National Institute of Informatics - Digital Silk Road Project
Digital Archive of Toyo Bunko Rare Books

tip of the tail. A bird-head is often found in this position, but it is seldom that
of the eagle as Alfoldi believes (84), but rather that of the goose as can be readily
seen on the T'ang piece from the Altai (73) of which we have already spoken.
The tiger's body is here decorated with circles and spirals which may be either
of a Chinese or Steppe origin and which in our classification is proof of a relat-
ively early date, at least of the Han period. Among other types of this period
we often find an animal combat placed on the wider end (85) either framed by
a border or unframed.

In order to make some ordered arrangement of these innumerable motives,
we shall at first deal only with the wild animal designs. We then pass to a period
after Han. Plate XVIII no. 7 could be artistically of the same early date, but the
meaning behind the motive makes this impossible. Around the neck of a natural-
istic tiger there is a collar to which the bar is attached by means of a three-fingered
hand. The latter motive has been studied by Hentze in different regions of
prehistory and he connects it with the descending phases of the moon (86).
Tiger, collar and hand remind us of very definite Chinese myths. An animal
with the same attributes, reproduced on a Han brick, is explained by Hentze as
the demon of darkness whom the spirits pull away from the solar tree by
means of a rope attached to its collar (87). The original meaning seems lost
here as the motives are quite disconnected. Evidently they were borrowed
mechanically from ancient objects. The fact that the Chinese added a Chinese
decoration to a Eurasiatic utilization gives us a hybrid art, a remark that is true
of the majority of northern frontier objects. If the smooth surface and weak
modeling of the tiger speaks for a date about the middle of the I millennium,
a repliqua where the fur is drawn in regular design (Pl. XVIII no. 8) must be
even later, perhaps of the T'ang dynasty, since the modeling is even more indis-
tinct. The hydra-head as well, has nearly disappeared, a sign of the tendency
of the development to transform the majority of motives in the course of time.

This tendency consists of a constant breaking away from naturalism. In
Plate XIX no. 1 the animal has not lost all true proportions. The stylised outline
has kept the characteristic curve, but the claws are simplified and the head enlarged.
Instead of a natural rendering of the fur we find an ornamental incised design
which again comes from the northern technique of bevelling and which covers
the whole surface without organic arrangement. The joints are emphasised with
the spirals that are so favoured by Scythian art. The collar can scarcely be seen,
the hand and hydra-head have disappeared. This simplification and weakening
in style seems to be of a period about 1000. The technique of incision, adopted
again for casting, finally so breaks up the surface that the tiger nearly becomes
unrecognisable. We find this animal (Plate XIX no. 2) in a rectangular frame
that as a whole has the appearance of openwork. The frame allows for a
development of the claws and of the joining of the long thin tongue to the