Lineages and Structure in Tibetan Buddhist Painting:
Principles and Practice of an Ancient Sacred Choreography
David Jackson
University of Hamburg
Abstract: Though compositional structure ­ which here means specifically the
placement of divine figures ­ is an essential aspect of Tibetan painting, this theme
has rarely been discussed or described by scholars. The conventions for depicting
lineages of teachers in particular must be carefully taken into account when
documenting thang kas that contain lineages with inscriptions. The historian should
carry out, if possible: (1) decipherment of inscriptions, recording names; (2)
historical identification of individual masters, furnishing dates if known; (3A)
identification of the lineage, and (3B) listing its members in chronological order
(i.e., following the sequence of lineal descent); (4) diagramming the position of
all figures, following the numbering of step three. The present article classifies
and describes the lineage structures found in the vast majority of paintings with
lineages. Understanding lineage structure through these four steps allows the
historian to identify the religious teacher and approximate generation of the patron
who commissioned the painting, essential steps toward restoring the painting to
its lost historical context.
Introduction
Although to the uninitiated, Tibetan Buddhist paintings may seem to be a chaotic
and inexhaustibly variable universe, in fact their iconography is limited, orderly,
and, above all, hierarchic. To fathom this art, one of the first steps is to recognize
the hierarchic arrangements in which its sacred figures have been placed. For
understanding the main conventions of precedence and hierarchy, moreover, one
must learn to interpret in detail the depictions of guru lineages. Besides their
intrinsic religious, iconographic, and aesthetic interest, depictions of bla ma lineages
can furnish some of the few reliable historical clues for dating a Tibetan painting,
which is already reason enough to study them. Yet despite their importance for a
sound understanding of Tibetan art, the basic conventions of lineage portrayal ­
Journal of the International Association of Tibetan Studies, no. 1 (October 2005): 1-40.
www.thdl.org?id=T1220.
1550-6363/2005/1/T1220.
 2005 by David Jackson, Tibetan and Himalayan Digital Library, and International Association of Tibetan Studies.
Distributed under the THDL Digital Text License.
"hierophantic choreography" ­ and other compositional elements in thang ka
paintings have rarely been discussed or described in detail.
Lineage structures are not self-evident. Several linguistic or historical hurdles
must be cleared if one wants to document them in a satisfactory way:
1. Correct decipherment of inscriptions recording names
2. Correct historical identification of individual masters, furnishing dates
if known
3. (A) Correct identification of the lineage, and (B) listing its members
in chronological order (i.e., following the sequence of lineal descent)
4. Diagramming the position of all figures, following the numbering of
step 3
The first Tibetologist to study in any detail thang kas depicting lineage gurus
was G. Tucci, who in his scholarly tour de force Tibetan Painted Scrolls1 described
three thang kas that each portrayed as main figures four lineage masters of the
Ngor pa subschool of Sa skya pa "Path with Its Fruit" (Lam 'bras) instructions.
Tucci succeeded in (1) deciphering the inscriptions and (3A) identifying the main
lineage. He also (3B) correctly ordered the main figures within each thang ka,
though not as a continuous series within the main lineage.2
In the five decades that have followed, most catalogs of Tibetan artworks did
not reach the level of Tucci's work in their analyses of inscribed lineages, though
in the 1970s a few scholars began to perform at least step 1 of the documentation.
Anne Chayet in two entries of a major exhibition catalog3 documented the names
of two lineages. For painting no. 122, she presented the names of lineage masters
in their correct order (step 3B). Though she did not attempt steps 2, 3A or 4, she
demonstrated implicitly an understanding of structure.4
Another book of the 1970s to furnish names from inscriptions was a sales catalog
of paintings from Ngor Monastery published from Paris in 1978 by the Galerie
Robert Burawoy.5 This book, of unusually large format and price, documented the
names of lineage masters in several paintings, presenting them in white letters on
1
G. Tucci, Tibetan Painted Scrolls (Rome: Libreria dello Stato, 1949).
2
See G. Tucci, Painted Scrolls, 369-70, nos. 25-27. The three paintings were evidently the first,
fourth, and fifth paintings in a set that originally consisted of eight paintings. The paintings' minor
figures were not randomly selected masters as Tucci guessed, but rather ten or eleven adepts in each
painting from the eighty-four siddhas: 8 x 10.5 = 84.
3
Gilles Béguin, ed., Dieux et démons ďHimâlaya: Art du Bouddhisme lamaique (Paris: Réunion
des Musées Nationaux, 1977), nos. 109 and 122.
4
In her later book on Tibetan art and archeology, Anne Chayet (Art et Archéologie du Tibet [Paris:
Picard Éditeur, 1994], 189), when discussing prospects for future research on Tibetan art, mentioned
the analysis of lineages as a problem calling for further investigation, sketching two typical compositional
types, one earlier and one later. See also David Jackson, "Apropos a Recent Tibetan Art Catalogue,"
review of Wisdom and Compassion, by Marylin M. Rhie and R.A.F. Thurman, Wiener Zeitschrift für
die Kunde Südasiens 37 (1993): 109-30, which was not yet available to Chayet.
5
Galerie Robert Burawoy, Peintures du monastre de Ngor (Libourne: Arts graphiques ďAquitaine,
1978).
2Jackson: Lineages and Structure in Tibetan Buddhist Painting
transparent pages overlaying the color plates. Otherwise the catalog avoided
numbers for plates and pages, and it did not sequentially list, date or otherwise
identify the lineage masters.6
In several catalogs of the 1980s and 1990s, collaborators transcribed some of
the inscriptions bearing the names of masters.7 But they listed and diagramed the
names of lineage masters in an ad hoc order, not following the sequence of the
lineage. The catalog of Essen and Thingo 1989 deserves praise for presenting all
available inscriptions, but even its lists and diagrams were not in conformity with
the order of the lineages portrayed.
Since the early or mid 1990s, several other scholars noticed the potential
usefulness of lineage analysis for dating.8 In recent major catalogs the
documentation of some entries is also becoming better.9 To encourage this trend
I would like in the present article to share some of the internal rules and outer
expressions of structure that I have encountered in my own research.10
6
The origin of this, the second of two anonymous sales catalogs of Tibetan paintings by the Burawoy
gallery in the 1970s, is unclear to me, but someone in France with competence in Tibetan must have
helped the gallery owner, whose main expertise is with Japanese weapons and armor.
7
See for instance Pratapaditya Pal, Art of Tibet: A Catalogue of the Los Angeles County Museum of
Art Collection, rev. ed. (Los Angeles: University of California Press, 1990), with H. Richardson's
documentation of inscriptions in an appendix; G.-W. Essen and T. T. Thingo, Die Götter des Himalaya:
Buddhistische Kunst Tibets; Die Sammlung Gerd-Wolfgang Essen, 2 vols. (Munich: Prestel-Verlag,
1989); Pratapaditya Pal, Art of the Himalayas: Treasures from Nepal and Tibet (New York: Hudson
Hills Press, 1991), with documentation of inscriptions by H. Stoddard; Hugo Kreijger, Tibetan Painting:
The Jucker Collection (London: Serindia Publications, 2001), with documentation of inscriptions by
P. Verhagen.
8
For instance, H. Stoddarďs footnote in Pal, Art of the Himalayas, appendix; J. C. Singer, "Painting
in Central Tibet, ca. 950­1400," Artibus Asiae 54, no. 1/2 (1994): 87­136; J. C. Singer, "Taklung
Painting," in Tibetan Art: Towards a Definition of Style, ed. J. Singer and P. Denwood (London:
Laurence King, 1997), 52­67; and Kimiaki Tanaka, "The Usefulness of Buddhist Iconography in
Analysing Style in Tibetan Art," Tibet Journal 21-22 (1996): 6-9.
9
Pal et al. 2003; and John C. Huntington and Dina Bangdel, The Circle of Bliss: Buddhist Meditational
Art (Columbus: Columbus Museum of Art; Chicago: Serindia Publications, 2003). See also the
painstaking documentation of three lineage thang kas by C. Luczanits, "Art-Historical Aspects of
Dating Tibetan Art," in Dating Tibetan Art: Essays on the Possibilities and Impossibilities of Chronology
from the Lempertz Symposium, Cologne, 2001, ed. I. Kreide-Damani (Wiesbaden: Reichert, 2003),
25-57.
