Surprisingly enough, one could argue (as I often do) that the day inserted into a leap year is not that we label 29th February, but actually the 24th of February. Here I explain why, briefly.
Almost everyone assumes that the day added into February in a leap year is the 29th, but that assumption is based on a lack of knowledge of how our calendar has evolved over the centuries, subject to manifold influences of history, contingent events, and most-especially various religious considerations. Historically-speaking, the day added is that we call February 24th [sic], as I will explain. However, to jump to the punch line, the British decision to conform to the dating system of the Gregorian/Catholic calendar from the mid-18th century actually states rather simply that the day to be added is the 29th [sic, again]. This is what it says in Lord Chesterfield’s Act of 1750. That statement ignored a wealth of history and underlying biases, but it is what it is, and so legally-speaking the calendar inherited by New Zealand (and the other British colonies, including what is now the United States of America) inserts the 29th of February into a leap year.
Why, then, would I say that ‘properly’ the day added is February 24th? There are many ways to explain this… Historically, our calendar derives from the Roman Republican calendar, as revised by Julius Caesar in 46 BC (on advice from an Egyptian scholar, Sosigenes). In the era of the Roman Republic a leap month (Intercalaris, although other names were also used for it) was added from time to time at the end of February, which ended on the 23rd [sic]. Thus there was a tradition of days being added after 23rd February.
Julius Caesar’s calendar made things more regular with a leap year involving a single day every fourth year, although it took some decades for Augustus Caesar to get it right, due to the peculiarities of Roman inclusive counting. It was either AD 8 or AD 12 which was the first leap year in the four-year cycle mandated half a century before by Julius Caesar, and we know not which.
In this system of Roman reckoning, and dating, the day inserted was the sixth day counted inclusively backwards from the kalends of March (the 1st of March). This is, in our dating system, the 24th of February. Not only that, but the day was doubled; legally-speaking, the 48-hour period from the start of February 24th to the end of February 25th counted as only one day. A statute of Henry III in AD 1236, never repealed, as such, warrants as follows: “…the day of the leap year and the day before should be holden for one day.”
(In that, Henry got it wrong: in his statute’s words it looks like “the day of the leap year” is the 25th, whereas in fact it’s the 24th. In the statute it should have said “…the day of the leap year and the day following…”)
That statute is entitled De Anno Bissextili, and the word ‘bissextile’ means that the sixth day (prior to 1st March and counted inclusively, as was the Roman way) is counted twice. What we often term a ‘leap year’ is more-correctly termed a ‘bissextile year’. The term ‘leap year’ derives from the fact that a Dominical letter, used in the determination of the date of Easter, is jumped over or ‘leaped’ in such a year. Just when that leaping occurs is something else one might debate: is it at the end of February (at the junction between 29th February and 1st March) or is it at the midnight of 24th/25th February? That question is moot in that the important thing is that the leaping is, figuratively, done long before when Easter is to occur.
If you look at the mediaeval calendar for February shown in the header to this post, and inserted in its full extent below, you will see that there is a set of Roman numerals descending in value in a column with another column to its right labelled ‘kl’. That ‘kl’ indicates kalends, the word from which we derive ‘calendar’, and the Latin term for the first day of the month. The days are dated, therefore, as descending numbers from the day after the ides.
(The ides is on the 15th in four months of the year, and on the 13th in the other eight months, and is shown hereunder by the Latin word IDUS. Similarly the dates count down to the nones (depicted as NON), which is the fifth day of the month, and thereafter counting down to the ides. The most famous ‘ides’ is that warned of by the soothsayer in Shakespeare’s Julius Caesar: “Beware the ides of March”, on which date Julius Caesar was assassinated in 44 BC).
I recognise that many will find this idea, that the day added to a leap year is the 24th rather than the 29th, to be extremely odd and obscure… but it does have various implications in religious observances. In an ordinary (non-leap year) the saint commemorated on February 24th has generally been Saint Matthias (although the Catholic Church shifted him to May 14th some time ago). The thing is, in many Christian sects Saint Mattias is associated with 24th February in an ordinary year, but in a leap year he gets moved to February 25th. Similarly the saints for the other dates in February (the 25th through to the 28th in an ordinary year) get shunted a day later, whereas another saint is plonked onto the 24th (in a leap year), so that he gets only one commemorative day every four years. For more about calendars of saints’ days, one could start here.
For more about how the Roman Catholic Church counts its calendar and so on, take a look at this Wikipedia page, where the following is explained:
In leap year the month of February is of 29 days, and the feast of St. Matthias is celebrated on the 25th day and the feast of St. Gabriel of Our Lady of Sorrows on the 28th day of February, and twice is said Sexto Kalendas, that is on the 24th and 25th; and the dominical letter, which was taken up in the month of January, is changed to the preceding; that, if in January, the dominical letter was A, it is changed to the preceding, which is G, etc.; and the letter F is kept twice, on the 24th and 25th.
Now, you might well have found all the above fairly confusing. But wait… Now consider the various Orthodox churches which still use the Julian rather than the Gregorian calendar. Currently there is a 13-day difference in dates (for example, Orthodox Christmas occurs on 25th December Julian, which is the same as 6th January Gregorian). The computus employed to calculate Orthodox Easter adds a layer of confusion, in that rather than being 13 days after Gregorian Easter, instead Julian/Orthodox Easter can coincide with Gregorian/Catholic/Western Easter or else can be one, four or five weeks later.
So, what about dating and so on in a leap year? The answer is that there is a 13-day shift but then the dates towards the end of February need also to be altered in line with it being a bissextile year. A place to start trying to make head-or-tail of that is here.
I think I have written enough to leave everyone (perhaps including myself) stumped. My final thought for the day, though, is this: the day added to a leap year is properly February 24th, in our reckoning; but in addition February 24th and 25th constitute only one day, if we take a strict view of the history.
And with that, I’m off to the pub.