & Thurman (1999: 51, fig. 9).

[23]This script is Indic in its origins, to be sure, but it was very much studied and used by Tibetans (see on this point Singer 1998: 73).  These lines are definitely not in Tibetan transcription, as stated in Kossak & Singer (1998: 59, n. 8), where it says “This inscription is composed in Sanskrit and transliterated in Tibetan script.”  This contradicts the main body of the text (p. 57), which says, correctly, that it is in “red ornamental Sanskrit script (ranjana).”

[24]Except for the reading ’tsher-ba (‘feeling morally pricked’ or ‘regretful’), which should rather be ’tshe-ba (‘harming, hurting’).

[25]Dhammapada (1988: 52, verse 6).  This is only one of the numerous available translations of the Dhammapada.  This is not an appropriate place for a study of its translation history.  Tibetans know this text as the Ched-du Brjod-pa’i Tshoms, corresponding to the Sanskrit title Udānavarga.  An edition of the Udānavarga in Tibetan has been undertaken by Zongtse & Dietz (1990), a work not currently available to me.  Another early thang-ka (illustrated in Kossak & Singer 1998: 67; see also the photograph on p. 19 where the verse is contained on the 3rd and 4th lines from the bottom of the mchod-rten) with this same verse inscribed on it is noted in Singer (1994: 109), including the text and an English translation (but note that the words “composed in Tibetan” are misleading here).  A slight difference in the first line indicates that the verse on the back of the thang-ka corresponds to the verse in the context of the Tibetan Pratimokṣa, rather than the Tibetan Udānavarga.  More relevant to our particular thang-ka, given its date, Phag-mo-gru-pa (1110-1170), in one among his four consecration texts (the fourth one is actually composed of five smaller untitled texts) called the Rab-gnas Sa-bcu-ma (1507: I 223r-231v, at 226v), includes the Bzod-pa Dka’-thub (the name of the ‘patience creed’ based on its initial words) among the inscriptions that ought to be written on the backs of thang-sku after the work of painting has been completed (and before the rab-gnas, the consecration proper); later on, at the very end of this text, he writes out the verse in full (231v; compare 193r.5, where it is used at the end of still another text).  Jonathan Silk informs me that the patience creed is also found at the ends of a number of vinaya texts (in order to save us from a lengthy footnote discussion, we will urge J. Silk to write an article of his own on the subject).  In the many works called Gzungs-’bul, a genre that emerged out of the context of more general consecration manuals in about the 15th century, similar passages on the things that ought to be inscribed on thang-ka backs are almost always to be found, and some of these inscriptions they also recommend for use at the ends of holy books.  For an extensive listing of Gzungs-’bul texts, see Bentor (1996: 367-9). 

[26]Compare the ‘request for forbearance’ in Bentor (1996: 213-4).

[27]The Ford Tārā probably should probably, in terms of its iconography, be referred to as an Acacia Forest Tārā (although I would be the first to admit that the iconographical distinctions between her and the Green Tārā and the singular Eight Fears Tārā are less than clear to me; apart from the absence or presence of the small lotus footstool and the botanical identification of the lotus[es] above their shoulders they seem virtually identical; according to Srivastava 1989: 131, the distinction is that the Acacia Forest Tārā should have two attendants depicted on either side.  The Ford Tārā has four attendants, excluding the two figures upholding the sides of the lotus on which she is seated).  See Allinger (1997, 1999) and the brief sādhanas for Green and “Rosewood” Tārās contained in First Dalai Lama [1980]).  Perhaps more germane to the Ford Tārā, a 17th-century anthology of religious practices contains a section devoted to a sādhana for Tārā as practiced by the Bka’-gdams-pa school (Gtsang Mkhan-chen 1985: 16-18), extracted from a guidance text by a famous 13th-century abbot of Snar-thang Monastery named Mchims.  Here the primary deity is the Acacia Forest Tārā, while the Eight Fears Tārās are visualized one by one (but in wrathful forms, not the pacific forms found in the Ford Tārā).

The thang-ka that we now refer to as the Ford Tārā was to the best of my knowledge first published as an appendix to Pal (1984).  At that time the inscriptions were not known, since they were concealed under a backing cloth, which was first removed by Robert Bruce-Gardner of the Courtauld Institute of Art (London) in 1995 (Singer 1998: 67).  Still, Pal quite correctly dated it to the twelfth century.  He was also probably the first to speculate in print that it might have been imported from India, or painted by an Indian in Tibet.  “Indeed in my opinion this thanka may well be of Indian origin, or was perhaps painted by an Indian artist in Tibet.  If the former, then it may be the only example of a Pāla painting on cloth known to date.” (Pal 1984: 209, note 19).  But in the appendix, evidently written later, he writes in a more cautious vein, “Thus, the thanka was very likely painted in Central Tibet for a Kadampa monastery, but its astonishing affinity with Pala art makes us wonder if the artist was not an Indian from Bihar.”

[28]A perfectly correct reading is supplied in Kossak & Singer (1998: 59, note 9).  My reading is ra·sgreng·ba’i·lha/.

[29]I believe that here Ra-sgreng-ba refers to the monastic community of Reting.  There is no person known to me who bears a name anything like R[w]a-sgreng-ba (‘the one[s] from Reting’), with the exception of Byang-sems Rwa-sgreng-pa who, being a contemporary of Tsong-kha-pa, was active in the late 14th century (but then, he would not be referred to simply as Rwa-sgreng-pa without further specification).  For a literary example of Rwa-sgreng-ba being used in the sense of the monastery’s community, see Lhun-grub-chos-’phel (1994: 128).

[30]It is true that the penmanship is markedly different from that of the other black ink inscription (observing the differing slants in its letters ra and ’a[-chung]), showing that it must have been done by a different hand, but it does have equally old features (it also seems that it could have been retraced).  There are too many question marks to say anything with confidence about the history that would explain how this thang-ka left its home in Reting and came into the ownership of American collectors.  My current idea is that it very well might have been looted from Reting Monastery by Tibetan government soldiers in 1947 (see Carnahan 1995: 116-131; Goldstein 1989: 516-521; Surkhang 1983: 39).  The Tibetan Trade Delegation visited the Newark Museum in September, 1948.  Eleanor Olson, in charge of the Asian collections of the Newark Museum from 1938 to 1970, is as far as we know the first American owner of the Tārā thang-ka.  More specific information is needed before forcing any conclusions.  In a communication of April 17, 1999, Valrae Reynolds, the present curator of the Newark Museum’s Asian collections, expressed her doubts that the Tārā thang-ka would have been acquired from the Tibetan Trade Delegation, but rather like other works of art in Eleanor Olson’s personal collection, must have been purchased from a New York art dealer toward the end of the 1950’s or in the early 1960’s.

