The misconceived classification offlexible, unpredictablecalendars as infe-
rior and primitive and offixed, predictablecalendars as superior and ad-
vanced, implying in turn the notion of progress from the former to thelatter,
lies at the foundation of evolutionism, a socialtheory which has been particu-
larlyinfluential,albeitlatently, in modern calendar scholarship. Although
evolutionism haslargelyfallen from favour in the socialsciences, calendar
scholars persist in assuming an almost deterministic, evolutionary trend which
is variously described as a progression of calendars from unpredictableto
predictable,flexibletofixed, irregular to regular, and empiricalto calculated
(or cyclical, or schematic), often together with a shift fromlunar to solar
reckoning. This evolutionary modelwaslargely promoted by Nilsson in a
seminalwork entitled, revealingly,Primitive Time-Reckoning( 1 920; see esp.
148 – 9); it has been reiterated by severalmore recent historians including
Bickerman, who writes, for example:
The evolution of the calendar, thus, follows threelogically and also historically
successive stages: 1. separation of the beginning of the month from the sighting of
the new moon; 2. the empiricaladjustment of thelunar count to the course of
the seasons, that is practically to the solar year; 3. the cyclic calculation oflunar
months.^19The supposedlylogicalunderpinning of this modelis particularly noteworthy;
the extent to which Bickerman means it as deterministic or evolutionist (note
his use of the term‘evolution’) is perhaps, in fairness, uncertain, but neverthe-
less this is the impression that he conveys. For Bickerman and others,logical-
historicalsequences of this kind account for the development of several
ancient calendars such as the Babylonian,Persian, Roman, and Jewish calen-
dars. Atfirst sight, this approach mightfind justification in that the shift from
lunar to solar and fromflexibletofixed calendars appears to have been a
generaltrend in the ancient world, and that there is hardly an attested case of
shift in the reverse direction.^20 However, the generality and apparent irrever-
sibility of this trend do not necessarily turn it into alaw of nature, nor into
somelogical, deterministic, evolutionary force inherent to the history of
ancient calendars.In this book,Ishallargue that this generaltrend must be
(^19) Bickerman ( 1 968) 1 9; see also Schürer ( 1973 – 87) i. 5 94.What exactly Bickerman means in
this passage, how these three stages follow from one another (especially 3 from 2), and how they
might be partially or wholly exemplified by the ancient calendars, Greek, Babylonian, Egyptian,
etc., which he mentions immediately afterwards, are not entirelyclear. Stages 1 and 2 seem to
refer to a shift fromlunar to solar calendars, whilst 1 and 3 refer to a shift from empirical
methods tofixed, cyclicalschemes.
(^20) A shift from solar tolunar calendar seems only to have occurred in exceptionalcircum-
stances: e.g. at Singara (upper Mesopotamia) in 363CE, when the city was transferred from
Roman toPersian Sasanian rule and, most probably for this reason, abandoned the Julian
calendar in favour of its earlier, Mesopotamian tradition oflunar calendar reckoning (Stern
2004: 468).
16 Calendars in Antiquity