practice:lunar calendars could be reckoned in a variety of ways.In the city
states of Greece and the Greek-speaking world, calendars were particularly
irregular, unpredictable, and diverse (Chapter 1 ). The only deviation from this
norm,later to become highly significant, was the calendar of Egypt (known as
the‘civil’calendar) that was non-lunar,fixed, and completely schematic; this
calendar, in 500 BCE, was completely exceptionalin the ancient world.
By the year 300CE, in contrast, nearlyallcalendars within our geographical
scope had become schematic andfixed. The solar Julian calendar was now
established throughout the Roman Empire, around the entire Mediterranean
and in north-western Europe, either in its pure form or in the form oflocal
calendars that had been‘Julianized’and adapted to thefixed 365¼-day year
(as in Egypt, Asia Minor, and most of the Roman Near East;Chapter 5 ).In the
Sasanian Empire to the east, thePersianZoroastrian calendar, also schematic,
fixed, and based on a 365-day year, now served as officialimperialcalendar.
Thelunar calendar survived onlyinlimited regions of thelate antique world:
in Sasanian-held Mesopotamia, inPalestine (atleast among Jews and Samar-
itans), in the region stretching from the Greek peninsula to the Balkans and
thelowerDanube, and perhaps in some parts of Gaul(Chapter 6). But even
thelunar calendar, in these regions, had undergone a certain process of
schematization andfixation. The Babylonian calendar, atleast in itslast
attestation in earlyParthian-period cuneiform sources, had adopted afixed
19 - year cycle of intercalations and was now based not just on empiricalnew
moon sightings (as in the neo-Assyrian period) but also on calculated new
moon predictions (Chapter 2). Otherlunar calendars in Sasanian Mesopota-
mia seem to have been based on Easter cycles or similarfixed schemes
(Chapter 5 ). Easter cycles, by 300CE, were steadily becoming the norm in
theChristian world; they were normally tied to the Julian calendar (or to the
Alexandrian calendar, an adaptation of the Egyptian calendar to the Julian),
thus synchronizing the computation of the (lunar) date of Easter with the
dominant calendars of the Roman Empire. The Jewish calendars ofPalestine
as wellas theDiaspora, sometimes following thelead of Easter cycles, were
also increasingly based onfixedlunar schemes, as is known e.g. of the rabbinic
calendar. Less is known about thelunar calendars inlate Roman Greece, the
Balkans, and thelowerDanube, but it has been argued that the calendars of
late Roman Greece (perhaps, initially, under Seleucid influence in the third–
second centuries BCE) had become more regularized than in the earlier
period. Last but notleast, the Galliclunar calendar as known to us from
calendars werelunar comes principally from Bede (c.72 5 CE), who devotes a chapter of hisDe
TemporumRatione(ch. 15 ) to the‘months of the English’with an elaborate, if not entirely
coherent, description of a pre-Christian, Anglo-Saxon calendar that waslunar but regulated by
the solstices with intercalations (Wallis 1 999: 53 – 4 and above,Ch. 3 n. 69); his use of the past
tense suggests that this calendar was ancient and nolonger practised in his own day. On the value
of Tacitus,Germania 11 , seeIntroduction n. 1 2.
426 Calendars in Antiquity