[31]It is extremely doubtful that this statement is intended to refer to a previously existing receptacle, which Bya Brtson-’grus-’od then took over as an object of his personal devotions.  If this were the intended meaning, it should read rather in the following hypothetical way:  *Bya Brtson-’grus-’od-kyi thugs-dam-gyi rten-du mdzad-pa, ‘It was made to be the receptacle of high aspirations of Bya Brtson-’grus-’od.’  For examples of this phraseology, see Lhun-grub-chos-’phel (1994: 129, 131) and Nyang-ral (1988: 52).

[32]There may be some paleographical significance in the use of tsheg before shad, as well as the distance between them, which violates (later, but just how much later?) punctuation standards.  One might note also in all the black ink inscriptions, there is a tendency to shift the second letter in stacked letters (specifically the first ‘p’ in Spyil-phu-ba, the ‘g’ in Ra-sgreng, and the ‘h’ in lha) toward the right.  This would also go out of style in subsequent centuries.

[33]Commentarial mchan notes in Tibetan are usually connected to the words on which they comment by a string of dots.  Missing portions of text that ought to be inserted may be connected in the same way, or marked by two identical footnote symbols (drawings of eyes being among the most common), one of them at the point of insertion, and the other next to the insertion itself which may be located elsewhere on the page, or occasionally on some other page.  Even in the absence of these usual signs, I believe I can make out a stroke (or is it just a smudge?) marking the insertion point (and it is on this point that the group of three syllables is centered).

[34]There is a tradition according to which the final elements of the names of Sarvāstivādin monks ought to be chosen from the four words dpal, ’od, grags-pa and bzang-po.  All Tibetan monks belong to Sarvāstivādin tradition as far as their monastic lineage is concerned, and in the times of the Tibetan Empire and the early Second Spread (Phyi-dar, a term discussed later on), it does seem that this naming tradition was followed to some degree (as in the names Ye-shes-’od, Byang-chub-’od, Zhi-ba-’od, Dbang-phyug-’od, Shes-rab-’od, Gzhon-nu-’od, Rin-chen-bzang-po, Grags-pa-bzang-po, Seng-ge-bzang-po, etc.).  Monastic naming practices deserve more study.

[35]Las-chen (1494/1972: II 115-118).  For some of the same information in English, see Blue Annals: (276).

[36]Bya ’Chad-kha-ba Ye-shes-rdo-rje (1101-1175) was the author of what arguably has by now become the Tibetan text most frequently translated into English (for one example, see Gyatso 1988), his classic statement on Mind Training, which involves a number of techniques for overcoming self-centeredness.  He was the first to make these formerly secret techniques available to a larger audience in the form of Tshogs Chos (‘Group Teachings’), a tradition continued today by H.H. the Dalai Lama, among others.  Bya ’Chad-ka-ba (note, and then ignore, the slight spelling difference) does appear in a lineage of the Acacia Forest Tārā contained in Dalai Lama V (1970-71: I 144).  He received his name by virtue of his founding of Old ’Chad-kha Monastery in Mal-gro (also spelled Mal-dro), where he stayed for eleven out of the last 30 years of his life [I believe this means he stayed there from about 1154 to 1164, since New ’Chad-kha was founded in 1164] (see Las-chen 1494/1972: II 113; Blue Annals, 275; Yar-lung Jo-bo 1988: 102), and where he is said to have gathered an assembly of about 900.  It may be of significance to note that, even though later sources generally spell the name as ’Chad-kha, the original spelling was most probably Mchad-kha, as in our inscription.  Mchad-ka or mchad-kha is equivalent to mchad-pa, meaning ‘cemetery’ as well as ‘tomb mound’ and ‘coffin’ (Btsan-lha 1997: 184).  As an Old Tibetan word used in connection with funerals, it may be spelled chad[-pa] as well as mchad[-pa].  It is entirely possible that Old Mchad-kha was founded on or near the site of a cemetery to which Mchad-kha-ba went in order to meditate on impermanence, a practice known to Theravāda, Mahāyāna and Vajrayāna Buddhism alike.  (In this paper, I have not ‘fixed’ the different spellings of his name, since the fact that these spellings differ from the spelling in our inscription bears significance of some kind or another.)  That Spyil-phu and Mchad-kha monasteries had painting traditions associated with them is proven by Se Spyil-phu-ba’s two successors to the abbacy.  The 2nd abbot, Lha Lung-gi-dbang-phyug Byang-chub-rin-chen (1158-1232), had been inspired to study with Bya ’Chad-kha-ba when he saw his portrait (’dra-’bag), but learning that ’Chad-kha-ba had passed away, he went to study with Se Spyil-phu-ba instead.  During his time there began a tradition of painting the 2nd abbot’s portrait and that of his nephew (to be mentioned presently) on either side of the central figure of 11-headed Avalokiteśvara.  During the tenure of his nephew, the 3rd abbot Lha ’Gro-ba’i-mgon-po (1186-1259), there were thang-kas made with him as the central figure surrounded by the sixteen Arhats.  These abbots were of royal descent, and the 3rd abbot in particular was very widely influential, and was given headship over many monasteries and their properties, most of them located in the eastern regions of Central Tibet.  For their biographies, see Las-chen (1494/1972: II 119-125) and Yar-lung Jo-bo (1988: 105-110).  It is perhaps of tangential interest to note that Yar-lung Jo-bo (1988: 98-99) says how Lha ’Gro-ba[’i]-mgon[-po] gave the three types of receptacles (rten) to Rwa-sgreng where they were preserved even as he was writing in 1376 (Lha ’Gro-ba-mgon-gyis gnang-ba’i rten gsum-po ’di / da-lta’ang Rwa-sgreng-gi gdan-sa su byed-kyis ’bral-med-du ’chang-gin yod), but then one ought to compare this to Blue Annals (p. 267) where the three receptacles are also specified.  The only painting mentioned in these passages is one of Trag-shad (Raudrāntaka).