10
See, for instance, David Jackson, "A Painting of Sa-skya-pa Masters from an Old Ngor-pa Series
of Lam 'bras Thangkas," Berliner Indologische Studien 2 (1986): 181-91; David Jackson, "The
Identification of Individual Masters in Paintings of Sa-skya-pa Lineages," in Indo-Tibetan Studies:
Papers in Honour and Appreciation of Professor David Snellgrove's Contribution to Indo-Tibetan
Studies, ed. T. Skorupski (Tring, U.K.: Institute of Buddhist Studies, 1990), 129-44; Jackson, "Apropos,"
109-30; David Jackson, A History of Tibetan Painting: The Great Painters and Their Traditions,
Beiträge zur Kultur- und Geistesgeschichte Asiens 15 (Vienna: Verlag der Österreichischen Akademie
der Wissenschaften, 1996); David Jackson, "Lama Yeshe Jamyang of Nyurla, Ladakh: The Last Painter
of the 'Bri gung Tradition," in art issue (ed. E. Lo Bue), Tibet Journal 27 (2002): 153-76; and David
Jackson, "The Dating of Tibetan Paintings is Perfectly Possible­Though Not Always Perfectly Exact,"
in Dating Tibetan Art: Essays on the Possibilities and Impossibilities of Chronology from the Lempertz
Symposium, Cologne, 2001, ed. I. Kreide-Damani (Wiesbaden: Reichert, 2003), 91-112. An earlier
version of this paper appeared as "Tibetische Thang kas deuten, Teil 1: Die Hierarchie der Anordnung,"
Tibet und Buddhismus 50, no. 3 (1999): 22-27; and "Tibetische Thang kas deuten, Teil 2:
Übertragungslinien und Anordnung," Tibet und Buddhismus 50, no. 4 (1999): 16-21.
3Journal of the International Association of Tibetan Studies 1 (October 2005)
Previous Research on Principles of Composition
Like lineage analysis, the general principles according to which Buddhas and
other sacred figures are placed in a Tibetan painting have received relatively little
attention until now. Again we owe the first steps to G. Tucci, who in Tibetan
Painted Scrolls devoted chapter thirteen to "The Plan of the Tankas," where he
described many key iconographic and decorative elements.11 He had little to say
about composition; besides that, the thang kas followed a similar general plan,
with many shared compositional "characters," and the main figure (gtso bo)
dominated the central space, representing the essence of the painting.
K. M. Gerasimova in her 1978 article "Compositional Structure in Tibetan
Iconography" stressed the role of the iconometry of individual figures, but she
underestimated the complexity of other elements of structure in iconic (or
"representational") paintings:
The construction of individual figures and decorative-ornamental combinations
on a flat surface actually exhausted the entire problem of the organization of space
in the representational icon. Its compositional formula consisted in the quantitative
establishment of the centre and a symmetrical grouping of the secondary
components according to a principle of simple transfer.12
Gerasimova13 described much more complexity in the structures of biographical
or narrative paintings. In reality, even for the usual iconic depictions of Buddhas
and Bodhisattvas, the subject of composition is more complicated than admitted
by either Tucci or Gerasimova. But not hopelessly so.
A Grammatical Comparison
The non-verbal signs of a Tibetan scroll painting or thang ka can be read and
interpreted almost as one would read a written text. A painting of this tradition has
its rules of "grammar," so to speak, which allow one to interpret its arrangements
systematically. As in many written languages, one can distinguish in a painting
several levels of description, such as those corresponding to letters, words, and
sentences. To follow the analogy of language and reading, the sacred figures in a
thang ka could be considered to be like the words in a language. The individual
attributes of a figure ­ i.e., iconographic elements such as colors, hand gestures,
dress, and ornamentation ­ are like the letters of the words. To determine the
correct ordering of the figures, there exist rules governing composition ­ something
like rules of syntax.
11
See G. Tucci, Painted Scrolls, 300ff. Tucci also mentions some basic principles in his chapter on
the symbolic meanings of colors and lines, 287-88.
12
See K. M. Gerasimova, "Compositional Structure in Tibetan Iconography," Tibet Journal 3, no.
1 (1978): 47.
13
Gerasimova, "Compositional Structure," 48f.
4Jackson: Lineages and Structure in Tibetan Buddhist Painting
Two Main Means for Establishing Structure
The "syntax" of a Tibetan thang ka is not self-evident and will only be
recognized by someone who can identify and classify the individual figures. But
how to identify the figures? The two main means are: iconography and inscriptions.
The first means, iconography, is adequate for identifying to which class a figure
belongs. But it is not very reliable for identifying individuals within the most
important class, namely gurus and bla mas, because their iconography was
sometimes fluid. The same master or adept (grub thob, siddha) may be shown in
different postures depending on different contexts (for example, the four forms of
Virpa corresponding to four famous episodes of his life) or according to different
painting traditions. Nevertheless, iconographic factors, such as the dress, hair or
hats of the bla mas, can be enough for a provisional first identification of a group
of figures and possibly of a few more famous individuals.
Still, doubts often remain, especially for paintings from less common traditions.
Here written evidence such as inscriptions are sometimes the only means for a
firm identification. Indeed, at the present early stage of research, Tibetan art
historians should concentrate as much as possible on paintings with readable
inscriptions. Lineages, moreover, should be analyzed with caution and
sophistication, not simplistically or uncritically, especially where the main figure
and the last two or three historical figures of a painted lineage cannot be identified.
Traditional Tibetan Classifications of Buddhist Art
Tibetan painters and learned religious masters were aware of the basic hierarchical
rules and chronological conventions expressed in paintings.14 Such rules were
important aspects of the complex and highly developed tradition of religious art
that they maintained. In traditional Tibet, art was mostly religious, and according
to Tibetan "iconological" theories recorded in treatises on sacred art (bzo rig bstan
bcos), art works were traditionally classified into three main types, each
corresponding to an aspect of Buddhahood: enlightened body, speech, or mind.
Thus the main types of sacred objects, in ascending order, are the three "supports"
(rten):
1. Bodily supports (sku rten)
2. Verbal supports (gsung rten)
3. Mental supports (thugs rten)
Bodily supports can be further divided according to their spatial extension into
two classes: (1) painted (bris), i.e., two-dimensional, objects, and (2) sculpted or
otherwise outwardly extending ('bur), i.e., three-dimensional objects. Here we
14
My remarks here are based mainly on a direct investigation of paintings, though on a few points
(especially regarding iconometrical and iconographical classes) I have been influenced by the
explanations of learned bla mas and by written bzo rig treatises.
5Journal of the International Association of Tibetan Studies 1 (October 2005)
will be concerned exclusively with bodily supports of the first type: painted artworks
(bris sku), especially with painted scrolls (thang ka; thang sku; sku thang).
Classification According to Function
Painted images can furthermore be classified according to their main function,
though this is not a traditional Tibetan classification:
1. Simple "bodily supports," which are plain iconic representations of
a divine figure
2. Narrative paintings, which place the figures within a historical or
legendary story, such as a sainťs life
3. Didactic paintings, which symbolically represent religious truths
4. Astrological diagrams, which are meant to bring luck and repel bad
fortune
5. Representations of offerings, especially offerings made to protective
deities in order to gratify and placate them
Paintings with guru lineages made up just a small portion of the first class,
though that proportion was much higher in the fourteenth through sixteenth
centuries. Thus by no means is it the case that all thang kas depicted lineages, and
complete portrayals become increasingly rare in recent centuries.
A didactic painting, the Wheel of Existence (Srid pa'i 'khor lo). D. Jackson.
6Jackson: Lineages and Structure in Tibetan Buddhist Painting
Stylistic Trends
When one examines a number of dateable paintings from the twelfth through
nineteenth centuries, one notices tremendous changes in style. Basically, the styles
evolve from older Indian styles into later, more Chinese-influenced ones. This is
almost to be expected, given the geographical location of Tibet between the two
great civilizations of Asia ­ India and China ­ and the fact that Tibet received its
original Buddhist impulses mainly from India.
Though this stylistic evolution affected the depictions of the bodies and clothes
of divine figures less markedly, its effect on the backgrounds and other decorative
details can hardly be overlooked. Among other things, the styles evolve from a
mainly red, yellow and orange color scheme, with abstract decorative designs in
the background, to a primarily green and blue palette, with more or less stereotyped
elements of Chinese-style landscapes in the background.
Earlier Indic painting with a predominately orange and red color scheme. S. Kossak and J. Singer
1998, no. 49, "Dancing Ganapati," late 13th or early 14th century, 68 x 59 cm. Private collection.
