being annexed into the Roman Empire. Prior to this, the calendar in Dura had
been lunar, as elsewhere in the Parthian Empire. This is evident from the
mention of the month of Dystrosembolimos(intercalary) in two inscriptions
from the city, characteristic of a lunar calendar and probably identifiable with
the official Parthian calendar (see above, n. 46). A horoscope inscription dated
9 Panemos 487SE(i.e. 176CE) and referring astronomically to a date ranging
between 3 and 5 July of that year demonstrates further that in this period the
Macedonian calendar in Dura-Europos was still lunar.^164 But from the early
third century double-dated documents clearly indicate that the city had
adopted the Julian calendar, retaining only—as in Antioch and other parts
of Syria—Macedonian month-names and an autumn New Year.^165
The end of AntiquityThe Julianization of Near Eastern calendars, conterminous with Roman ex-
pansion in the Near East, had reached its greatest extent by the third century
CE. Galen could thus write already in the late second century that‘Romans,
Macedonians, Asiatics, and many other nations’all used a solar calendar.^166
But some notable exceptions remained: lunar calendars were still in use in
Palestine, as Galen himself noted, and so in other parts of the Roman Empire
that will be considered in Chapter 6. Even in the vast majority of the Roman
East where afixed 365-day calendar had been adopted, considerable diversity
remained between the calendars and few, if any, could be regarded as identical
with the Julian calendar.
The Julian calendar in its pure form, complete with Roman month-names,
did eventually penetrate the Roman East, but not before thefifth or sixth
(^164) Indeed, if months began atfirst visibility of the new moon, 9 Panemos in 176CEwould
have corresponded to 4 July. See references above, n. 48. Although Dura-Europos was already
under Roman control by the 160sCE, it was not formally incorporated into the Roman imperial
administrative structure till the 190s (Millar 1993: 114–15, 131–3, 467). In this respect, the lunar
horoscope of 176 165 CEmay be regarded as belonging still to the Parthian or post-Seleucid period.
Welles, Fink, and Gilliam (1959) 10, referring to texts nos. 29 (251/2CE) and 30 (232CE);
the former implies a New Year in Dios (ibid. 153, as pointed out to me by Chris Bennett). Text
no. 32, a divorce bill from 254CE, is dated 29 Xanthikos (upper text, l. 1) and 30 April (lower text,
l. 3: [ðæe]äýï ŒÆºÆíäHí[Ì]Æåßøí), a discrepancy noted (ibid. 169) that is presumably erroneous
and that may be due to the fact that these dates were written by different scribes.We do not know
how early the calendar change was made, but as in Arabia, it may have been simultaneous to the
formal annexation of Dura-Europos in the Roman Empire. The presence of the Julian calendar
in Roman Dura-Europos is also evident from the full-length Roman calendar that was produced
for an auxiliary cohort stationed in the city (ibid.no. 54; also in Beard, North, and Price 1998: ii.
71 – 4). Some documents, moreover, are only dated by the Julian calendar (no. 26, from 227CE).
(^166) Galen,InHippocratis Epidemarium Libros Commentarius3 (ed. Kühn xvii/1. 24, see also
21 – 2); Palestine is mentioned on p. 23.
Fragmentation: Babylonian and Julian Calendars 293