[37]Spyil-phu and New ’Chad-kha monasteries were evidently located rather close to each other, with Spyil-phu located at a higher elevation, in the valley of ’Phan-yul (in the northern reaches of the river-system that empties into the Kyichu, which then flows past Lhasa).  Both were only a short journey away from Reting Monastery.  It is not clear why a New ’Chad-kha had to be built, but it seems that the old one was located in nearby Mal-gro, the valley that leads up to ’Bri-khung Monastery.  I have not been able to find published evidence of their recent existence, but Spyil-phu at least continued first as a Bka’-gdams-pa (as such it continued to flourish into the mid-14th century; see Blue Annals: 281), then as a Dge-lugs-pa establishment, up until at least the seventeenth century, at which time it housed about forty-four monks.  The name of Spyil-phu monastery signifies that it was originally nothing more than a grass meditation hut (spyil-bu or spyil-po) in the high part (phu) of a valley, used for a long retreat by Se Spyil-phu-ba, whose name was derived from the name of the place.  As in the histories of many monasteries in those times, it was around the nucleus of the retreat hut (or in some instances, the retreat cave) that the monastery with its chapels and living quarters gradually evolved.

[38]Uray (1984: 351, based on a publication of Heather Karmay [Stoddard] not presently available to me) gives the example of a painting of Avalokiteśvara by the Monk (Bhikṣu) Dpal-dbyangs which bears dates corresponding to 836 ce.  I do not know of other inscriptionally dated paintings by Tibetan artists of these early centuries, although given that there is this one example there must surely be others.  Some early Dunhuang paintings with Tibetan-language painters’ inscriptions (not donors’ inscriptions, as stated in the catalogue) are illustrated in Whitfield & Farrer (1990: 73-75, plates 52-54).  The painters’ names The-god-za Leg[s]-mo and Do-khong-legs appear.  The first painter was most definitely a woman, as indicated by the syllables za (‘wife,’ ‘woman from the clan of…’) and mo (in this case a feminine gender marker).

[39]In Tibetan manuscript practice there is a strong reluctance to cross out letters once written.  The letters are regarded as intrinsically sacred, and doubly so in case of a name of a holy personage.  I have noticed manuscripts which actually do have vertical lines crossing out each letter, but these are remarkably rare (in manuscripts on thick paper, the letters might be scraped and even gouged out as much as possible, but in the case of these inscriptions on cloth, gouging would of course be precluded).  The usual method is to place a line of dots above the passage that is to be cancelled out, or to place three dots in triangular formation over each syllable.  None of these methods of cancellation are in evidence here.  Therefore it may be possible to read “Mchad-kha-ba” as a true insertion, in which case we would have to simply add it to the main line of the text and follow the resultant reading of the whole, “[In it] abides the consecration of Se Spyil-phu-ba [and] of Mchad-kha-ba.”  Now it is true that genitives generally hook directly to the noun following, and do not ‘jump’ as the first genitive seems to do here.  Perhaps more accurately called ‘serial genitives,’ there ought to be a shad punctuation mark dividing the first substantive from the second.  Although by no means common, a clear example of a serial genitive may be seen in a medical encyclopaedia (Dbang-’dus 1983: 464), in the context of a definition of nerves or ‘white channels’ (rtsa dkar) by Stag-tshang Lo-tsā-ba (b. 1405).

[40]That the residents of Spyil-phu could be referred to by this collective name (although with a slight difference in the spelling) is proven in a passage in Yar-lung Jo-bo (1988: 112):  gdan-sar bzhugs ring-la Spyil-bu-ba-rnams-la’ang bde-skyid rgya-chen-po byung-zhing, ‘During his long stay in the monastic headquarters, the residents of Spyil-bu experienced a great deal of contentment.’

[41]Yar-lung Jo-bo (1988: 109; compare also Blue Annals, p. 279) tells the story about how the previously mentioned 3rd abbot of Spyil-phu Monastery, Lha ’Gro-ba’i-mgon-po, was invited to consecrate a large image in ’Dam Ko-khyim [an 11th-century foundation] that had been invaded by a gdon-spirit [held responsible for serious psychiatric disorders that might now be labelled as schizophrenia or autism; see the last chapters of Clifford 1984] that no other consecration master had been able to remove. As signs of his successful exorcism [a kind of exorcism rite is included in consecration rituals as a precaution; see Bentor 1996: 117-9, 150-9, 224], all those present saw two pigeons fly out of the image’s two armpits, as well as a pair of fighting jackals coming out toward them.  Incidentally, it still hasn’t been well appreciated that all three types of receptacles (‘support’ being another possible translation, these three fairly cover the field of holy objects recognized by Tibetan Buddhism) ought to have dhāraṇīs (in the broader sense, including relics of all types) inserted in them.  In three-dimensional artworks, like images and mchod-rtens, the word ‘insertion’ describes precisely what is done.  Obviously holy books and paintings have no hollow spaces in which to insert anything.  Still, one finds that fragments of old scriptures are placed inside the thick cover-pages of Tibetan books, traces of relics are used in the ink (see Martin 1994: 306, n. 9, in particular), and dhāraṇīs are often written out at the very end of the text.  In the case of paintings, mantric syllables may exist beneath the paint (R. Bruce-Gardner in Kossak & Singer 1998: 204-5), and very frequently dhāraṇīs are placed on the back in addition to the usual three syllables behind each holy or divine figure, perhaps together with their appropriate ‘name’ or ‘heart’ mantras.  The Tibetan consecration literature, even as it argues the need for consecration rituals to be done, makes fairly generous allowances for varied levels of elaboration in the actual practice.  On dhāraṇī insertions in general, including a translation of one of the earliest Tibetan-authored works on the subject, see Bentor (1995).

[42]See Bentor (1996: 21).  The word ‘consecration’ is reserved for translating the Tibetan term rab-gnas, while ritual actions that customarily precede the rab-gnas, such as sa-chog and gzungs-’bul, might be collectively termed ‘pre-consecration rituals.’  For a brief and accessible introduction to consecration as well as gzungs-’bul rituals, see Reynolds et al. (1986: 56-59, 134-135.  It has been said that Bya-yul-ba (Gzhon-nu-’od, 1075-1138; he started building Bya-yul in 1113 and it was completed 13 years later) was the first Tibetan to perform consecration (rab-gnas) rituals (Las-chen 1494/1972: I 367), although it is also recorded that his teacher Spyan-snga-ba Tshul-khrims-’bar performed them at the founding of Lo Dgon-pa in 1093 or 1095 (Yongs-’dzin 1990: 275).

[43]Bentor (1996: 207, 332); Phag-mo-gru-pa (1507: I 230v).