Concerning the basic conventions of figure positioning, the arrangement changed
over time from a strictly linear arrangement in straight rows and columns to a
somewhat more staggered and natural arrangement of figures in a landscape. The
earliest convention is clearly Indian, while the later developments (right-left
alternation beginning at top-center) no doubt reflect a penetration of Chinese
7Journal of the International Association of Tibetan Studies 1 (October 2005)
traditions. These changes jump out at the historian today thanks to twelve centuries
of hindsight, but in fact Tibetan Buddhist painting remained deeply conservative
and changed at a very slow pace throughout most of its history.
Later Chinese-influenced painting with blue and green color scheme. Rhie and Thurman 1999, no.
8 (247), 18th century, 84 x 61 cm. Rubin Museum
Principles Determining the Size and Placement of Figures in a
Painting
Since Tibetan Buddhist art was and remains a conservative, formal, and orderly
world, in which nothing of significance can occur by chance, what were the
organizing principles that determined its pictorial compositions? Here by
"composition" I do not mean the layout of such secondary, decorative elements
as landscape, but specifically the choice and positioning of figures. The main
8Jackson: Lineages and Structure in Tibetan Buddhist Painting
organizing or "syntactic" rules of composition are not complicated, and they can
be summed up in terms of three expressions of precedence or hierarchy:
1. A figure's status as a "main" or "minor" figure
2. To which iconographic class the figure belongs within the levels
established by 1
3. Which special rank, if any, an individual figure has within the
iconographic classes established by 2
The Hierarchy of Main and Minor Figures: Distinguishing Levels
of Priority
The first essential distinction of hierarchy is simply the determination of which
figures are of main importance and which are of lesser importance. Most paintings
contain at least the two levels:
1. Main figure
2. Minor figures: retinue ('khor) or guest deities (lha mgron)
Here for the sake of simplicity I have limited the minor figures to just one level
(II). Some thang ka paintings possess two or even more levels of lesser figures,
i.e., (III), (IV), and so forth.
A deity becomes a "main figure" or "minor figure" in a given painting according
to the immediate spiritual wishes or priorities of the devotee or patron
commissioning the work. (For a lay person, these priorities would have been
established through the advice of a religious preceptor, who might even sketch on
paper a simple plan of a painting showing the position of each deity by writing its
name where it should stand.) To put it another way, a figure is chosen as the "main
figure" (or group of main figures) of a painting simply by being of immediate
importance, for one reason or another, to the patron. For example, a deity such as
White Tr is often chosen as the main figure to ward off serious illness or other
threats to the patron's longevity.
One can immediately recognize and distinguish the members of priority-levels
I and II through differences of size and placement. A main figure is always larger,
while minor figures are smaller. Furthermore, a single main figure is usually
positioned in the middle of the painting on the central vertical axis.
A central figure can also be positioned to the right or left of the center axis
(tshangs thig), such as in some Chinese-influenced series of arhats or portraits of
masters, where the main figure sits in partial profile on a wooden throne or platform
within a landscape. This off-center positioning confused Tucci, who sought a
doctrinal or iconographical explanation for it and the "missing" lotus seat, beyond
a mere change in aesthetic preference.15 The solution is easy to see when one takes
into account the entire set of paintings. The central axis still exists: it is the vertical
15
See G. Tucci, Painted Scrolls, 301.
9Journal of the International Association of Tibetan Studies 1 (October 2005)
axis of the main painting in the middle, toward which all the main figures in the
thang kas to the right and left turn their faces. Moreover, the central vertical axis
of each lateral painting remains the aesthetic central axis around which balance is
achieved in that composition, and the main figure of that painting turns his face
toward it, too.
An artist has to establish for a painting or set of paintings relatively larger or
smaller units of measures for each priority-level. The sizes of the faces (zhal tshad)
or of the palms of the hands (these are classical units of measure, each made up
of twelve finger-widths or sor mo) are much larger for a main figure than for the
minor ones.
In theory, all figures can belong to the priority-level of the main figure (I), and
a painting can have no minor figures. But in actual practice this rarely occurs for
paintings having more than two or three figures. In most paintings, a main figure
(or group of main figures) supplies both a spiritual center of gravity and a welcome
aesthetic focus.
Hierarchy of Iconographic Classes within Each Priority-Level
In contrast to the first distinction of main versus minor figures, which in some
ways is based on personal, almost arbitrary factors, the second main distinction
has to do with a more absolutely and permanently established hierarchy: namely,
the ordering of the different iconographic classes of deities within the Tibetan
Buddhist pantheon. The sacred figures of the pantheon each belong, in fact, to one
or another relatively higher or lower class. The main classes of sacred figures
include, in descending hierarchical order:
1. Masters of the lineage
2. Tantric deities (yi dam)
3. Buddhas in sambhogakya and nirmakya forms
4. Bodhisattvas
5. Goddesses (i.e., female bodhisattvas)
6. Pratyekabuddhas; śrvakas/sthaviras
7. ka and ki (Tib. mkha' 'gro and mkha' 'gro ma), i.e., beings
of high realization associated with tantric practice
8. Wrathful protectors of the Dharma (dharmapla), e.g., Vajrapi or
Mahkla
9. Yaka deities (Tib. gnod sbyin), e.g., the four great kings, guardians
of the directions
10. Wealth-bestowing deities (nor lha), e.g. Jambhala
11. Other lesser deities (mahnga, gter bdag, etc.)
This list embodies a spiritual hierarchy. The earlier classes embody higher
realizations, while the subsequent ones embody relatively lower ones. For example,
the realization of a perfectly enlightened Buddha is higher than that of a bodhisattva
(who is, after all, still a candidate to Buddhahood), and of course it is higher than
that of a worldly deity. The tradition does, however, distinguish between ordinary
10Jackson: Lineages and Structure in Tibetan Buddhist Painting
and great bodhisattvas; great bodhisattvas such as Avalokiteśvara are considered
to have reached a Buddha-like level of realization, though they do not manifest
themselves as a Nirmakya Buddha. Another important distinction is between
those deities who have reached the level of a saint ('phags pa) and those who are
gods of the worldly sphere ('jig rten pa'i lha).
The spiritual hierarchy of the above list is expressed ritually by the order in
which such deities are invoked in the ceremonies of Tibetan monasteries. In
consonance with Vajrayna doctrine, the gurus take precedence over all else.
How is this hierarchy expressed in a painting? As before, it is shown through
size and placement, though here with some differences. The hierarchy or spiritual
precedence of one class over another is manifested, first of all, through its vertically
higher placement in the painting, relative to the other classes of the same
priority-level. A good exemplification of the hierarchy or classes is the so-called
assembly-field (tshogs zhing) type of painting.16
An assembly field incorporating all classes of deities in a hierarchic arrangement. K.-H. Everding
1993, p. 92.
16
See Karl-Heinz Everding, Tibet: Lamaistische Klosterkulturen, nomadische Lebensformen und
bäuerlicher Alltag auf dem `Dach der Welť (Köln: DuMont Buchverlag, 1993), 92; David Jackson,
Tibetan Thangka Painting: Methods and Materials, rev. ed, in collaboration with J. Jackson (London:
Serindia Publications, 1988), 35; and Valrae Reynolds, Amy Heller, and Janet Gyatso, The Newark
Museum Tibetan Collection, vol. 3, Sculpture and Painting (Newark: The Newark Museum, 1986),
169-71.
11Journal of the International Association of Tibetan Studies 1 (October 2005)
Secondly, a higher or lower status of a class is expressed through larger or
smaller physical proportions (but here, again, relative to other classes on the same
level of importance). There exists, in fact, an exact system of figural proportions
by which higher ranking classes possess larger proportions than the ones beneath
them.17 The scale of measurements (i.e., the actual length of a "face-length" or
"finger-width" unit), however, remains the same within one importance-level.
Usually there is only one main figure, and thus the division into classes only
concerns the minor figures. But occasionally thang kas contain two, three, or more
"main figures." In that case the rules of placement according to class hierarchy
operate within that superior group, too.
Hierarchies within the Same Class of Sacred Figures
The third basic hierarchical distinction, that which influences the placement of
figures within a single iconographic class on the same priority-level, does not
actually pertain in every iconographic class. Sometimes all the figures within a
class enjoy the same status, and their ordering within their class is somewhat
arbitrary, though members of established groups are often depicted according to
an established order, based, for instance, on the sequence of their appearance in a
canonical text or famous older painting that functioned as model.
