suggests that there too, Babylonian months could diverge from the central
Babylonian calendar, presumably for exactly the same reasons.^77 It was per-
haps to remedy this problem that new moon prediction came to be favoured in
the Babylonian calendar.Whereas empirical new moon sightings carried out
in Babylon were not conducive to wide dissemination, new moon predictions
could be made well in advance, and could thus be circulated in good time
across the Empire. There is no evidence that predictive calendars were
disseminated in practice in this way, but the possibility of doing so may
have been attractive to the rulers of the great ancient empires in the Near East.
Thus in more than one way, the evolution of the Babylonian month towards
greater predictability and regularity may have been related to social, political,
and imperial administrative factors, rather than to the development of scien-
tific expertise. This conclusion, albeit tentative in the context of the new moon,
willfind confirmation in the context of the intercalation.
- INTERCALATION
In order for a lunar calendar to keep up with the solar year and the seasons, it
is necessary to intercalate a 13th lunar month every two or three years. The
practice of intercalation appears to have been widespread in Mesopotamia
throughout Antiquity, from already the third millenniumBCE.^78 Evidence
from this early period, however, is limited. Intercalation appears to have
been quite irregular: in the early second millenniumBCE, intercalations were
sometimes made in three or four consecutive years.^79 The decision to
Zoroastrian year; it makes sense that preparations for these offerings were being made during the
final days of the twelfth month (de Blois, forthcoming). The two-day discrepancy between the
Babylonian date of this document (Shebat:20) and the date expected in the central Babylonian
calendar (Shabat:u 22, assuming a month beginning withfirst visibility of the new moon in
Babylon) can be explained as the result of a local, approximate reckoning of the Babylonian
calendar.
(^77) On similar divergences in the Seleucid and post-Seleucid periods, see Ch. 5.
(^78) The only possible exception is the Assyrian calendar of the Middle Assyrian period, in the
15th–12th cc.BCE(before Assyria’s adoption of the standard Babylonian calendar inc.1100BCE),
where an apparent drift of the calendar in relation to the seasons and the solar year has been
interpreted as due to absence of intercalation (Cohen 1993: 237–47; Reade 2000). But the Old
Assyrian calendar (20th–18th cc.), according to recently discovered evidence, was clearly inter-
calated (Veenhof 1995–6: 13–15; 2000: 140–7).
(^79) Huber (1982) 56–61; Englund (1988) 123–5 n. 2 (I differ from Englund, who considers the
evidence to indicate that intercalation was generally regular); Greengus (1987) 214 n. 20; Britton
(2007) 119. Greengus loc. cit. also shows evidence that an intercalation could be known as much
as 35 days in advance, which indicates that the decision to intercalate was not necessarily takenin
extremis.
94 Calendars in Antiquity