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

conditions quite clear of any arguments dependent on the amount of available water, methods of
irrigation, etc., i.e. factors which here, as elsewhere in the Tārim Basin, still await systematic survey
by a competent irrigation engineer. At the same time I thought it necessary to point out that while
this shrinkage of the water-supply resulting from desiccation 'fully explains why the ancient site was
never reoccupied, it does not prove that the original abandonment was directly and solely due to it'.
Nor could I neglect to add the warning that 'as in most historical changes the causes [for that
abandonment] may have been far more complex than the modern inquirer is apt to assume'.

Uncertainty
as to cause
of abandon-
ment. I must still maintain this distinction between the definitely proved case of desiccation and the
uncertainty which surrounds the circumstances in which the abandonment of the ancient oasis actually
took place. The interesting account which Prof. Huntington has given of his short visit to the site
in 1905, brings out, indeed, very clearly the fact that the Niya River in its present size and con-
dition could not so much as reach, much less irrigate, the extensive area of ancient cultivation marked
by the ruins.¹² But I do not find in the observations there recorded anything to prove the implicit
assumption which would attribute the abandonment of this settlement towards the close of the third
century A.D. directly to a gradual failing of irrigation. Nor have my own renewed explorations
furnished definite evidence for settling the question on a critically safe basis. But some of their
results have a sufficiently direct bearing to deserve a brief collective review here.

Evidence of
archive
hidden in
N. xxiv. In the first place attention is due to the peculiar conditions in which the small hidden archive of
N. XXIV was discovered. From the details fully set forth above it may be considered certain that
the last occupant of the residence in question had been obliged to leave his home in an emergency,
and not on account of a gradual abandonment of the oasis such as would necessarily follow the pro-
longed failure of irrigation. What particular emergency caused this hurried departure we are
not likely ever to learn, nor what prevented the owner's return. Where historical knowledge is so
limited, the range of possible explanations—war, insecurity, pestilence, fiscal oppression, etc.—must be
wide. Instead of indulging in conjectures, we should learn from such an instance the need of caution
in our inferences where the chances of human affairs in a distant past are in question.

No proof
for gradual
abandon-
ment of site. It is an argument of considerable weight for the point at issue that the extensive range of ruins
excavated failed to reveal the slightest archaeological indication that the abandonment of different
portions of the ancient oasis had taken place successively. Difficulties about irrigation, as illustrated
by numerous modern instances, including the typical case of 'Old Domoko', would certainly make
themselves felt first at the outlying parts of the canal system, as duly pointed out by Prof.
Huntington. But though the length of the area from south to north over which structural remains
have been carefully explored, extends now to over twelve miles, nothing has come to light to
warrant the suggestion that the finds at the northernmost ruins date from an earlier period than
those yielded by the ruins at the other end of the old oasis. The evidence against such an assump-
tion may claim special weight because it is furnished also by documents with exact dates in regnal
periods which, as far as our present knowledge goes, are uniformly represented over the whole area.

No cutting
of dead
trees. In this connexion it is also of importance to mention that nowhere among the northern group
of ruins, nor, as a matter of fact, anywhere else on the site, did I come across any of those stumps
of cut garden- or fruit-trees which are such significant mementos at oases gradually abandoned.
Where the cultivated area has undergone gradual shrinkage those holding the lands still under