But when a true hierarchy does exist, it may reflect a doctrinal superiority or a
spiritual seniority. The deities of the anuttarayoga tantras, for instance, are accorded
a higher status over those of the yoga tantras and the two still lower classes of
tantra, in accordance with the doctrinal ranking of New Translation (Gsar ma pa)
tantras. In the representation of a lineage of teaching masters, by contrast, the
order expresses the precedence of relative seniority within that lineage: a spiritually
senior figure takes precedence over a junior one. This does not necessarily mean
seniority in age (though in fact a chronological succession of older to younger
masters is the typical case). Here the decisive factor is spiritual seniority, which
is established by one master being the religious teacher of the other.
Artistically, precedence may also be shown for figures on roughly the same
vertical level by placing superior figures either closer to the center, or to the right
hand of their inferiors. Thus, for a pair of masters both shown as main figures, the
one to the right relative to the figures (i.e., to the viewer's left) has the superior
seat. Similarly, within a lineage or series, the position at the first figure's right
hand usually has precedence over that to his left, reflecting ancient Indian
conventions for showing respect and, originally, customary uses of the respective
hands for cleaner or dirtier tasks.
Special Exceptions Regarding the Guru
In a few paintings, the depictions of the patron's personal guru or the great founding
masters of his tradition have been pushed to a higher or more central position
17
See Jackson, Tibetan Thangka Painting, chapter 4 and appendix A.
12Jackson: Lineages and Structure in Tibetan Buddhist Painting
within their class (i.e., to a position indicating higher respect), motivated by special
devotion to that master. Thus a single guru or a cluster of three founding masters
may be moved to the top center, out of their expected spatial position according
to normal linear sequence.18 Sometimes the great founding masters have not only
been taken out of their normal sequence and moved to a more central position, but
they have also been depicted on a larger scale, i.e., on a higher importance level.19
There also exists at least one painting where a figure of the guru has been elevated
to a position within the highest priority-level, and in fact to a seat on the crown of
the main figure, though this is extremely rare, and the guru is here portrayed with
smaller proportions.20
Lineages and Structure
To summarize, the placement of figures is governed by hierarchic rules that operate
within three contexts according to:
1. The immediate spiritual importance of the figures for the patron,
2. The iconographic classes within a given priority-level, and
3. The relatively higher or lower position of the individual figure within
a given iconographic class
Moreover, the same hierarchical principles apply both within a single painting
and within a set consisting of numerous paintings. In the latter case, those principles
determine the arrangement of the whole set.
But what does it mean, concretely, to say that the individual figures within an
iconographic class are "positioned hierarchically"? There exist, indeed, a number
of conventions to express positions of decreasing precedence in a Tibetan Buddhist
painting, and in the course of history quite a few of them were actually employed.
The modern scholar must make sure he or she has identified in each case which
hierarchic conventions have been used. Here, a guru lineage in the painting can
be an extremely helpful clue, since it often sets the pattern for the rest of the
composition.
The Preeminent Position and Importance of Teachers
The masters of teaching lineages thus belong to the highest of all iconographic
classes. Even when depicted as "minor figures" in relation to the immediate spiritual
priorities of the patron, they still occupy spatially the highest positions in a painting.
Their presence can therefore hardly be missed at the top, or at both the top and
right and left side-columns, of many important old paintings.
18
See for instance S. Kossak and J. C. Singer, Sacred Visions: Early Paintings from Central Tibet
(New York: The Metropolitan Museum of Art, 1998), 91, no. 18 [single guru]; and 89, no. 17 [group
of three founding bla mas].
19
See Pal, Art of Tibet, 82, plate 18 [P13]. The structure of this painting has been described in Jackson,
"Identification of Individual Masters," 130-32.
20
See Kossak and Singer, Sacred Visions, 81, no. 13.
13Journal of the International Association of Tibetan Studies 1 (October 2005)
Importance of Lineage Histories for Tibetan Buddhists
Throughout much of their history, Tibetan Buddhists have demonstrated a
proclivity for depicting guru lineages. The resultant portrayals are of great
importance not only as a record of a given lineage's history and the iconographic
representation of its masters, but also, when the lineage is complete, as very
important clues for dating the paintings.
Tibetan Buddhists in other contexts of ritual and practice, too, carefully recorded
and transmitted their teaching lineages to the extent they could. In the esoteric or
Mantrayna traditions of Mahyna Buddhism, such lineages were of crucial
religious importance: the lineage gurus needed to be ritually invoked as a
preparatory step in practice. This respect for the lines of gurus contributed to a
deep and very concrete sense of history among many Tibetan Buddhist tantric
masters ­ in contrast to the usual scholar-monks or dge bshes, whose main training
consisted in the systematic study of non-tantric doctrine through memorization
and debate and who were typically less textually and historically oriented.21
Because lineages were so important for Tibetan Buddhist practice, individual
masters would write down the particular lineages for Tantric teachings that they
had received from their various teachers. The resulting books often consisted of
little more than bare lists of masters' names and the titles of books or teachings,
yet in Tibetan literature they made up a genre of writing called "teachings received"
(thob yig or thos yig).22 Artistically this same attention to lineages was expressed
in the careful portrayals of many lineages of gurus. Painted lineages are an artistic
expression of the same concern that also finds its expression ritually in the recitation
of lineage prayers (brgyud 'debs/bla ma'i brgyud pa la gsol ba 'debs pa).
Special Chronological Conventions for Lineages
As elsewhere in the planning of a painting, so too in the portrayal of lineages one
usually finds an orderly and exact system at work. Indeed, painted lineages can be
read chronologically, and thus interpreted as historical records. But for a correct
interpretation one needs to determine in each case which particular convention of
chronological descent has actually been used. For this, it is best to begin by:
1. Identifying the starting point. Then it is much easier to follow the
continuation of the lineage and thus determine
2. The convention used for depicting the temporal sequence of the
subsequent figures
21
Here it is interesting to compare the remark of Dan Martin, Tibetan Histories: A Bibliography of
Tibetan-Language Historical Works (London: Serindia, 1997), 15: "The nearly universal concern of
Tibetan religious schools for `lineage' is a highly historical sort of preoccupation."
22
On the thob yig genre, see Jan-Ulrich Sobisch, "The `Records of Teachings Receiveď in the
Collected Works of A mes Zhabs: An Untapped Source for the Study of Sa skya pa Biographies," in
Tibet, Past and Present: Tibetan Studies I, ed. H. Blezer (Leiden: Brill, 2002), 59-77.
14Jackson: Lineages and Structure in Tibetan Buddhist Painting
Main Conventions Regarding the Starting Point
Lineages usually begin, as would be expected, with the earliest teacher. Thus
they almost always begin with a Buddha, who, for the tantric traditions, is the
tantric original-guru. This primordial teacher is Vajradhara for the New Translation
Tibetan tantric schools, a blue-colored Buddha in sambhogakya form holding a
vajra and bell in his crossed hands, while for tantras of the Old Translation (Rnying
ma) school, he is the original Buddha Samantabhadra. For non-tantric traditions,
one can usually expect as the starting point the historical Buddha, Śkyamuni (in
nirmakya form).
Vajradhara, the primordial Buddha. Central figure in "Vajradhara with adepts and arhats," Rhie
and Thurman 1999, no. 168 (665), 55.9 x 55.9 cm. Rubin Museum. DETAIL ONLY.
Where this first figure sits indicates the beginning of the lineage. There existed
in fact several artistic conventions regarding the starting point of lineages, but the
two most common starting points are:
1. The top left corner (relative to the viewer, which is the top right
corner, relative to the deities). This seems to be the oldest convention,
and it is well suited to paintings where the figures are arranged in
straight rows and columns.
2. The top central position. This has become, since about the early
16th century, the most common convention, though it occurs in a few
earlier paintings, too. It is suited to figures placed in a more realistic
landscape (which at the top of the painting means in the sky).
15Journal of the International Association of Tibetan Studies 1 (October 2005)
Exceptions do exist, such as the special case where the first figure sits just to
the left (relative to the viewer) of the top central position. Here another figure (the
patron's own guru) has for reasons of special respect usurped the top central
position.23
Conventions of Descent
Let us turn now to some of the main conventions for portraying lineal descent.
