(subdivided into blocks of 16 years, themselves double octaetereis or eight-year
cycles) starting from 222CE, which is presumably around when the table
was composed.^90 This cycle, grossly inaccurate, was modified and updated in
Ps-Cyprian’sDe Pascha Computus, a treatise on the calculation of the date of
Easter written in 243CE.^91 The extent to which these cycles were ever used in
practice, however, remains unknown.^92 They were later superseded in Rome by
the far more accurate 84-year cycles, perhaps already sometime in the third
century.^93 The most important of these was the so-calledsupputatioRomana,
Alexandria). Far more reliable is the testimony of Eusebius (Hist. Eccl.7. 20) that it was a later
bishop of Alexandria, Dionysius, who instituted the octaeteris and rule of the equinox inc.253
(Richard 1974: 310–15, arguing that this octaeteris remained in use in Alexandria untilc.323).
There is thus little reason to believe that the Roman octaeteris, which forms the backbone of
Hippolytus’cycle, was introduced to Rome from Alexandria by Demetrius (as argued by
Richard, and followed by Lejbowicz 2006: 11, 44, Mosshammer 2008: 122; see also n. 98
below). The theory that Hippolytus himself was of eastern origin is similarly without foundation
(see Mosshammer 2008: 116–25); likewise, the theory that the Roman 84-year cycle (on which
see below) was of Asian origin (ibid. 131–8).
(^90) E. Schwartz (1905) 29–36; Mosshammer (2008) 116–25; Holford-Strevens (2008) 167–72.
The table was discovered inscribed on a chair that was supporting a statue; it consists more
precisely of two tables on each side of the chair, one with the dates and weekdays ofluna XIV, the
other with the dates of Easter Sunday. See also Marcovich (1986) 12–13, Lejbowicz (2006) 13,
Nothaft (2011), 38–9.
(^91) Text inPL4. 1023–47; discussion in E. Schwartz (1905) 36–40, Mosshammer (2008) 125–7;
translations in Ogg (1955), Strobel (1984) 43–67. The author appears to have been of African
origin, as his biblical citations follow an Old Latin version associated with Africa (Ogg 1955: vii),
but the place of writing of this work remains unknown (E. Schwartz 1905: 40; Lejbowicz 2006:
17 – 18). The dating of 243CEis almost explicitly given in ch. 22 (Ogg 1955: 19).
(^92) Richard (1966) 266 (andpassim). The authors of these cycles were not necessarily in a
position to enforce their use: all we know is that ps.-Cyprian may have occupied some high office
in the Church (Ogg 1955: 23 n. 18), whilst the identity of Hippolytus remains controversial and
obscure (Mosshammer 2008: 118–21). C.W. Jones (1943) 14–15 argues that 3rd-c. Christian
communities must have used a great variety of Easter tables, many of these local and more or less
self-made; in my view, however, it is even more likely that most Christians in this period did not
use any tables at all.
(^93) The dating of the earliest 84-year cycles depends on a controversy surrounding the
Laterculusof Augustalis, an 84-year cycle beginning in 213CEthat is known from theDeRatione
Paschae, a Carthaginian treatise dating from 455CE(text in Krusch 1880: 279–97, with references
to Augustalis in 280–1, 289–90). Nothing is otherwise known about this Augustalis, but accord-
ing to E. Schwartz (1905): 63–6, followed by Mosshammer (2008) 227–8, some features in his
cycle point distinctly to a 5th-c. redaction: it was structured (at least according to its description
provided inDeRatione Paschae, which, one could argue, does not necessarily represent the
original conception of Augustalis) into four ogdoads (eight-year periods) and hendekads (11-
year periods) in alternation, followed by one extra ogdoad (thus a total of 84 years), clearly in an
attempt to reconcile this cycle with the Alexandrian 19-year cycle (8 + 11 years), which would
have been of no concern whatever to a 3rd-c. Roman computist. Furthermore, the cycle
commences in a year with aluna XIVon 25 March (whereas thesupputatioRomanabegins in
a year with an epact of 1, which works out one year later), which was traditionally regarded as the
Julian date of the Passover of the Crucifixion (Krusch 1880: 290); thisfits in with other 5th-c.
Easter computists who similarly attempted to correlate their cycle with the date of Christ’s death
(see alsoWarntjes 2007: 69–70; Holford-Strevens 2008: 207 n. 82). On the other hand, it seems
strange that a 5th-c. computist should have started his cycle as early as 213CE(rather than in 381,
Dissidence and Subversion 327