group of stars; on this basis, a 19-year lunisolar (or more precisely, lunistellar)
cycle of intercalations could have been constructed and instituted.^120
It cannot be denied that astronomy played an important part in calendar
regulation. As we have seen, evidence from the neo-Assyrian and earlier
periods suggests that the relationship between lunar months and the solar or
sidereal year was regularly evaluated, maybe on an annual basis, in order to
establish whether an intercalation was necessary. This may have been achieved
through the observation of equinoxes and solstices, or (much easier) of stars or
groups of stars. Later, by the Seleucid period, the formulation of afixed scheme
for calculating equinoxes and solstices (known as the‘Uruk scheme’)^121 may
have provided an easier method of establishing whether an intercalation was
needed.
But the astronomical foundation of the Babylonian 19-year cycle remains
entirely unclear. It is frequently claimed that the cycle was designed in such a
way that the year would always begin on or after the true vernal equinox.^122 In
objective terms this is correct, since at its earliest occurrence in the cycle, in
year 17, the month of Nisannu always began on the day of the vernal equinox,
whilst in all other years it began later.^123 However, the Babylonians themselves
were probably unaware of this, because their calculation of the vernal equinox
was different from ours: according to the‘Uruk scheme’, widely used in the
Seleucid period, the vernal equinox occurred a few days later than what we
know as the true equinox, and schemes used in earlier sources could assume
an even later date.^124 As a result, the earliest Nisannu (in year 17) is deemed
in the sources to begin a few days (typically, three days)beforethe vernal
equinox.^125 In Babylonian terms, therefore, there was no clear relationship
(^120) Britton (1993) 66–8, and with more detailed explanation, Brown (2000) 174–5. A partic-
ular interpretation of text E, re. 18 (Neugebauer and Sachs 1967: 205) may lend support to this
theory, but I consider it unlikely (see Stern 2008: 36 n. 10). Brown (2000) 193 suggests that an
astronomical text relating to 616– 588 BCE(BM36731) implies an attempt to work out the length
of the year, which may have been used in some way for regulating the intercalation; this is
another expression of the same paradigm, that calendar regulation depended on advances in
mathematical astronomy. 121
Neugebauer (1975) i. 357–63; Slotsky (1993). Neugebauer (1975) i. 366 surmises that the
Uruk scheme may have been introduced in the early 4th c.BCE; Britton (2007) 125 dates it to
c.350BCE.
(^122) Britton (1993) 67–8, (2007) 120–4.
(^123) As astronomically demonstrated e.g. by Huber (1982) 8–10 (note that his analysis is based
on the conjunction preceding 1 Nisannu, not on thefirst day of Nisannu). Hartner (1979) 2–3,
(1985) 742–3 argues, on this basis, that the 19-year cycle must have been instituted in 503/2BCE
because in this year 1 Nisannu coincided with the vernal equinox, and this would have
constituted the natural beginning of the cycle.
(^124) On pre-Seleucid deviations from the Uruk scheme, see Slotsky (1993): e.g. in Sachs and
Hunger (1988–2006) v. 194–5 (no. 57, 423BCE), the equinox is reckoned one day later than
according to the Uruk scheme, and so ibid. 236–9 (no. 62, 379BCE).
(^125) This is explicit e.g. in Sachs and Hunger (1988–2006: iii. 116–7), where in 142BCE(year 17
of the 19-year cycle) the equinox is said to have been on 4 Nisannu. See also ibid. i. 108–9, where
The Babylonian Calendar 117