Smyly — Examination of Dates of the Assouan Aramaic Papyri. 245 



second; and if A had belonged to the fifteenth year, B (1) would have been 

 in the twenty- third year. 



If A be compared with B (2), on the supposition that the years were 

 Thoth years, A would have fallen in the year 19th Dec, 472, to 18th Dec, 

 471, and B (2) in the year 17th Dec, 465, to 16th Dec, 464. Hence, if A was 

 in the fourteenth year, B (2) would have been in the twenty- first year, and if A 

 had been in the fifteenth year, B (2) would have been in the twenty-second. 

 Thus the assumption of a Thoth year leads to the results that we must take 

 the lower of the two numbers for the year in Papyrus A, and the higher in 

 Papyrus B, and that we must suppose that the accession year of Artaxerxes 

 was counted as his first year. Both of these results are improbable, for we 

 have seen that the higher numbers are to be preferred, and it is not likely 

 that different systems of writing numerals were used in Papyri A and B. 

 And if the accession year of Artaxerxes was counted as his first year, there 

 would have been no reason for dating Papyrus B by the number of the year 

 of Xerxes. 



If, on the other hand, we assume that the years began on the 1st of Nisan 

 (March- April), A will fall in the year 471/0, and B (1) in the year 464/3. 

 If, then, the 14th year was 471/0, the 20th year would have been 465/4, and 

 the 21st 464/3 ; if the 15th year was 471/0, the 20th would have been 466/5, 

 and the 21st 465/4. Since the date of B (1) is in 464/3, we should have to 

 assign A to the 14th year and B (1) to the 21st year of Xerxes, thus taking 

 the lower number in A and the higher in B. 



In a Nisan year B (2) would belong to the year 465/4, so that if A were 

 in the 14th, B (2) would be in the 20th, and if A were in the 15th B (2) 

 would be in the 21st year. This gives rise to no difficulties ; and we are led to 

 the conclusions that the years were Nisan years, and that B (2) is the correct 

 reading of Papyrus B. So far no definite dates have been adopted from 

 independent history ; the results would have been the same if there had been 

 a margin of two or three years on either side in the dates assumed for the 

 kings. Even so it has been found that B (1) cannot be regarded as a possible 

 reading ; but history also provides a strong reason for rejecting it. It is 

 practically certain that Xerxes was murdered in the summer of 465, and it is 

 extremely unlikely that dating by the numbers of his years would have been 

 continued till December of 464, a year and a half later. But B (2) belongs 

 to January of 464, about six months after the death of Xerxes ; and it is 

 quite natural that documents should continue to be dated after the king's 

 death by the number of the current year of his reign, till the beginning of 

 the next year, that is, till the next 1st of Nisan, but not beyond this point. 

 There is an analogy for this in the financial documents of the early Ptolemies, 



[37*] 