[44]I believe that, when used to refer to a receptacle, the word thugs-dam may be explained as a elliptical form of the expression thugs-dam-gyi rten (as in a preceding example).  Likewise yi-dam (yi being an allomorph of yid, ‘mind’) may be understood as an elliptical form of yi-dam lha or yi-dam-gyi lha (any divine form of Buddha that might form a focus of high aspiration in sādhana practice); see further on.  It is true that thugs-dam may have still other meanings in other syntactical environments.  If thugs-dam is something that one enters, cultivates or abides in, it means ‘meditation.’  If one requests a highly respected Lama for a divination, one refers to it respectfully as a thugs-dam (rather than using the ordinary word for divination, mo).

[45]I would like to thank Christian Luczanits for pointing out the latter inscription.  This inscription (“lha-btsun-ba nā-ga-ra-dza’i thugs-dam”), together with a photograph of the image, has been published in Klimburg-Salter (1998: 201, fig. 218).  Here thugs-dam is translated as ‘personal image.’  The birth and death dates of Nāgarāja have been deduced by Vitali (1996: 241-2) as 988 and 1026.  A number of images with inscriptions connecting them to Nāgarāja have been published, but they have not yet been studied and analyzed as a group.

[46]Nyang-ral (1988: 416), but see also an earlier passage (p. 274), where the temple complex of Bsam-yas is clearly identified as the thugs-dam of Khri-srong-lde-brtsan.

[47]This Vajravārāhī thang-ka is illustrated and specifically discussed in Singer (1994: 131-132 and fig. 26), Singer (1997: 57-57, including plate 40) and Singer in Singer & Kossak (1998: 96-99, including plate 20).  In what follows I will differ with three or four specific points of Singer’s treatment, while, it ought to be noted, agreeing with everything else (except a minor point of spelling, skyes rim for bskyed-rim).

[48]It does seem unlikely that this inscription would have been placed on the front of the thang-ka  by Dbon-po Dpal himself.  For one thing, he probably would not refer to himself with the reverential 'precious lama' (bla-ma rin-po-che).  More likely it was placed there by a later person in order to memorialize the abbot's role in bringing the painting to completion.  Still, whether it was inscribed earlier or later has little effect on the 'meaning' of the inscription.

[49]His childhood name was Dkon-mchog-sras (Stag-lung 1992: 589; Rta-tshag 1994: 511).  He received the name Grags-pa-dpal-’od-zer-bzang-po in his 12th year, during his first monastic ordination, from the hand of Sangs-rgyas-yar-byon (Stag-lung 1992: 601; Rta-tshag 1994: 522).  Sangs-rgyas Dbon (‘Buddha[’s?] Nephew’) and Dbon-po Dpal (‘Glorious Nephew’) would seem to emphasize his status as the abbot’s nephew (although all the abbots were nephews of preceding ones; note that the 2nd abbot of ’Bri-gung was also called Sangs-rgyas Dbon; named Bsod-nams-grags-pa, the latter was abbot from 1222-1234).  He is called Sangs-rgyas Dbon Grags-pa-dpal in Ko-zhul (1992: 1741).  His father was a brother, sharing the same parents, of Sangs-rgyas-yar-byon (ibid.: 586).

[50]Based on the inscription in Kossak & Singer (1998: 99, n. 5).  Compare also the translation of a virtually identical inscription, this one found on the back of a tsa-ka-li, in Heller (1997: 49), an article which has further important information on Sangs-rgyas-yar-byon.

[51]The translation of the red inscription, when combined with the the passage it has in common with the gold inscription, as they appear in the catalogue (Singer in Singer & Kossak 1998: 99), may be constituted as follows:  ‘Each of the 108 consecrations of the XXX masters and disciples, beginning with Chöje Rinpoche through to Lama Rinpoche Onpo dwells [here].’

[52]I take the use of the ablative case suffix, as it is understood by Tibetan grammarians, to indicate the 'point of origin' ('byung-khungs), and more specifically in this instance to indicate the point of departure for the motion implied in the term yan-chod.

[53]Singer in Singer & Kossak (1998: 99) states that “there can be no doubt that Chöje Rinpoche refers to Tashipel,” Tashipel being Stag-lung-thang-pa Bkra-shis-dpal, while Singer (1997: 58) says, “The Epithet Chöje Rimpoche almost certainly refers to the Taklung Founder Tashipel.”  In fact, chos-rje rin-po-che is a title of respect used to preface the names of a large number of teachers in the early Bka’-brgyud-pa lineages.  It means ‘precious religious leader.’  It is not a proper name.  In an early biography of Phag-mo-gru-pa contained in his collected works (Phag-mo-gru-pa 1507: I 2-27), the title in the colophon reads:  Chos-rje Rin-po-che’i Rnam-thar Yon-tan Rin-po-che’i Phreng-ba.  Here, chos-rje rin-po-che can refer to no one but Phag-mo-gru-pa.  In fact, elsewhere in the same collection (1507: III 265) we may read “chos-rje rin-po-che Phag-mo-gru-pa.”  Within the titles of many early Bka’-brgyud-pa biographies, chos-rje rin-po-che is prefaced to the names of their subjects.  These biographies include those of ’Ba’-ra-ba Rin-chen-dpal, Gtsang-pa Rgya-ras-pa, Dus-gsum-mkhyen-pa, Rgod-tshang-pa (1189-1258) and Spyil-dkar-ba (1228-1300).  In the biography of O-rgyan-pa, his teacher Rgod-tshang-pa is most often referred to as either chos-rje or chos-rje rin-po-che (examples of the latter in Bsod-nams-'od-zer 1997: 32, 47, etc.).  In short, in the absence of contextual indications, and going by the title chos-rje rin-po-che alone, we can come to no conclusion about who is intended.  True, it could be referring to Stag-lung-thang-pa, among many other possibilities.  Most likely, it refers to one of the abbots of Khams Ri-bo-che, all of whom do often have chos-rje or chos-kyi rje prefixed to their names.

[54]Hahn (1974: 170-173) and Yamaguchi (1997).  Yamaguchi supplies a number of examples of actual usages in Tibetan literature.  Other postpositions created using the same pattern as yan-chad and man-chad are phan-chad, tshun-chad, phyin-chad and slan-chad.  In each of these cases, the chad (or chod) element indicates a cut-off point, more or less arbitrarily determined, drawn through the particular spatial or temporal context.

