Realisations: Reflections on Techniques in Early Central Tibetan Painting
by Robert Bruce-Gardner


The preparation of paint was elaborate and laborious, since the various pigments had differing requirements for grinding, cleansing of impurities, and binding. Apart from the organic colors - indigo and red lac - the pigments were minerals that were in common, almost universal, use. Most remarkable is the exclusion of lapis lazuli, which has not been identified in any of these early paintings, nor has it been recorded in later Tibetan art. The generally coarsely ground azurite and malachite, the blue and green derived from the same cuprous mineral, were the most critical to prepare properly for use as pigments. After their Separation and the removal of their many impurities, care had to be taken when grinding, for as the particles became smaller, they lost their intensity of color. This accounts for the usually thicker layers in these color fields and the matte appearance of their granular surfaces, which scatter the light, and for the vulnerability of these areas, as the large particles, with voids between them, are comparatively under bound and more likely to flake or be affected by water and abrasion.

The most widely employed opaque red was vermilion, occurring naturally as cinnabar, and, less often, the varying hues of red earth. Minium and realgar, the two orange pigments, are sometimes superficially indistinguishable, but they can be found within the same painting, demonstrating a subtlety of choice. The predominant yellow was orpiment, sometimes used in a mixture with yellow ocher, which was infrequently used as an independent pigment. Indigo and red lac would presumably have been acquired in powder form, because their preparation as pigments is complex and specialized. For lac, a transparent inert base was stained when the dye was released from the source, an insect secretion, and indigo required a process to induce the precipitation of the colorant from the leaf of the harvested plant. Indigo has enormous tinting power; a minute Proportion of the blue is discernible in a white matrix. Used pure, it has the intensity of black, with a subtle luster, and has been confused with black even by painters; when corresponding and mirrored elements of a particular composition were painted on one side in black and on the other in indigo, the difference was unnoticeable. The red of the lac was deep and transparent, used either in a mixture with white to form the familiar pink of early paintings or as a glaze that modified the color over which it was applied, as did a thin film of indigo.

The black was carbon, most likely the sooty product of burning wood or some other combustible material. White was equally generic, being essentially calcium-based, but like the ground layer, it could include a variety of inert ingredients. If pure chalk was used, it would be transparent and rather gray when wet, only appearing white as it dried. The ratio of white to binder varied, as can be seen in the differing degrees of gloss or matte in the passages of white within these paintings. Whether the proportion of glue depended on the exact nature of the white has yet to be resolved. Some white paint has been vulnerable to the same agents as malachite and azurite: it may have been thickly applied to compensate for transparency and then flaked, or if under bound it may have lost adhesion and if over bound have become brittle. This could be said of all pigments, but very fine and very large particles are the most difficult to manage: in preparation, in handling, and in terms of fragility. These few pigments, embellished with powdered gold bound with glue, produced the sumptuous colors of the Tibetan palette represented in these paintings and, indeed, for the works of art of the next three centuries, until the introduction of synthetic pigments from the West.

Tibetan painters did not use a palette in the accepted sense as a surface on which paints were laid and on which they were mixed to load a brush of blended and variable color. Each pigment and each mixture of pigment had to be prepared as a required and prejudged color, in a pot or Container that would serve as a constant reservoir of paint, for any particular composition. The warm glue binder was added until the correct balance for use as a paint was achieved. Such paint, often inaccurately termed as gouache, is technically and properly described as glue distemper; gouache is a medium in which opacified pigments have been bound in gum, quite distinct from the Tibetan use of animal glue. To ensure the retention of Optimum quality, this quite liquid and warm paint had to be applied as quickly and uniformly as possible. If the mixture cooled or began to dry out, it had to be modified, either by rewarming or by adding more hot water or glue, which would alter the viscosity of the paint.

The basic required colors and mixtures for these early paintings were relatively limited; variants were achieved by glazing with the transparent colors. Later paintings reveal a significant increase in the number of subdivisions and shades, and probably a complex sequence of applications. Evidence of the sequence of painting can sometimes be observed where color fields overlap, especially those that delaminate, thus identifying the first and subsequent layers. Some painters were so precise that their work reveals practically no invasions of one color into another's domain, which makes the sequence hard to deter-mine. The most frequent evidence suggests that the first rendering, throughout the entire composition, would have been of the color fields of blue and green. The rules governing the practice, whether pragmatic or ritual, are unclear at this early stage of study.

In some instances the color fields have been marked by notations, either a number or an abbreviated syllable, informing the artist of the color with which a particular element should be covered. This has led to some speculation that, once the drawing was complete, the task of underpainting would be given to an apprentice with insufficient knowledge or experience who would need the key to "painting by numbers." This may indeed have been the case on occasion, but the notations would equally serve the master as a pragmatic guide to the most effective and efficient expenditure of his paint while it was at its prime. It was not an opportunistic or personal decision to paint a pictorial element blue, for instance; this was determined by an approved and established fact or convention. To overlook an area of blue would, perhaps much later, require new preparation, perhaps inexactly, for a small area. Usually confined to larger, more complex and formal compositions, the earlier color notations seem to be only numerical, probably a reflection of the limited number of hues. Even those observed, however, do not exactly conform with a prescribed canon, in which, for instance, the number three is always green and the number four is always blue. It is possible that artists ascribed their own palettes with numbers rather than follow a ritual convention. The abbreviated word notations of later paintings could simply be a reaction to the extension of the range of colors, making a numerical reference erratic and complicated.

In the earliest of these thankas the paint was first applied in two layers, the second to achieve an even opacity and surface to receive the transparent glaze. This final layer was exploited in the modeling of form: to modify the tint of the base layer, to impose decve motifs, and to add features and outline definition to the composition. Cross-section samples of paint, greatly magnified, taken from the classic Taklung Portrait of Thangpa Chenpo reveal examples of the technique of the eleventh and twelfth centuries. The first layer structure of the blue was initially established in indigo, over which the more brilliant and costly azurite was laid; the second, from the lotus platform, shows the top surface as a deep red glaze over the white and red lac mixture of the pink Substrate. How-ever typical such technique might be, it must be accepted that each painting is individual in its making. Differences and adaptations occur even within a small selection of paintings of the earliest dates, from the late eleventh and the twelfth Century.