Once the beginning of the lineage has been located, it is normally not difficult to
see how the lineage continues, whether straight across or down columns, or in an
alternating fashion. There are several conventions for each starting point:
Descent Starting from the Top Left
Straight across the top from left to right (for short lineages)
Among the various conventions of descent from the original Buddha Vajradhara
at top left, one of the oldest and simplest is to proceed straight across the top row,
from the viewer's left to right.24 This is well suited for lineages of up to about
eleven or twelve figures, though in special circumstances it can be stretched to
fifteen or even more.25 The structure can be shown:
13121110987654321
23
See, for instance, Pal, Tibetan Paintings: A Study of Tibetan Paintings, Eleventh to Nineteenth
Centuries (Basel: Ravi Kumar/Sotheby Publications, 1984), plates 29 and 30.
24
Chayet, Art et Archéologie, 189, noted such a convention among early thang kas.
25
See Marylin Rhie and R.A.F. Thurman, Wisdom and Compassion: The Sacred Art of Tibet (New
York: Henry N. Abrams, 1991), 231, no. 75; and Pal, Art of the Himalayas, 152, no. 85.
16Jackson: Lineages and Structure in Tibetan Buddhist Painting
Lineal descent starting at the top left, continuing straight across from left to right. "Mandala of
Rakta Yamari," Pal 1991, p. 152, no. 85. 95.3 x 76.2 cm. Zimmerman collection.
Diagram of the "Mandala of Rakta Yamari"
17Journal of the International Association of Tibetan Studies 1 (October 2005)
Straight across the top from left to right, and the same once again in a second
horizontal row placed beneath the first
When there are too many figures to be easily depicted in the first row, one
solution (rarely used) is to add another row just below the first.26
1110987654321
17?16?*15?*14?d3d2d115141312
An extreme example of this structure is found in a Bon po thang ka documented
by S. G. Karmay in his excellent study, The Little Luminous Boy: The Oral
Traditions from the Land of Zhangzhung Depicted on Two Tibetan Paintings,27
where the series of eighty-six figures continues in this way in eleven consecutive
horizontal rows of up to ten figures each, interrupted only by the larger central
figure:
10987654321
20191817161514131211
30292827262524232221
3837363534333231
4645444342414039
525150494847
585756555453
etc.
26
See Pal, Art of the Himalayas, 149, no. 83, where the three central figures of the next row, however,
are deities, and it may be that figures 14 and 15 appear twice (otherwise the last four figures are 16,
17, 18 and 19).
27
Samten G. Karmay, The Little Luminous Boy: The Oral Traditions from the Land of Zhangzhung
Depicted on Two Tibetan Paintings (Bangkok: Orchid Press, 1998), 3, caption 1.
18Jackson: Lineages and Structure in Tibetan Buddhist Painting
Straight across the top row from left to right, and the same once again in subsequent rows. S. G.
Karmay 1998, p. 2, caption 1, size not mentioned. Driesch collection, Cologne.
Straight across the top from left to right, and down one column
When the lineage consists of more than twelve or thirteen masters, another solution
for the placement of excess figures is to run the continuation from the upper right
corner down the right-hand column. A good example of this is the painting of Ngor
chen kun dga' bzang po's (1382-1456) ordination lineage in the Zimmerman
collection.28
87654321
9
10191718
11
12
13
142120
15
16
28
See Pal, Art of the Himalayas, 155, no. 87.
19Journal of the International Association of Tibetan Studies 1 (October 2005)
Straight across the top row from left corner to right, then down the right column. Ngor chen and
Mus chen and their ordination lineage. Pal 1991, p. 155, no. 87, 87.6 x 80 cm. Zimmerman collection.
Diagram of "Ngor chen kun dga' bzang po and Mus chen and their ordination lineage."
20Jackson: Lineages and Structure in Tibetan Buddhist Painting
Straight across the top to the right, and then alternating
This is a variation on the same beginning, in which the top row begins on the
far left and progresses to the right, before finally alternating between left and right
columns. A good example is found in Rhie and Thurman.29
87654321
109
1211
1413
etc.15
Straight across the top to the right, and then alternating.
Left to right, interrupted in the middle
An interesting variation of the left-to-right sequence is found in certain old "Stag
lung pa" paintings in which the sequence has been interrupted in the middle by a
centrally placed figure or figures representing the guru of the main figure or the
three preceding gurus (founders of the tradition).30 Here a single guru, figure no.
7 ­ Phag mo gru pa (1110-1170), shown with a heavier beard ­ has been moved
to a central position over the main figure ­ no. 8, Stag lung thang pa (1142-1210).
The lineage begins:
6547321
29
Rhie and Thurman, Wisdom and Compassion, 221, no. 70.
30
See Singer, "Taklung Painting," 55, plate 37; and Singer, "Painting in Central Tibet," 126, plate
25.
21Journal of the International Association of Tibetan Studies 1 (October 2005)
Left to right, interrupted in the middle. Here a single guru has been moved to a central position
over the main figure. Collection A. Bordier.
Another such case is found in Singer plate 41,31 a painting probably dating to
the last two decades of the life of its main subject, no. 11a, Sangs rgyas dbon po
(1251-1296). The composition here groups three important gurus ­ the three
founding masters of Stag lung ­ in the top center.
6541089321
7-s3--s2-11b
11a
31
Singer, "Taklung Painting," 59.
22Jackson: Lineages and Structure in Tibetan Buddhist Painting
Descent Starting from the Top Center
From the center, across the top to the left and down the left side column, then
back to the top center, from which then across to the top right, and down the
right column
An example is the Mahkla Pajarantha painting in Rhie and Thurman.32 This
painting's lineage begins with Vajradhara at the top middle, progresses three figures
to the viewer's left, and then drops down the left column. Then it returns to figures
no. 10 and no. 11, the first lay adherent and paita in the top row ­ Rje btsun
grags pa rgyal mtshan (1147-1216) and Sa skya pai ta kun dga' rgyal mtshan
(1182-1251) ­ goes right, and finally descends down the right column. The structure
is as follows:
131211101234
145
156
167
178
189
Note that the iconography of figure no. 10, Rje btsun grags pa rgyal mtshan,
actually corresponds to the usual later depictions of his father, Sa chen kun dga'
snying po (1092-1158), who here should be in position 8. Could we here have Sa
chen kun dga' snying po then in an anomalous position (10), owing to the great
veneration paid him by the tradition? Or is Sa chen kun dga' snying po in position
8, and this is just a case of more fluid iconography?
From the center across to left, then across to right (and then alternating below)
An example is Rhie and Thurman no. 77.33
1110987123456
13?15?14?12?
32
Rhie and Thurman, Wisdom and Compassion, 222, no. 71.
33
Rhie and Thurman, Wisdom and Compassion, 234.
23Journal of the International Association of Tibetan Studies 1 (October 2005)
From the center across to left, across to right, then back to top center
In this painting the lineage jumps to the top center, before descending to figure
no. 8, the main figure, Stag lung thang pa. Here the top center has been reserved
for the guru of the founder.34
6547123
From the center first to the left, then to the right, then back to top center, before descending to the
main figure. The top center is reserved for the guru of that main figure. "Portrait of Taklung
Thangpa Chenpo," S. Kossak and J. Singer 1998, no. 18, 47 x 37 cm. Private collection.
34
See M. Rhie and R. Thurman, Weisheit und Liebe: 1000 Jahre Kunst des tibetischen Buddhismus
(Bonn: Kunst-und Ausstellungshalle der Bundesrepublik Deutschland, 1996), 449, no. 203 (84a).
24Jackson: Lineages and Structure in Tibetan Buddhist Painting
Alternating to the left and right, all the way down
This has been the most widespread convention since the sixteenth century, and
it has practically replaced all other conventions.35 A good example is Rhie and
Thurman no. 64,36 showing Paita Gayadhara (as master of the Path with Its Fruit)
with surrounding lineage.37 Here the sequence is: 1. top center, 2. his right, 3. his
left, and so on. Thus the structure of the top row and the next few lines is:
975312468
1110
1312
1514
etc.16
Beginning at the top center, then alternating to the left and right, all the way down. The most
widespread convention since the sixteenth century. Mus chen, the second abbot of Ngor (as master
of the Path with Its Fruit) surrounded by another lineage. 78.5 x 67.5 cm.Collection Barbara and
Walter Frey, Zurich (F758).
