astronomical calendars provided. This may explain why 19-year cycles appear
to be followed, albeit not consistently, in parts of the Athenian epigraphic
record (}1). However, the use of astronomical calendars would not have been
without problems, because the beginning of the month in the civil calendars
was meant in principle to coincide with thefirst appearance of the new moon
(as discussed in}1), whereas in the astronomical calendars the month nor-
mally began at the conjunction.^98 This should have led to a typical discrepancy
of two days between an astronomical calendar and a well-regulated civil lunar
calendar. This discrepancy does not mean that astronomical calendars could
not have been used to help regulate the civil calendars, but it does confirm, at
the very least, that astronomical calendars were fundamentally not designed or
well suited for this purpose (Jones 2000a: 156–7).
There are good reasons why in spite of their scientific accuracy, the adop-
tion offixed astronomical calendars should have been resisted in the Greek
cities. As I shall be arguing throughout this book, calendars were not a
scientific pursuit, but primarily a political affair. The ability to manipulate
the calendar through suppression and intercalation of days and months was a
political privilege as well as an important tool of government, as I shall discuss
further towards the end of this chapter. City councils and magistrates would
not have lightly given this up. The adoption of afixed astronomical calendar
would have meant the abolition of the politicians’ability and right to control
it, and thus effectively a major political reform. There is no evidence that such
a reform was ever even proposed.^99
Astronomical and chronological usesIn this light, the astronomical calendars have often been characterized as
‘theoretical’, their main use being confined astronomical dating.We know at
least that the Callippic cycle was used for dating astronomical observations
(^98) For the Metonic cycle and its epoch, see above, n. 16. For the Callippic cycle, see A. Jones
(2000a) 144: the epoch of thefirst Callippic cycle was 28 June 330BCE, when summer solstice and
the conjunction almost exactly coincided (at least according to modern astronomy; we do not
know how accurately these events were reckoned at the time). The coincidence of the beginning
of months with the conjunction can also be inferred from Callippic dates attested in astronomi-
cal sources. But as Jones rightly stresses, this does not prove that the Callippic cycle was
deliberately designed in such a way that its months would always coincide with the conjunction.
(^99) As well argued by Pritchett (2001) 89–90, 98–101, who points out that a reform of this kind
could not have failed to generate debate in the city’s political arena, and probably also to meet
with considerable resistance, which would not have gone unnoticed in the writings of ancient
historians and orators; the absence of any hint to this effect in ancient sources—e.g. in the
passage of Diodorus above-mentioned—suggests that no such reform ever took place. This
general conclusion is endorsed by Lehoux (2007) 76–7 and (more reservedly) Hannah (2005)
68 – 70, 81.
Calendars of AncientGreece 53