[55]They both use the same example drawn from the Padma Thang-yig, which reads:  lo brgyad yan-chad which they both translate as ‘above eight years old.’  I was unable to locate it in its original context, but I did find an entirely parallel construction in Mang-thos (1988: 201):  me mo lug (spyi-lo 1367) la Rong-ston Shākya-rgyal-mtshan Shar Rgyal-mo-rong-du byon / shar Tsong-kha-pa-las lo bcus gzhon / lo bco-brgyad yan-chad-du Bon-la ’chad rtsod rtsom gsum gnang…, which we must translate, ‘In the Fire Female Sheep year (1367 ce), Rong-ston Shākya-rgyal-mtshan [1367-1449] came to Rgyal-mo-rong in the east.  He was younger by ten years than the easterner Tsong-kha-pa.  Prior to age 18, he performed the three [actions expected of a scholar] teaching, debating and composition for the Bon [religion].’  This cannot be read, following Das and Jäschke, as if it meant that he was a Bon-po when he was ‘above eighteen years old’, since it is well know that the famous Sa-skya teacher (and sometimes critic of Tsong-kha-pa) had been a Bon-po until he was 18 (see for example van der Kuijp 1986: 205) and a Sa-skya-pa thereafter.

[56]Dag-yig (1979: 716) reads:  bya-dngos-la rang-dgar mtshams bzung-ba’i stod-kyi cha’am sngon-gyi cha’i ming-ste, which resists precisely literal translation.  Bya-dngos might be translated as ‘action and/or object,’ although I just translate it as ‘something.’

[57]Quite a few Old Tibetan examples of the use of yan-chad and man-chad are known, if relatively seldom in temporal contexts.  The most clearly temporal usages I could locate are in the inscription at Skar-chung (Richardson 1985: 72-81).  These, being in Old Tibetan language, have further interpretational problems that do not concern us here.  To paraphrase my reading of lines 34-35 of the inscription, the Emperor is recommending that religious preceptors who are fully ordained monks should be appointed to his successors upward (yan-chad, i.e., backward in time) to the time they are small in body as well as downward (man-chad, i.e., forward in time) from the time they are enthroned.  I believe that here the moment of enthronement is the boundary in time from which we must calculate backward (yan-chad) and forward (man-chad).  Even if the words do occasionally occur in historical narratives, for use in temporal contexts the most examples by far occur in works of the chronology (bstan-rtsis) genre, but also in ‘records of teachings received’ (gsan-yig or thob-yig).  The thob-yig of Mon-ban (1985: 46-47) gives, for example, a lineage of 13-deity Cakrasamvara from Vajradhāra through several Indian masters, entering Tibet with Ras-chung-pa, and then passing into the general ’Brug-pa Bka’-brgyud-pa lineage from Gling-ras-pa (1128-1188) and his disciple Gtsang-pa Rgya-ras-pa (1161-1211).  Then it gives another lineage, ‘The lineage through which I received the biography of Lo-ras-pa and the Public Teachings Ocean of Nectar [i.e., the public discourses of Gtsang-pa Rgya-ras-pa that his disciple Lo-ras-pa put into writing] is as follows:  from Gtsang-pa Rgya-ras upward (i.e., back in time), it is the same [as the preceding lineage].  The lineage [from him] downward (forward in time) is as follows: Lo-ras-pa [1187-1250], Rtsa-ri-ras-chen…’ (Rje-btsun Lo-ras-kyi rnam-thar dang / Tshogs-chos Bdud-rtsi’i Rgya-mtsho thob-pa’i rgyud-pa ni  / Gtsang-pa Rgya-ras yan-chad cig / man-chad-kyi rgyud-pa ni  / Lo-ras-pa  / Rtsa-ri-ras-chen  /…).  Several examples of the use of yan-chad occur in Mang-thos (1988: 3, 33, 55, 72-74, 77, 83, 115, 158), and most of these do support our hypothesis that yan-chad involves moving back in time.  However, a passage that Mang-thos has drawn from the Chos-la ’Jug-pa’i Sgo (the 1167 work by the Sa-skya patriarch Bsod-nams-rtse-mo that ends with the famous historical section) seems to disprove it (Mang-thos 1988: 72; the example on p. 55 also could be brought forward).  Turning to the original source of Mang-thos’ citation (Bsod-nams-rtse-mo 1968: 343 column d) only confirms our idea that, in this particular passage, yan-chad moves forward in time.  However, even in this work there is no consistency, since on the preceding column (343 column c) we find another usage where yan-chad moves backward in time (‘Prior to that one [Emperor Ral-pa-can] the Tibetan subjects were basking in the sun of happiness and it was an auspicious period’; de yan-chad Bod ’bangs skyid-kyi nyi-ma ’de-zhing bkra-shis-pa’i dus yin-no // //.  Obviously, the period of fragmentation that came soon on the heels of Ral-pa-can’s assassination was not the auspicious period referred to).  Rather than attribute the inconsistency to the author (highly unlikely given the close proximity of the passages), I would suggest a problem in the manuscript transmission.  This would at least have the effect of saving the at-risk hypothesis.  Ramble (1996: 151) has argued in another context that the passage of time, for Tibetan speakers, involves a progression from the high to the low.  For more examples of usage, see Yamaguchi (1997).

[58]To see this point more clearly, it is advisable to consult Singer (1997: plates 38 and 40 on pp. 56-57; compare also Rhie & Thurman 1999: 316, plate 102), where a double portrait thang-ka (consecrated by Dbon-po Dpal) of the 2nd and 3rd abbots is reproduced on a page facing the Vajravārāhī thang-ka.  The hand gestures (mudrā) of the figures are identical in both thang-kas, Sku-yal-ba with both hands forming the ‘teaching’ (or ‘turning the wheel of the Dharma’) gesture and Sangs-rgyas-yar-byon with right hand only in teaching gesture, left hand in meditation gesture on his lap.  The top register of the Vajravārāhī contains, from your left to your right, 1. Vajradhāra.  2. Tilopa.  3. Naropa.  4. Mar-pa.  5. Mi-la-ras-pa.  6. Sgam-po-pa.  7. Phag-mo-gru-pa.  8. Stag-lung-thang-pa.  To this we add, in the upper ends of the two side registers beneath the top register:  9. Sku-yal-ba (1191-1236) and 10. Sangs-rgyas-yar-byon (1203-1272).