35
Chayet, Art et Archéologie, 189, mentions this order beginning at the top, center, and alternating
to both sides as a usual later convention. She records a tradition according to which this convention
was introduced in the Sa skya pa school by Theg chen chos rje kun dga' bkra shis (1349-1425). This
suggestion is interesting because that master visited the Ming court in the first decade of the fifteenth
century, and brought back to Tibet much Chinese Buddhist art, which was later much admired and
even taken as models.
36
Rhie and Thurman, Wisdom and Compassion, 207.
37
The identical structure of another thang ka from this series has been described in Jackson,
"Identification of Individual Masters," 138f. For a still later painter with the same basic structure, see
Pal, Art of Tibet, 88, plate 24 [P21], a painting described in Jackson, "Identification of Individual
Masters," 139-41.
25Journal of the International Association of Tibetan Studies 1 (October 2005)
The above types cover probably ninety-nine percent of existing thang kas. But
they do not exhaust all possibilities. In fact, one should keep one's eyes open for
further variations expressing the same basic principles. Two recently analyzed
thang kas, for instance, were found to embody concentric structures:
Concentric, around a central deity
In this instance,38 the lineage masters are arranged in semi-circles to the right and
left of the central deity, Hevajra (H):
93128
115H410
137612
1514
1716
1918
38
Kreijger, Tibetan Painting, no. 56.
26Jackson: Lineages and Structure in Tibetan Buddhist Painting
Concentric, around a central deity. In this instance, the lineage masters are arranged in semi-circles
to the right and left of the central deity above, Hevajra. "Panjaramahakala," H. Kreijger 2001,
No. 56, 70 x 51 cm.
The last bla ma (19) is seven generations after 'Gro mgon chos rgyal 'phags pa
blo gros rgyal mtshan (1235-1280), i.e., he brings the lineage down to about Ngor
chen kun dga' bzang po's time (ca. mid-15th century). But the style seems to
indicate a date a century or more later than Ngor chen kun dga' bzang po
(1382-1456).
Concentric, around a maala
The painting39 embodies another concentric structure, but here the center is the
maala of the deity Vajravidraa:
39
Kreijger, Tibetan Painting, no. 66.
27Journal of the International Association of Tibetan Studies 1 (October 2005)
13112
153214
175416
76
98
1110
1918
2120
2322
2524
2726
292830
Figure no. 24 in the lineage is Ngor chen kun dga' bzang po. Figure no. 30 is
the final guru of the lineage, and not (as one might otherwise expect) the
commissioning patron. He is the Ngor abbot Sangs rgyas seng ge (1504-1569).
The final generations are:40
25. nanda (=Ngor chen kun dga' bzang po)
26. Krti (=Gu ge pa chen grags pa rgyal mtshan)
27. (Glo bo mkhan chen bsod nams lhun grub [1456-1532]?) Legs pa'i
'byung gnas
28. (Sa lo 'jam dbyangs) Kun dga' bsod nams
29. Kun dga' lhun grub
30. Ye shes rgyal mtshan
The tradition portrayed is Vajravidraa in the tradition of Jayavarma and
Jnaśr, with a 19-deity maala for its initiation.
Quirks and Complications
First impressions can deceive when interpreting lineages. To avoid error, one
should try first of all to determine whether the painting depicts a complete lineage
of masters or just a fraction.
A Single Lineage Ending with the Main Figure
Most paintings end the lineage with the last minor figure. But in paintings portraying
bla mas as main figures, a few compositions end the lineage by jumping from the
last teacher pictured among the minor figures to the main, central figure. In this
case, the main figure (or sometimes a pair of central figures) and the minor figures
together represent a single complete lineage.41
40
See Ngor chen kun dga' bzang po, Thob yig rgya mtsho, in Bsod nams rgya mtsho, ed., Sa skya
pa'i bka' 'bum (Tokyo: Toyo Bunko, 1968), 9: 84.1-2.
41
This was probably the case in Pal, Tibetan Paintings, plates 35 and 41; and Pal, Art of the Himalayas,
155, no. 87.
28Jackson: Lineages and Structure in Tibetan Buddhist Painting
Combinations of Partial and Complete Lineages
Sometimes the same painting portrays a small fraction of one lineage
(represented by a single main figure) and another complete lineage (represented
by a series of smaller minor figures). Here the painting belongs to a multiple-thang
ka set, and the lineage of the main figure is continued by the several other paintings
in the same set. It was also possible to depict a still larger part of an incomplete
lineage as the main figures ­ say two or four masters of the Sa skya pa Path with
Its Fruit (Lam 'bras) instructions ­ and to surround them with minor figures
representing a complete lineage of gurus of another teaching line.42
A further possibility was to depict parts of two separate lineages in a single
painting: one partial lineage consisting of multiple main figures, and the other as
multiple minor figures. Illustrating this is a set of thang kas each showing four
Path with Its Fruit lineage masters and a half of a second (unrelated) lineage.43
Partial major and partial minor lineages in one thang ka. Four Path with Its Fruit lineage masters
and a half of a second (unrelated) lineage. 72.5 x 68 cm. Collection Barbara and Walter Frey,
Zurich (F753).
42
See for instance Jackson, A History, 81, figure 24.
43
See Pal, Art of Tibet, 84, plate 20 (P15). The structure of this painting has been described in Jackson,
"Identification of Individual Masters," 132-36.
29Journal of the International Association of Tibetan Studies 1 (October 2005)
Yet another possible complication is that occasionally a lineage branched or
forked, and where this was thought significant by the patron, it could be shown by
separately portraying both of the two branch lineages in question.44 Finally, it may
happen that two distinct lineages are depicted by the minor figures. This, too, can
be viewed as a type of branching; here the split occurs already from the original
teacher (e.g. Vajradhara). For example, a thang ka can depict two series of minor
figures descending to the right and left, each series representing a distinct minor
lineage.45 Or two complete lineages can be shown, each with a separate progenitor.46
As one approaches the most recent two or three centuries, moreover, one finds
that the lineages are hardly ever painted in complete form. This does not mean that
the lineages have become less important religiously, but only that they have become
too long to be easily represented in paintings. Instead of a complete lineage, it is
now usual to depict a selection of the greatest founding masters and transmitters
of the teaching. But this makes the few paintings of complete lineages from the
eighteenth through twentieth centuries all the more important.47
Two distinct Lineages. Partial major and two complete minor lineages. Ngor chen kun dga' bzang
po, Mus chen and two Ngor pa abbots, with two complete lineages, one to the right and the other
to the left. Collection A. Bordier.
44
See Pal, Art of Tibet, 82, plate 18 (P13). The structure of this painting has been described in Jackson,
"Identification of Individual Masters," 130-32.
45
See Rhie and Thurman, Weisheit und Liebe, 440, no. 192 [16a].
46
As in Reynolds et al., Newark Museum, plate 11 (P12).
47
See, for instance, the important recent 'Brug pa lineage in Pal, Tibetan Paintings, plate 95 (inscribed).
Recent 'Bri gung pa lineage thang kas are not that rare. See Jackson, A History, 343, plate 64; and
Jackson, "Lama Yeshe Jamyang," 153-76.
30Jackson: Lineages and Structure in Tibetan Buddhist Painting
Standard Groups that Look like Lineages
One has to be careful, moreover, not to identify as teaching lineages all
similar-looking arrangements of Indian masters and Tibetan bla mas. Near the start
of the teaching lineage, such figures can, of course, be early members of the lineage.
If they appear above (i.e., prior to) the Tibetan bla mas, then one can provisionally
assume that they belong to the lineage. But when they appear as a group below
Tibetan bla mas or elsewhere, one must exercise caution. A row or descending
column of Indian paitas or siddhas might in fact be some standard arrangement
of Indian masters such as the "Six Ornaments and Two Best Ones" (Rgyan drug
mchog gnyis), "Eight Great Adepts" (Grub chen brgyad), or a fraction of the
"Eighty-four Adepts," and not the start of a lineage.
Systematic inclusion of charts and complete lists of figures has the advantage
of forcing the investigator to confront unusual features. For instance, repeated or
missing teachers of the lineage should not be passed over without at least some
attempt at explanation. Though all bla mas can provisionally be assumed to have
been either members of the lineage or (as a final figure) the patron, the presence
of unknown or unexpected Indian masters (e.g. siddhas or paitas) must be
explained.
