study) to a purely scientific discipline. Fostered by a nineteenth-century
tradition of handbooks of so-calledmathematischeund technischeChronolo-
gie,^16 this approach assumes that calendars were constructed and regulated on
the sole basis of scientific, especially astronomicaland mathematical, knowl-
edge, and hence that their development was driven by scientific progress. Thus
the design of afixed 19 - year cycle of intercalations in the Babylonian calendar
during the Achaemenid period has been interpreted as the outcome of ad-
vances in Mesopotamian mathematicalastronomy; similar arguments have
been invoked, without much evidence, to explain, for example, the emergence
of afixed Jewish rabbinic calendar inlate Antiquity.^17 But Mesopotamia was
perhaps a specialcase, because in the rest of the ancient world, the impact of
astronomers and astronomicalscience on the development of calendars was
ratherlimited. The inaccuracy of the Egyptian civilcalendar as a solar calendar
was evident to Egyptian astronomers and could easily have been corrected
with the insertion of an additionalday every four years, as in fact was proposed
in the mid-third centuryBCEin the decree ofCanopus; but this did not prevent
the Egyptian calendar from remaining unchanged for severalmillennia, until
the arrivalof the Romans in thelatefirst centuryBCE—whereupon the calendar
was reformed, for politicalrather than astronomicalreasons.In Athens and
other cities of the Greek peninsula,flexible and irregularlunar calendars were
retained untilthe end of Antiquity in spite of the considerable progress in
astronomicalknowledge that was achieved in the Graeco-Roman world, and
in spite of the emergence, in Greek astronomicalsources, of highly developed
lunar calendar schemes that were used by astronomers but not, as far as we
can tell, in any aspect of civic and publiclife. Even in Mesopotamia, astronomy
hadless of an impact on the development of the officialBabylonian calendar
than has been assumed. Although Mesopotamian astrologers were involved in
setting the months and years of the Babylonian calendar—in contrast with
most other ancient societies, where calendars were controlled entirelyby
politicalrulers—decisions were often taken by the king or imperialofficials,
and not always on the basis of astronomicalconsiderations. AsIshallargue,
even the 19 - year cycle that was eventually adopted was not the best scheme
that Babylonian astronomy could have produced; it was, to alarge extent, the
historicalresult of intercalation policies that successive Achaemenid kings had
instituted. JuliusCaesar is one of the few politicalrulers in Antiquity to have
especially commissioned astronomicalexperts for the design of a new, sche-
matic calendar, although we now know that his average 365¼-day solar year
was slightly excessive (this was eventually corrected in the sixteenth century by
Pope Gregory, again with the helpofespecially recruited astronomicalexpertise).
(^16) Ideler ( 1825 – 6); Ginzel( 1911 ).
(^17) On the Babylonian calendar, seeCh. 2. On the Jewish calendar, see Schürer ( 1973 – 87) i.
5 94 and my critique of it in Stern (200 1 ) 227–32.
Introduction 13