THE MECHANICS OF
ANNIVERSARY CALCULATION
Before exploring the theme further, it is worth pausing to reflect on what exactly
an anniversary is. What does it actually mean to say that “today” is the anniver-
sary of another day? The issue is not so hard if we are talking about dates since the
Gregorian reform of Julius Caesar’s calendar of 45 b.c.e., which means since 1582
in most of Europe, since 1752 in Britain and the North American colonies (Britain
always lags behind European initiatives), 1918 in Russia, 1923 in Greece. Since the
reform, anniversaries have some kind of logic, for the calendar is the same, but
anniversaries across the divide, including those reaching back to the ancient world,
are factitious. In 1582 ten days were dropped to allow for the uncorrected shifts
since Caesar’s time: “Adjustment was necessary because the Julian year, consisting
of 365 days, with a 366th day added every fourth year, has an average length of 365
days 6 hours, which is some 11 minutes 12 seconds too long, causing Julian dates
to fall progressively further behind the sun.”^87 This was done in such a way that 4
October was immediately followed by 15 October — Virgil’s birthday, as it hap-
pens.^88 If you wish to celebrate an anniversary across that divide, then, you cannot
claim that a precise number of years has passed since the corresponding day in the
original year, whether you use a sidereal year, with the planet facing precisely the
same spot in space, having revolved a given number of times every 365 days, 6
hours, 9 minutes, and 9.5 seconds, or a tropical year, calculating from spring equi-
nox to spring equinox, a slightly different year of 365 days, 5 hours, 48 minutes,
and 45.2 seconds.^89
Anyone raising a glass of wine in Virgil’s honor on 15 October is performing a
commemoration but cannot strictly be said to be observing an anniversary, a turn-
ing of the year to the same place. In fact, the chaotic state of the pre-Julian
Republican calendar means that we do not know when in plottable time Virgil was
born: 15 October 70 b.c.e.certainly did not fall on what an astronomer would call
15 October. Dates from the Republican period, unless they can be controlled by
astronomical data, are not precise indicators of time, but conventional expressions:
“historical dates, even in modern authors, should be understood as pertaining to
the Republican, not the retrojected Julian calendar: the statement that Cicero was
born on 3 January 106 b.c.refers to a.d. III Nonas Ianuarias Q. Servilio Caepione C.
Atilio Serrano coss. [ = consulibus],not to 3 January in the 106th Julian year before
the Christian era.”^90 As a result of these difficulties, historians of Republican Rome
and of the modern period use a makeshift for chronology, simply sticking to
- Years, Months, Days I: Eras and Anniversaries