国立情報学研究所 - ディジタル・シルクロード・プロジェクト
『東洋文庫所蔵』貴重書デジタルアーカイブ

camels and passes through their mouths like bridles. This example very clearly
shows that after a certain lapse of time ancient forms can become quite new and
no longer hold their original meaning.

Plate XXVIII no. 4 gives an idea of the diversity of possible variations.
This pair of ibex is far more broken up than any Minussinsk piece, although the
element that causes this dissolution is formed of an ancient motive from this
region, the circular perforation of which we spoke at length in Chapter VI.
It takes possession of the tree at the centre and changes it into three wheels.
It appears above the horns and below the bodies, in fact if takes hold of all pos-
ible spaces. Only the ibex themselves, well modeled moreover, are not so treated.
Three stylistic tendencies meet here, the heraldic grouping and the circular
perforations of Minussinsk, the modeling of an early period, and finally the
tendency towards a schematic transformation which is an artistic characteristic
peculiar to the barbarians of the northern Chinese frontier. A date near 1000
best corresponds to the coincidence of these three elements. When even the
modeling disappears as in Plate XXVIII no. 5, we must be at an even later
period and have entered the II millennium. Here the perforations have even
broken up the forms of the standing animals, probably stags, the heads of
which are backward-turned. Animals so geometrically simplified, three instead
of a pair, and all turned in the same direction, are certainly an invention of our
region. Among the latest yet most primitive plaques, is that of Plate XXVIII
no. 6, composed in the same way as Plate XXVIII no. 4. The stags are nearly
unrecognisable. The centre tree, a very ancient motive, has become a dotted
line, and the circle above is less a symbol than a method by which the artist
avoided small perforations and replaced them with a linear design. Such a
weakening of original forms can only be possible at a late Mongol period. This
object, so to speak, is of " no date ".

There are animals in plaque form of so small a volume as to be mere
skeletons. A wild animal (Plate XXIX no. 1) may, in spite of this fact, be
recognised as a wolf. The disappearance of the body surface, however,
cannot be possible before the middle of the I millennium. There is still less
volume in the ibex of Plate XXIX no. 2, where even a true drawing of the legs
has been lost. The back, the legs joined together, and the jaws, have become
grooved bars. We always find at the end of this art, that is to say at the beginning
of the II millennium, the same tendency to forget nature by placing the
emphasis on the ornament.

The figure astride a camel (Plate XXIX no. 3) is of a type already observed
in the hooks of Plate XX nos. 3 & 4, and is of the same period, the beginning of
the II millennium.

The small figure (Plate XXIX no. 4) is not sufficiently distinct for us to draw
any conclusions; it must probably be of the end of the T'ang period. A heraldic