advance, and that irregular tampering was not to be expected.^70 But this
argument is weak, because in the context of a light-hearted comedy, attention
to such detail is unnecessary. Aristophanes may well have chosen to ignore, in
his text, the possibility of a hollow month because this would have been a
potentially distracting complication in Strepsiades’speech.
From this same passage it may be inferred that debt collection and/or
instigation of legal proceedings took place, at Athens, on the last day of the
month,‘old and new’—the day that Strepsiades was dreading.^71 Some have
argued that this last day of the month must have been known on the day itself,
and not determined later, retroactively, by sighting the new moon in the
subsequent evening; otherwise, creditors would never have known whether
to collect their debts andfile their cases on the 29th or on the 30th.^72 This has
been taken further as evidence that the end of the month was determined at
least some days in advance.^73 Not enough can be inferred, however, from this
brief passage of Aristophanes, and not enough is known about the practice of
creditors at Athens, for such conclusions to be drawn. It quite possible, for
example, that debt collections and/or legal proceedings were always instigated
on the 30th day of the outgoing month, even if this day turned out later (or
even on the day itself ) to be not‘old and new’, but the 1st of the new month
(this would explain, incidentally, why Strepsiades ignores the possibility of a
hollow month). Alternatively, it is possible that the archons made sure the
beginning of the new month, and hence‘old and new’of the old month, were
determined and publicized no later than the 29th day. This would admittedly
mean that they could not have determined the beginning of the new month on
the basis of actual sightings of the new moon; at most, the date of appearance
of the new moon would have been anticipated or estimated.^74 But this is
completely plausible, for as noted above (near n. 15), there is no evidence of
new moon sighting in ancient Greek sources, and anyway, conformity to the
moon was not always strictly maintained. This passage does not furnish
evidence, however, of predictability or regularity of the length of the Athenian
month.
(^70) Walsh (1981) 112 n. 10 (citing Meritt), in support of the theory of the gibbous moon
(above, n. 68). 71
See above, n. 65.
(^72) Dunn (1998) 217 n. 16, rightly rejecting, for the same reason, Pritchett’s suggestion that the
29th remained nameless until moon watches were carried out in the evening.
(^73) See n. 70.
(^74) New moon sightings may still occasionally have been made for the purpose of verifying the
accuracy of the calendar: see Dunn (1998) 216–17, unfairly dismissing Pritchett’s suggestion that
predictions of new moon visibility may have been based on observations of the waning of the old
moon. Note also that if the archons’predictions went very wrong, they always had the option—if
they so wished—of adding an intercalary last day of the month (henakai nea embolimos), i.e. an
extra day after the 29th or 30th, as is attested in the epigraphic record (see e.g. ibid. 221–2).
46 Calendars in Antiquity