[59]See Singer (1994: 96, n. 52; 1996: 182; 1997: 294, n. 4; 1998: 72) and Singer in Kossak & Singer (1998: 11, 18-19), where sādhaka is misleadingly translated as ‘the officiating hierarch’ (as well as ‘officiant’ and ‘sacrificer’), which is even on occasion further interpreted as being the consecrator.  This is ultimately based on Lalou (1930: 15), who uses the term l’officiant to translate sādhaka (Tibetan sgrub-pa-po), a word which would nowadays usually be translated as ‘practitioner’ (meaning someone who practices a sādhana).  It is unfortunate that, with all these repeated references to Lalou’s introduction, no mention was made of the passage in the French translation (Lalou 1930: 56) or the transcription of the Tibetan text (the final four lines on p. 98), where the iconography of the sādhaka is described (see below; also, Lalou 1935: 252).  Nearly every Tibetan, whether monastic or lay, practices at least one sādhana, and has been initiated into the practice of a divine form of Buddha (a high aspirational deity, a yi-dam [lha]).  This by no means makes all of them officiating hierarchs.  At the very least this latter status would require an extended retreat period for completing the Generation Process (Bskyed-rim) meditations.  It may be argued that the word ‘patron’ emphasizing financial support for the artist ought technically to be used only to translate sbyin-bdag or yon-bdag (both might be more literally translated as ‘donor’).  Here we quite consciously use the word patron in a broader sense, since the one who originally conceives the work is generally also the one who, by whatever means (including their own financial resources or donations solicited from others, and of course dealing with the artist[s] in still other ways), brings it to completion (but for an exceptional example, note the already given account of the high aspiration of Atiśa that was brought to completion or ‘erected’ [bzhengs] by his Tibetan disciple Great Yogi).  Patronage of art everywhere generally does involve much more than simple monetary reward.  What makes Tibetan-style Buddhist thang-kas truly different from paintings in most other cultures is their use as part of a sādhana practice in which the patron is personally engaged (that there are other motives and usages for thang-kas within Tibetan culture, and there certainly are, is not at issue here).  The three works on Tibetan art by Kye-rdor Mkhan-po (1779-1838 ce) studied in Tanaka (1997) are instructive in this regard (including a large number of scriptural citations), even if composed rather late and by a Mongolian author (Mongolian Buddhists had both a personal devotional and ‘ethnographic’ interest, in the sense that they often looked more closely into ‘common’ aspects of Tibetan Buddhist culture that were relatively neglected, because taken more for granted, by Tibetan authors).  These works insist that the relationship between patron and artist be motivated and guided by the highest Buddhist ethical ideals, in particular non-attachment (the artist should not be displeased if the offering is small…), and it is to be assumed that in some degree the reality falls short, otherwise there would be no need for the authorial insistence.  We might note, at least for entertainment value, that one early Tibetanist at least thought that the corner portrait represented the artist:  “On completion, the artist puts a miniature figure of himself in a corner at the bottom in an adoring attitude” (Waddell 1895/1972: 331).

[60]This kneeling posture might suggest yet another line of interpretation, although one that does not really  contradict the interpretation suggested here, that it is the ‘exhorting the patron’ sequence in the consecration ritual (and it must be taken into account that this posture only appears in a certain percentage of the known examples).  In general, the posture of kneeling on one knee signifies that one is in the actual presence of a Buddha, imitating the characteristic posture of persons who play the role of questioner in the sûtras.  In ritual practice even today, it is not likely to be used in any other context apart from the taking of bodhisattva vows.  It is possible that verses 7-9 of Atiśa’s famous work the Bodhipathapradīpa (‘Lamp on the Path to Awakening,’ of which there have been a number of editions and translations), could be considered as a proximate source text for these depictions.  The general subject of the following quotation is the formal voicing of an aspiration to achieve Complete Enlightenment.  In my reading it begins by clearly listing Body, Mind and Speech receptacles, which does remind one of the presence of the book and mchod-rten in the ritual scene:  “In front of painted images of the Completely Enlightened One and other [Body receptacles],/ mchod-rtens and Holy Dharma [scriptures],/ with things like flowers and incense,/ whatever one can afford, make offerings./ Also, the seven different offerings/ which are taught in the Samantabhadracarī./ With pure and irreversible intentions/ until the actualization of the heart of Awakening,/ have perfect faith in the Three Jewels./ Kneel with one knee on the ground/ and, having joined the palms together…” [the next verses are on the generation of the vow to become Enlightened].  Compare Phag-mo-gru-pa (1507: I 106r.4) which seems to be based on the Bodhisattvabhûmi.  In this latter source, the aspirant kneels on the right knee after having draped up the folds of the robe on one shoulder [leaving the other shoulder uncovered].

[61]Quite a few of the censers are reminiscent of a particular Chinese type, in which the vessel portion, although practically globe shaped, comes to a point at the top in an attempt to approximate the shape of a cosmic mountain (compare the Han Dynasty examples in Erickson 1996/7; handles seem relatively rare).  The upper and lower halves of the censer are cracked open on one side, and one may usually perceive black curls of smoke emanating from the ‘mouth’ so formed.  The handle, usually held in the figure’s hand, looks like a curved, even sometimes leafed, branch, with an extra leg permitting it to stand independently, as it sometimes does.  While Turfan patron figures generally hold flowers (Klimkeit 1990: 191), the ones depicted in Dunhuang (Wiercimok 1990: 211-212) may hold either flowers or censers.  It is at least of some interest to note that a Tibetan text from Dunhuang (The British Library’s Stein Collection, India Office Library no. 602) contains, near its end (at folio 47), an independent song of offering to the Buddhist Holy Ones.  The only offerings mentioned in this song are flowers and incense.

[62]This is my fresh translation based on the Tibetan text in Lalou (1930: 98-9).  Since this very lengthy text, long available in Sanskrit, Chinese and Tibetan, has a number of chapters devoted to various paintings (not to mention the many chapters on paintings found in still other tantras), there are undoubtedly many more scriptural clues for understanding early Tibetan art.  In an important study on the cloth painting making rites of the Mañjuśrīmûlakalpa, focussing on the preparation of the canvas (the ‘cloth rite,’ ras chog), Kapstein (1995: 251) has summarized the sādhaka description in one sentence:  “Below, the practitioner himself is to be depicted making offerings of incense.”  It seems likely that many of the tantra’s prescriptions for cloth paintings were in fact followed in Tibet in the 12th century, to judge from the four consecration works by Phag-mo-gru-pa mentioned above.  Two of them contain directions for the preparation of the canvas strikingly similar to those described in Kapstein’s article.