Depictions of trülku (sprul sku) Lineages
Since the sixteenth and seventeenth centuries, moreover, there is the increasing
danger that a series of previous lives ('khrungs rabs) of a reincarnate lama (sprul
sku) might have been depicted where one would expect a guru lineage.48 Above
him in the sky are four bla mas, including at least three of his previous
reembodiments as the Zhwa dmar.
48
See, for example, Rhie and Thurman, Wisdom and Compassion, 251, no. 87, where the main figure
is a Zhwa dmar rin po che ­ probably the fourth, Chos grags ye shes (1453-1524).
31Journal of the International Association of Tibetan Studies 1 (October 2005)
Depictions of the series of previous lives ('khrungs rabs) of a reincarnate lama (sprul sku). The
Fourth Zhwa dmar rin po che Chos grags ye shes (1453-1524) and his predecessors. Rhie and
Thurman 1991, p. 251, no. 87, and Rhie and Thurman 1999, plate 3, p. 85, 62.2 x 47 cm. Zimmerman
collection.
Another striking example is Rhie and Thurman no. 51,49 "Nyingma Lama,"
where the main figure is in fact the great seventeenth-century master and historian
of Sa skya Monastery, A myes zhabs ngag dbang kun dga' bsod nams (1597-1659),
and the minor figures portray his successive previous lives. The lineage of figures
contains a few obvious historical gaps, which makes it doubtful at first glance that
an unbroken lineage of masters and disciples was depicted. This series raises
difficulties even as a list of a reincarnate lama's rebirths, since several of the lives
overlap conspicuously ­ Bla ma dam pa bsod nams rgyal mtshan (1312-1375) and
Theg chen chos rje kun dga' bkra shis, for instance, being uncle and nephew. Does
this reflect sloppy historical scholarship on the part of the person who first
"retraced" the lineage of previous lives? It was, however, doctrinally conceivable
that an enlightened master could have manifested two different bodily forms or
lives simultaneously, though this explanation was somewhat inelegant.
49
Rhie and Thurman, Wisdom and Compassion, 184.
32Jackson: Lineages and Structure in Tibetan Buddhist Painting
Depictions of Several Teachers of the Same Generation
Another possibly deceptive arrangement is the placement of several teachers,
many from the same generation, above the head of the central figure. In this case
no single lineage is portrayed, though at first glance it resembles a lineage depiction:
for instance, the depiction of Sa chen kun dga' snying po with several lineage and
direct teachers in Rhie and Thurman no. 61.50 The figures portrayed are:
1. Rdo rje gdan pa
2. Bo ra rgyal?
3. Bal po dzny na badzra
4. Pu rang lo chung
5. Rngog lo ts ba
6. Brang sti dar ma snying po
7. Khyung rin chen grags
8. Lang dkon pa
9. 'Jam dpal
10. Ba ri lo ts ba
11. 'Bir ba pa
12. Mkhon sgyi chu ba
13. Snam kha'u pa
14. Mkhon dkon mchog rgyal po
15. Se mkhar chung ba
16. Mal lo ts ba
17. Byang chub sems dpa'
18. Mes lha[ng] tshe
50
Rhie and Thurman, Wisdom and Compassion, 201.
33Journal of the International Association of Tibetan Studies 1 (October 2005)
121081
131192
143
154
165
176
187
Depictions of several teachers of the same generation. "Portrait of Sa chen kun dga' snying po,"
Sa chen kun dga' snying po with several lineage and direct teachers, not a single lineage. Rhie and
Thurman 1991, p. 201, no. 61, and S. Kossak and J. Singer 1998, no. 51, 114.3 x 94 cm. Private
collection.
Up to figure no. 5 they are his lineage gurus, and after that all (except maybe
no. 8) are his direct teachers. Figures no. 9 and 11 taught him in visions.
34Jackson: Lineages and Structure in Tibetan Buddhist Painting
A lama with his teachers. Sa skya lo ts ba 'jam dbyangs kun dga' bsod nams surrounded by his
twelve teachers and the Buddha Vajradhara. H. Kreijger 2001, p. 78, no. 24, "Sakya Master," 90
x 70 cm. Jucker Collection.
See also H. Kreijger no. 24, "Sakya Master."51 This striking monochrome gold
painting portrays as its main figure the great sixteenth-century master Sa skya lo
ts ba 'jam dbyangs kun dga' bsod nams (1485-1533) of the Sa skya dus mchod
Palace, the twenty-third throne holder of Sa skya, tenure 1496-1533. He is shown
surrounded by his twelve teachers and the Buddha Vajradhara.
53124
9768
1110
1312
51
Kreijger, Tibetan Painting, 78.
35Journal of the International Association of Tibetan Studies 1 (October 2005)
15Main
Figure
14
1. [Vajradhara]
2. Rje dkon mchog 'phel
3. Bdag chen chos [rje]
4. 'Jam dbyangs shes rab rin chen
5. Grub chen phyag rdor ba
6. 'Khrul zhig tshul khrims rgyal mtshan
7. Mkhan chen kun blo ba
8. Rje ... ... Dge ba
9. Sngags 'chang ... ... Dpal bzang
10. Shkya seng [ge]
11. Chos rje yon tan pa
12. Gdong skyes pa
13. Ser chen chos bzangs pa
14. Pa chen grags pa rdo rje
15. Glo bo mkhan chen
To repeat, the twelve bla mas do not form a lineage. As a young boy, Kun dga'
bsod nams's first two main teachers were (14) Mi nyag paita grags pa rdo rje
(d. 1491) and (15) Glo bo mkhan chen bsod nams lhun grub (1456-1532), who are
portrayed as youthful bla mas to his right and left. I assume that this exquisite gold
thang ka was commissioned in the great bla ma's honor by one of his main students
either in the last decades of his life or after he died in 1533 at the relatively young
age of 48.
See H. Kreijger no. 18, "Portraits of Six Masters."52 Painting with such
arrangements are rare. They do not indicate any diminishing of the importance of
complete lineages by the time they were painted.
The Historical Accuracy of the Lineages Depicted
Generally speaking, the painted depictions of lineages in carefully executed thang
kas are an accurate representation of contemporary knowledge and opinion about
the particular lineages. Especially for the most recent generations they portray,
they can be trusted as a fairly reliable historical record, as can sometimes be
confirmed by checking the parallel written sources. In many cases the bla mas
planning the painting must have based themselves on the best available written
sources. Sometimes, now, the contents of inscribed painted lineages serve as a rare
record of otherwise unattested lineages.
The above comments about historical accuracy, however, mainly refer to the
Tibetan portions of lineages. It is possible, especially in very long Indian lineages,
that the early Indian segments embody semi-legendary or even legendary materials
52
Kreijger, Tibetan Painting, 66.
36Jackson: Lineages and Structure in Tibetan Buddhist Painting
of limited historical value. Still, one should investigate carefully at least the last
few Indian generations, and one should not dismiss out of hand all references to
Indian masters.
Conclusions
Depictions of lineages are thus a valuable key for students of Tibetan art. They
can help unlock the overall structure, and thus meaning, of many paintings. If
accurately interpreted, lineages can be important for a better understanding of the
art history, iconography, and even religious culture of Tibet in general.
The Chronological Significance of the Latest Figure
A dearth of precisely datable thang kas has long plagued historians of Tibetan
painting. Comparison of stylistic elements often allows a provisional dating to
within two or three generations. But many paintings contain additional evidence
that can be used for a more exact dating, such as written inscriptions that identify
individual gurus, the bla ma who performed the consecration, or even the
commissioning patron.
Even without inscriptions, the presence of an identifiable guru lineage often
makes possible a more exact dating. If the lineage is complete and has been properly
interpreted, the identity of its last figure enables an approximate dating of the
patron to within about one generation.53 This sounds inexact, but being based on
internal evidence relating to identifiable historical people the method has several
advantages over the more hypothetical conclusions obtained through mere stylistic
similarity.54
Yet one must be careful not to mistake, for instance, a partial lineage for a whole
one. One must also be careful where one or two bla mas as the main central figure
or figures may represent the final figure(s) of the lineage, thus bringing the lineage
forward another generation or two. It is essential to try to identify and date as many
figures from the end of the lineage as possible, for despite the other complexities
of this method, one point is simple: a painting cannot have been painted before the
latest historical figure it portrays.
Most historians of Tibetan art would agree that a description of inscribed lineages
is essential for documenting in detail the paintings that contain them. Ideally, the
documentation should include all four steps:
1. Deciphering the names
2. Identifying and dating the individual masters
53
For a more detailed discussion of the chronological possibilities of this method, see Jackson,
"Dating of Tibetan Paintings," 91-112.
