The Republican Roman calendar consisted of twelve months of fixed
lengths of 28, 29, or 31 days, adding up to afixed, 355-day year.^125 The days
of the month were numbered in descending order with reference to three
points in the month: the Kalends (1st of the month), the Nones (5th or 7th—
depending on the month), and the Ides (13th or 15th).^126 Since the 355-day
year-length fell short of the solar year (by about ten days), a block of 22 or 23
additional days was frequently intercalated, typically (but not necessarily)
every other year. The complicated intercalation procedure brings out again
the idiosyncratic nature of this calendar: an intercalary 27-day month (called
Intercalaris or Intercalarius)^127 was inserted after the 23rd or 24th day of
Februarius, thus cutting that month short (it was usually 28 days long) and
effectively adding 22 or 23 days respectively to the total number of days in the
year.^128
In the later period of the Republic—which is all we really know about
securely—the length and frequency of the intercalation was neither standar-
dized norfixed, but dependent instead on the annual,ad hocdecisions of the
members of one of the most important priestly bodies, the pontifical college
(perhaps, more specifically, of the minor pontiffs).^129 Their intercalations were
often irregular, leading to an erratic relationship between the Roman calendar
(^125) For the list of months, see below, Table 4.1. It does not appear to have followed any logical
sequence (note in particular the succession of three 29-day months from November to Januar-
ius). For a concise introduction, see Bickerman (1968) 43–6.
(^126) For an example see below, n. 158.
(^127) The alternative name Mercedonius is mentioned in Plutarch (Numa18,Caesar59) but
not attested elsewhere (Michels 1967: 18, Samuel 1972: 161 n. 1, Rüpke 1995: 321 128 – 2).
There is no comprehensive account of the intercalation in ancient sources, but the
procedure described above is the plausible reconstruction of Michels (1967: 16–22, 145–67) on
the basis of sporadic literary and epigraphic evidence. Among the relevant literary sources, the
earlier are Varro,De Lingua Latina6. 13 and Livy 43. 11. 13, 45. 44. 3; later sources include
Censorinus 20. 6, Plutarch,Numa18, and MacrobiusSaturnalia1. 13. 22; epigraphic evidence
consists chiefly of theFasti Antiates, the only pre-Julianfastiextant. For a summary of Michels’s
argument see Samuel (1972) 159–64; also Brind’Amour (1983) 27–8.
(^129) Ancient sources refer only to‘pontiffs’(e.g. Suetonius,Julius Caesar40), or even less clearly
tosacerdotes(‘priests’: Solinus 1. 43). Van Haeperen (2002) 219–21 assumes, accordingly, that
the whole pontifical college was in charge. However, Rüpke (1995) 372–3 notes the consistent
involvement of minor pontiffs in calendrical matters (e.g. the publication offasti) throughout the
Republican period. On the other hand, some sources suggest that thepontifex maximusmay also
have had some control over the calendar: according to Censorinus (20. 7–8), indeed, it is as
pontifex maximusthat Julius Caesar instituted his new calendar in 46BCE; whilst Suetonius
(Augustus31. 2) suggests that Augustus’reform of the Julian calendar in 8BCEwas related to his
recent appointment, in 12BCE, to the same office (see Bennett 2003: 232–3; note also that Bennett
2005: 178 relates intercalation patterns between 191 and 46BCEto the terms of thepontifices
maximi, suggesting perhaps that the latter controlled the intercalation). Rüpke argues, however,
that it is asdictator, not aspontifex maximus, that Caesar decreed the institution of the new
calendar (1995: 380–1), and that Censorinus’mention of Caesar aspontifex maximusrefers only
to a preparatory stage in his calendar reform, namely the intercalation of the year 46 BCE (pp.
384 – 7; followed by Van Haeperen 2002: 223). In any event, this means that thepontifex maximus
had some part to play in controlling the calendar.
206 Calendars in Antiquity