[63]The translation, based on Sde-srid (1990: 837) is tentative, especially since I have not yet traced the quote in its original context.  Another quote, one from the Samvarodaya Tantra, as contained in the same work (p. 922) emphasizes the benefits of thang-ka making for sādhana practice:  ‘The beginning mantrists should make images and so forth.  If they bring to completion painted images, more quickly will they attain extensive accomplishments (siddhis).’

[64]It may be of some significance that a similar stand — with, in this case, three mchod-rtens and the crossed yak-tail fly-whisks — may be seen centered below the central Buddha figure in the main surviving mural in the inner chapel of Grwa-thang Temple.  This mural would seem to date from the time of the monastery’s construction between 1081 and 1093 CE.  See Vitali (1990: plates 29-33) or Henss (1997a: plate 182).  This mural does not seem to have any figure which could be identified as the patron, but then one lower corner has been effaced.

[65]Cf. the mnga’-dbul rite within the consecration ritual; Bentor (1996: 296 ff.), although the fly-whisk is not mentioned there.  For an enlightening sketch of the history of the Anglo-Indian words chowry and cowtails, see Yule & Burnell (1886/1996: 214-5, 271-272).

[66]Béguin (1993).  The particular thang-ka depicted here is dated to the 14th century, although I think this dating doubtful.  It may be based on the presence of a yellow-hatted hierarch above the right shoulder of the main figure.  It is often mistakenly assumed that a yellow hat necessarily denotes the Dge-lugs-pa sect which, although not then known by this name, started at the end of the 14th century.  In fact, yellow hats were worn by Bu-ston Rin-chen-grub (1290-1364) and Dol-po-pa Shes-rab-rgyal-mtshan (1292-1362), to give just two clearly pre-Dge-lugs-pa examples (and the use of yellow hats seems to have gone back much further in history; just how much is unclear, but see Tseten Zhabdrung [1986: 47-48], who places use of yellow hats back as far as the 10th century, as well as Stearns [1999: 73]).  ‘Red hats’ and ‘yellow hats’ are simplistic Manchu-Chinese ethnographic categories uncritically adopted by earlier western Orientalists, despite their inaccuracy, together with the term ‘Lamaism’ itself.  (It is true that one may find these color categories used on occasion in late 18th-century and later Dge-lugs-pa writings, albeit generally by Dge-lugs-pa writers closely associated with the Manchu court.)  Byams-chen Chos-rje, the founder of Se-ra Monastery, is always depicted with a black hat, which doesn’t even remotely make this undeniably Dge-lugs-pa teacher into a ‘Black Hat’ Karma-pa.

[67]The use in Tibet of water as a substitute for other offerings is justified with a quotation from Atiśa, for which, see Pabongka (1990: 129; as part of a general discussion of offering protocol).

[68]For an exceptional example with two mirrors, see Kossak (1998: fig. 1, bottom register).  These mirrors are portrayed as black disks with arrays of dotted patterns on them (since they may be mistaken for shields, compare the mirrors in Beer 1999: 197 & 273, with the shields in ibid. 299).  It is doubtful that they are included here as part of a consecration ritual — one of them, after all, stands alone in front of the pair of lay patrons — and there is no spouted pitcher (which would be used together with the mirror in the ritual).  Probably the mirror is here intended, more generally, as a reminder that all phenomena, including the painting, are ‘reflected images’ (see Bentor 1995a).  Note the very occasional presence of mirrors in other patron scenes, such as Pal (1983: color plate no. 9, lower register, near the center; possibly also color plate no. 11, lower register, lefthand side, where there is indeed a spouted pitcher as well).

[69]There is an explicit example of this in Lhun-grub-chos-’phel (1994: 134), in which a thang-ka that was the thugs-dam-gyi rten of Spyan-snga-ba was also consecrated by him (and compare three similar examples on pp. 130, 131).  There is also no reason that the painter could not be the consecrator, and in fact, as confusing as this might seem, nothing to prevent the painter from being the patron as well.  Tibetan literature on art generally does not foreground the relationship between patron and artist, but rather that between patron and aspirational deity, or between patron and consecration ritual master.  Tourism and commercialization might appear to have done much to alter this picture in recent years (see Bentor: 1993; McGuckin 1996), but on the Tibetan side tradition prevails.

[70]Singer (1998: 73).

[71]It seems that some wall paintings from the Zhal-ras Chapel within the Jo-khang temple in Lha-sa actually date from its 1160 restoration under Zhang Rin-po-che’s teacher Sgom-tshul (for photographs, see Vitali 1990: plates 38-43, 46 as well as Stoddard 1994), and a few may even go back to the ninth century.  Accounts of the 1160 restoration do mention paintings (ri-mo).  There are a number of references to art and artists in the works of ’Jig-rten-mgon-po (1143-1217), but students of 12th century Tibetan art would do well to start with a brief text in his collected works (’Jig-rten-mgon-po 1969-71: II 10-13) in which he emphasizes correct proportions including some advice directed toward the metal workers.  Although mainly about metal images, he also spares some words about paintings, and (rather surprisingly) emphasizes the relationship between the patron (yon-bdag) and artist (lha-bzo).  See also the fairly contemporaneous work by Bsod-nams-rtse-mo (1142-1182 ce) mentioned below.

[72]For the story in English, see Blue Annals, p. 932.  I have also consulted a reprint of what is very likely the exact same manuscript upon which the Blue Annals based its account, the history of the Zhi-byed school by Zhig-po, which is reproduced in The Tradition of Pha Dampa Sangyas: A Treasured Collection of His Teachings Transmitted by Thugs-sras Kun-dga’, “reproduced from a unique collection of mss. preserved with ’Khrul-zhig Rinpoche of Tsa-rong Monastery in Ding-ri, edited with an English introduction to the tradition by B. Nimri Aziz,” Kunsang Tobgey (Thimphu 1979), vol. 4, pp. 324-432, at p. 407.  The Tārā thang-ka mentioned in this story was based on a particular Tārā Tantra, with five main figures in the central field, and some lesser deities (lha-chung) riding geese and so forth in the outer parts.  It is not, therefore, the Ford Tārā thang-ka.  This passage deserves more than the brief paraphrase given here (the painter’s name indicates that he was a member of the Bran-ka clan, the same clan that produced a number of important ministers of the imperial times).  There was also a painter (lha-bzo) listed among the Tibetan disciples of the early Bka’-gdams-pa teacher Spyan-snga Tshul-khrims-’bar (1038-1103, the dates being those given in Blue Annals, p. 285; if we follow the two main Bka’-gdams-pa histories, his dates were 1033-1103; see Bsod-nams-lha’i-dbang-po 1484/1977: 315-319; Las-chen 1494/1972: I 480-492), and it is interesting that this same teacher’s name appears in a portrait thang-ka inscription (Singer 1994: 114; Kossak in Kossak & Singer 1998: 62-64).  He was called Spyan-snga because, for seven (or nine, as suggested in an added annotation) years he served as personal attendant of ’Brom-ston (spyan-snga; see Bsod-nams-lha’i-dbang-po 1484/1977: 317, line 2).  Still another early painter appears in the Blue Annals (p. 396), a Bla-ma Lha-bzo, belonging to the Dpyal lineage of Vajravārāhī (on the Dpyal lineage, see the Vajravārāhī thang-ka with accompanying description in Kossak & Singer 1998: 100-101).