54
It will be possible in some paintings to test the accuracy of dating by lineage (and to confirm
whether the patron belonged to the generation after the last guru of the lineage), namely in those thang
kas that possess both an inscribed lineage and inscriptions mentioning the name of the patron and the
occasion for the painting being commissioned.
37Journal of the International Association of Tibetan Studies 1 (October 2005)
3. A. Identification of the entire lineage, and B. listing the names in the
chronological order of the lineage
4. Diagramming the relative positions of each figure, with numbers
corresponding to the list of step 3.
For structural analysis the job is not finished until step 4 is completed. In this
article I have concentrated on showing as many possible instances of that step
without furnishing for each thang ka the steps leading up to it.55
The Wider Significance of Lineages
Paintings of guru lineages bear witness to what seems to be a special feature of
Tibetan (especially Tantric) Buddhism and even Tibetan culture in general: a strong
sense of concrete tradition and history. The fastidious care paid by generation after
generation of Tibetans to recording actual lineages in art as well as in ritual practice
and similar written lineage records is, as far as I can judge, special within the Asian
Buddhist cultural realm. Though rooted in Indian concepts of the guru lineage,
these Tibetan expressions of lineage have few close parallels known to me elsewhere
in the world. Given the importance of painted lineages in these and other respects,
one can only hope that historians of Tibetan art will devote to them the careful
attention they deserve.
Bibliography
Béguin, Gilles, ed. Dieux et démons ďHimâlaya: Art du Bouddhisme lamaique.
Paris: Réunion des Musées Nationaux, 1977.
Chayet, Anne. Art et Archéologie du Tibet. Paris: Picard Éditeur, 1994.
Chogay Trichen. Gateway to the Temple: Manual of Tibetan Monastic Customs,
Art, Building and Celebrations. Bibliotheca Himalayica, Series III, vol. 12.
Translation by D. P. Jackson of bCo brgyad khri chen, Thub bstan legs bshad
rgya mtsho [=Ngag dbang mkhyen rab legs bshad rgya mtsho], Bstan pa'i rtsa
ba.... Kathmandu: Ratna Pustak Bhandar, 1979.
Essen, G.-W., and T. T. Thingo. Die Götter des Himalaya: Buddhistische Kunst
Tibets. Die Sammlung Gerd-Wolfgang Essen. 2 vols. Munich: Prestel-Verlag,
1989.
Everding, Karl-Heinz. Tibet: Lamaistische Klosterkulturen, nomadische
Lebensformen und bäuerlicher Alltag auf dem `Dach der Welť. Köln: DuMont
Buchverlag, 1993.
Galerie Robert Burawoy. Peintures du monastre de Ngor. Libourne: Arts
graphiques ďAquitaine, 1978.
55
I have given more complete documentation of numerous thang kas in recent publications, and plan
to present more in the future.
38Jackson: Lineages and Structure in Tibetan Buddhist Painting
Gerasimova, K. M. "Compositional Structure in Tibetan Iconography." Tibet
Journal 3, no. 1 (1978): 37-51.
Huntington, John C., and Dina Bangdel. The Circle of Bliss: Buddhist Meditational
Art. Columbus: Columbus Museum of Art, and Chicago: Serindia Publications,
2003.
Jackson, David. Tibetan Thangka Painting: Methods and Materials. Rev. ed. In
collaboration with J. Jackson. London: Serindia Publications, 1988.
------. "A Painting of Sa-skya-pa Masters from an Old Ngor-pa Series of Lam
'bras Thangkas." Berliner Indologische Studien 2 (1986): 181-91.
------. "The Identification of Individual Masters in Paintings of Sa-skya-pa
Lineages." In Indo-Tibetan Studies: Papers in Honour and Appreciation of
Professor David Snellgrove's Contribution to Indo-Tibetan Studies, vol. 2,
edited by T. Skorupski, 129-44. Tring, U.K.: Buddhica Britanica, 1990.
------. "Apropos a Recent Tibetan Art Catalogue." Review of Wisdom and
Compassion, by M. M. Rhie and R. A. F. Thurman. Wiener Zeitschrift für die
Kunde Südasiens 37 (1993): 109-30.
------. A History of Tibetan Painting: The Great Painters and Their Traditions.
Beiträge zur Kultur- und Geistesgeschichte Asiens, No. 15. Vienna: Verlag
der Österreichischen Akademie der Wissenschaften, 1996.
------. "Lama Yeshe Jamyang of Nyurla, Ladakh: The Last Painter of the 'Bri
gung Tradition." In E. Lo Bue ed., art issue of the Tibet Journal 27 (2002):
153-76.
------. "The Dating of Tibetan Paintings is Perfectly Possible­Though Not
Always Perfectly Exact." In Dating Tibetan Art: Essays on the Possibilities
and Impossibilities of Chronology from the Lempertz Symposium, Cologne,
2001, edited by I. Kreide-Damani, 91-112. Wiesbaden: Dr. Ludwig Reichert
Verlag, 2003.
Karmay, Samten G. The Little Luminous Boy: The Oral Traditions from the Land
of Zhangzhung Depicted on Two Tibetan paintings. Bangkok: Orchid Press,
1998.
Kossak, S. and J. C. Singer. Sacred Visions: Early Paintings from Central Tibet.
New York: The Metropolitan Museum of Art, 1998.
Kreijger, Hugo. Tibetan Painting: The Jucker Collection. London: Serindia
Publications, 2001.
Luczanits, Christian. "Art-Historical Aspects of Dating Tibetan Art." In Dating
Tibetan Art: Essays on the Possibilities and Impossibilities of Chronology
from the Lempertz Symposium, Cologne, 2001, edited by I. Kreide-Damani,
25-57. Wiesbaden: Dr. Ludwig Reichert Verlag, 2003.
39Journal of the International Association of Tibetan Studies 1 (October 2005)
Martin, Dan. Tibetan Histories: A Bibliography of Tibetan-Language Historical
Works. London: Serindia, 1997.
Pal, Pratapaditya. Art of Tibet: A Catalogue of the Los Angeles County Museum
of Art Collection. Rev. ed. Los Angeles: University of California Press, 1990.
------. Tibetan Paintings: A Study of Tibetan Paintings, Eleventh to Nineteenth
Centuries. Basel: Ravi Kumar/Sotheby Publications, 1984.
------. Art of the Himalayas: Treasures from Nepal and Tibet. New York:
Hudson Hills Press, 1991.
Reynolds, V., A. Heller, and J. Gyatso. The Newark Museum Tibetan Collection.
Vol. 3, Sculpture and Painting. Newark: The Newark Museum, 1986.
Rhie, Marylin M. Weisheit und Liebe: 1000 Jahre Kunst des tibetischen
Buddhismus. Bonn: Kunst- und Ausstellungshalle der Bundesrepublik
Deutschland, 1996.
Rhie, Marylin M., and Robert A. F. Thurman. Wisdom and Compassion: The
Sacred Art of Tibet. New York: Henry N. Abrams, 1996.
------. Worlds of Transformation: Tibetan Art of Wisdom and Compassion. New
York: Henry N. Abrams, 1999.
Singer, Jane Casey. "Painting in Central Tibet, ca. 950­1400." Artibus Asiae 54,
no. 1/2 (1994): 87­136.
------. "Taklung Painting." In Tibetan Art: Towards a Definition of Style, edited
by J. Singer and P. Denwood, 52-67. London: Laurence King, 1997.
Sobisch, Jan-Ulrich. "The `Records of Teachings Receiveď in the Collected
Works of A mes Zhabs: An Untapped Source for the Study of Sa skya pa
Biographies." In Tibet, Past and Present: Tibetan Studies I, edited by H. Blezer,
PIATS 2000: Tibetan Studies: Proceedings of the Ninth Seminar of the
International Association for Tibetan Studies, Leiden 2000, ed. Henk Blezer
and Abel Zadoks, vol. 2/1, Brilľs Tibetan Studies Library, ed. Henk Blezer,
Alex McKay, and Charles Ramble, 59-77. Leiden: Brill, 2002.
Tanaka, Kimiaki. "The Usefulness of Buddhist Iconography in Analysing Style
in Tibetan Art." Tibet Journal 21, no. 2 (1996): 6-9.
40Jackson: Lineages and Structure in Tibetan Buddhist Painting