THE BERLIN ASTRONOMY 69
far, however, from accepting the resultant Berlin date for
it, he pushes the whole thing a cycle of 1460 years further
back, and begins the Xllth Dynasty in 3459, the XHIth
in 3246, and the XVth in 2533.1
The differences in date are so vast, and the issues at
stake so important for our conceptions of .Egean civilisa-
tion, that we must consider at some length what is to
be our attitude to these conclusions. It is clear at the
outset that we must keep distinct the three very different
kinds of grounds on which we can test them. The validity
of the astronomical arguments, and the length of time
required for the development of Egyptian art and history,
are quite different things ; while the length of time
necessary for the development of Minoan art, which is
the only kind of Minoan history that we can yet judge
1 Sinai, 1906, pp. 163-75. It *s difficult to know what to
make of some of Petrie's statements either here or in his earlier
discussion of the Sothic question in Hist. i. 1903, pp. 248-52.
When for instance our MSS. of Censorinus give Thoth I as = VII
Kal. Jul. = June 25 (Julian) in the year he is writing, a.d. 238,
and deduce from it Thoth I as = XII Kal. Aug. — July 21
(Julian) for a.d. 139, Scaliger (Cholodniak, 1889, notes p. 75)
conjectures XIII Kal. Aug., because by the dates given in the
Ptolemaic Almagest Thoth I must have fallen in 139 a.d. on
July 20 (Julian). The Germans all accept this conjecture. Petric,
however, in both books keeps July 21. Has he tacitly accepted
Oppolzer's (S.S.A. 1885, pp. 16-7) ingenious defence of the text ?
On this view the Egyptian day began at sunrise, and thus over-
lapped two Roman days, which, like ours, began at midnight.
Thus the rising of Sirius, occurring as it did just before sunrise,
in the last few minutes of Thoth I, might be counted in July 21,
as well as in July 20. Or is Petrie's July 21 an accident ? On
what principle, again, does he call a.d. 238, a.d. 239? He ignores,
too, the further question, raised by the Germans, as to whether
July 19 of a.d. 140 is not the date that Censorinus really ought
to have given. The Kahun Inscription again, according to
Borchardt's reading, makes Sirius rise on the 16th of Pharmouthi.
Petrie, however, says the 17th. Is this a new reading of the
text, or has his eye hit on the later entry in which the offerings
given on the feast are entered on the next day ?
far, however, from accepting the resultant Berlin date for
it, he pushes the whole thing a cycle of 1460 years further
back, and begins the Xllth Dynasty in 3459, the XHIth
in 3246, and the XVth in 2533.1
The differences in date are so vast, and the issues at
stake so important for our conceptions of .Egean civilisa-
tion, that we must consider at some length what is to
be our attitude to these conclusions. It is clear at the
outset that we must keep distinct the three very different
kinds of grounds on which we can test them. The validity
of the astronomical arguments, and the length of time
required for the development of Egyptian art and history,
are quite different things ; while the length of time
necessary for the development of Minoan art, which is
the only kind of Minoan history that we can yet judge
1 Sinai, 1906, pp. 163-75. It *s difficult to know what to
make of some of Petrie's statements either here or in his earlier
discussion of the Sothic question in Hist. i. 1903, pp. 248-52.
When for instance our MSS. of Censorinus give Thoth I as = VII
Kal. Jul. = June 25 (Julian) in the year he is writing, a.d. 238,
and deduce from it Thoth I as = XII Kal. Aug. — July 21
(Julian) for a.d. 139, Scaliger (Cholodniak, 1889, notes p. 75)
conjectures XIII Kal. Aug., because by the dates given in the
Ptolemaic Almagest Thoth I must have fallen in 139 a.d. on
July 20 (Julian). The Germans all accept this conjecture. Petric,
however, in both books keeps July 21. Has he tacitly accepted
Oppolzer's (S.S.A. 1885, pp. 16-7) ingenious defence of the text ?
On this view the Egyptian day began at sunrise, and thus over-
lapped two Roman days, which, like ours, began at midnight.
Thus the rising of Sirius, occurring as it did just before sunrise,
in the last few minutes of Thoth I, might be counted in July 21,
as well as in July 20. Or is Petrie's July 21 an accident ? On
what principle, again, does he call a.d. 238, a.d. 239? He ignores,
too, the further question, raised by the Germans, as to whether
July 19 of a.d. 140 is not the date that Censorinus really ought
to have given. The Kahun Inscription again, according to
Borchardt's reading, makes Sirius rise on the 16th of Pharmouthi.
Petrie, however, says the 17th. Is this a new reading of the
text, or has his eye hit on the later entry in which the offerings
given on the feast are entered on the next day ?