[73]Of course, there is the possibility that these woven thang-kas are faithful copies of somewhat earlier painted thang-kas.  For another of comparable age portraying the tantric deity Acala (on Acala, see Heller, this volume), see Bod-kyi Thang-ka, plate 102, which bears an inscription saying it was given to Mkhon (an old spelling for ’Khon) Grags-pa-rgyal-mtshan (whose dates are 1147-1216) by a Khams-pa student of his named Cang Brtson-’grus-grags (and we might mention that a work on Acala by Grags-pa-rgyal-mtshan’s elder brother contains directions on constructing painted images of this deity together with instructions for their consecration (Bsod-nams-rtse-mo 1968a: 453).  For more on these early fabric thang-kas (gos-sku), see Tanaka (1994) and Linrothe (1998: 101, n. 32) and references given there.  For smaller and much less clear reproductions of the Zhang tapestry, see Pho-brang Po-ta-la, plate 104 and Reynolds (1997: 124).  Tangut examples of tapestries from about the same time have also survived, most notably the Green Tārā kept in the Hermitage Museum (St. Petersburg), on which see, for example, Rhie & Thurman (1991: 126-7).

[74]One may compare this portrait of Zhang Rin-po-che with the tiny portrait (a magnifying glass is recommended) in the extreme upper right-hand corner of the Cakrasamvara Maṇḍala thang-ka illustrated in Heller (1999: color plate no. 92), which probably also dates from the vicinity of 1200 ce.  The treatments of the hairline, face, facial hair and robes are quite similar, even if the hand positions differ.

[75]The depiction of Bya-rog-gdong-can supports that idea that it is indeed Zhang Rin-po-che who is depicted in the center.  However, he received the lineage for this deity together with a specific Mahākāla cycle from his teacher Rgwa Lo (who in turn received them from Tsa-mi); see Bu-ston (1971: 31) for the complete lineage as received by Bu-ston.

[76]In particular, see the discussion of the ‘spiky mountain’ motif, common to the Ford Tārā and the Zhang tapestry along with many other early thang-kas, by Rhie in Rhie & Thurman (1991: 49), although it may also be seen in paintings as recent as the 16th century (Heller 1999: color plates 103-4).  In early Tibetan painting in general, the area immediately surrounding the human and deity figures may be treated in three different ways:  [1] with the spiky mountain motif, [2] with vines encircling each individual figure, or [3] with very simple floral patterns (often little more than sets of dots).  Elaborate Chinese-inspired (but nevertheless Tibetan) landscapes would begin to take over the background from the 15th century onward.  In more recent centuries it would even be possible to move the main figure to an off-center position, as if to provide more space for the landscape. 

[77]Another early thang-ka with very similar stylistic features to both the Zhang tapestry and the Ford Tārā is the Jucker painting featured in Singer’s 1986 article, which says, “Despite the use of eastern Indian artistic prototypes, the Jucker painting is manifestly Tibetan in origin” (p. 25).

[78]Phyi-dar is often translated as ‘Later Spread’ or ‘Second Propagation.’  Traditional historians have mostly agreed that this period began in the last half of the 10th century, although its precise date often provokes discussion (being a historical event with an aftermath and not a period, the issue of its ending is not important to them, but since we have a good name for the period of Mongol domination, we may say that it ends in the middle decades of the 13th century, that is, if periodization serves any useful purpose at all apart from helping us organize our own thinking about history).  Tibetan historians also speak of a ‘Period of Fragments,’ when there was no ruler who could exert power over the greater part of the Tibetan plateau.  This latter period lasted from 842 to 1260.

[79]Valrae Reynolds has told me as much in an electronic communication (see also Reynolds 1996), and a very recent history of Tibetan art (Dkon-mchog-bstan-’dzin 1994: 135) also says that stitched thang-kas (’tshem-thang; ’tshem-pa means ‘sewing,’ not as we would expect for tapestries, ‘weaving,’ which would be ’thag-pa; the author does go on afterward to say a few words about ‘tapestry thang-kas,’ ’thag-thang) were generally produced by Chinese seamsters on the basis of Tibetan exemplars (ma-dpe, ‘mother example[s]’) taken to China for this purpose.  Although he goes on to emphasize the superior skills of these seamsters who did not need to work with any background pattern (they would work from the exemplar placed in front of them, rather like Tibetan carpet weavers), given that this is very much a copyist technique, it would seem that it is the Tibetan nature of the exemplar that needs to be underlined. For another study of stitched thang-kas, mostly dating from a later period, see Henss (1997).  One ought also to seriously consider the possibility that Chinese seamsters migrated to Tibet in pursuit of work.  On the migrant Chinese tapestry maker, see Rossabi (1998).

[80]For the inadequacy of exclusively stylistic evaluations for establishing historical provenance, see especially Jackson (1996: 369-377).  Factors such as the conscious employment of (sometimes relatively antiquated) models, blockprinted patterns, tracing techniques and the like, as well as the very versatility, idiosyncrasy and mobility of the artists themselves, are bound to confound any effort to locate specific artworks in time and space based solely on style.  Jackson’s recent book (1996), by the way, although very much about ‘style,’ makes an admirable and quite successful attempt to illuminate indigenously Tibetan understandings of style, quite unlike most Tibetological art historians, who generally discover (or, on occasion, invent) and then go on to apply their own stylistic criteria.  The problems of extraneous reframing and ideological introjection are better eluded by sincerely doing our best to find out local understandings.

[81]For the study of more recent thang-kas, there are also the records of inscriptions placed on holy objects (the genre known as rgyab-yig, ‘writings [on the] back’) by primarily but not exclusively Dge-lugs-pa authorities of the last three centuries.  Their consequences for art history have not yet begun to be drawn.