Julian, Gregorian, Easter and Leap Years

The calendar we plan our lives with has a name: it’s called the Gregorian calendar. It’s named after Pope Gregory XIII because he’s the one who implemented back in 1582 by papal decree. He did so because the calendar that had been used up to that point, the Julian calendar, implemented by none other that Julius Caesar in 45 BC* was wrong. While the solar Julian calendar was a huge improvement over the Roman lunar calendar, it wasn’t quite good enough. Here is what you need to know.

It all started with Easter. Or rather when to celebrate Easter, which was becoming increasingly difficult to plan due to errors with the Julian calendar. Way back in AD 325, the First Council of Nicea spent a lot of time deciding when everyone should celebrate Easter, concluding that it should be the first Sunday after the first full moon following the spring equinox. Mathematical tables were duly drawn up indicating when Easter should occur each year based on the Julian calendar.

But a slight inaccuracy in Julian calendar meant that, over the centuries, it slowly drifted out of sync with the solstices and equinoxes that mark the passing of the year. The prescribed dates for Easter were making less and less sense.

By the 1500s, astronomers like Christopher Clavius, the German Jesuit mathematician credited with being the architect of the Gregorian calendar, began to document the reasons why, showing that the Julian calendar’s dates for Easter didn’t agree with either the actual spring equinox or the length of the mean tropical year (the 365 days, 5 hours, 48 minutes and 45 seconds it takes for the Earth to orbit the sun). In fact, Clavius calculated, there had already been a 10-day drift in the calendar since Nicea.

The reason for the error is that the Julian calendar is slightly too long, with a year conveniently defined as 365 and one-quarter days to allow a leap year every four years. But the correct value for the length of a year is only 365.242 days, about 10 minutes shorter than the Julian year. As a result, actual celestial events were running ahead of the slower Julian calendar by about 10 minutes per year.

Since shortening each year by a fraction of a day isn’t really practical, the only way to fix the problem was to play with the way leap years were implemented, specifically to have fewer of them. While there are 100 leap-years every 400 years in the Julian calendar, the reformed Gregorian calendar has only 97 every 400 years. Those three missing days every 400 years are enough to put the calendar back on track, more or less**. The practical implementation is pretty clever too: one day is added at the end of February in years that are multiples of four, except for years that are exactly divisible by 100, unless that year is also exactly divisible by 400. In other words, in a 400 year period, one leap year is canceled every 100 years except for the last 100 years, thus three leap years are canceled every 400 years. The years 1700, 1800, and 1900 would have been Julian leap years but were not Gregorian leap years. The year 2000 was a leap year in both calendars.

Although introduced and adopted by Roman Catholic Europe in 1582, it took more nearly 350 years for the rest of the world to finally convert from the Julian calendar. A lot of this was because the new calendar originated by papal decree, which politicized the adoption process. Non-Catholic European countries slowly converted over the years for practical reasons, but Protestant Great Britain and its colonies (including those that would soon become the United States) didn’t adopt the Gregorian calendar until 1752. The USSR adopted it in 1918 and the final convert was Turkey in 1926.

And the logistics of the conversion were onerous too. All of the error that had accumulated since the First Council of Nicea had to be corrected, and this could only be accomplished by skipping calendar days to jump ahead to the correct date. When first implemented in Roman Catholic Europe, ten calendar days were skipped. The Julian calendar date of Thursday, 4 October 1582, was followed by the first Gregorian calendar date of Friday, 15 October 1582. By the time England and the its colonies converted the error had grown to 11 days, so September 1752 was a short month: people went to sleep on Wednesday September 2 and woke up on Thursday September 14.

For those who count such things, the Julian calendar is currently 13 days ahead of the Gregorian calendar.

* Note that when we talk about the Julian calendar, we’re interested in the passage of time during the year rather than the number of each year. The Julian calendar was adopted 45 years before the birth of Christ but it wasn’t until the 6th century that the year AD 1 was assigned to the presumed year of Christ’s birth, rather than to 45 years before the birth of Christ when the calendar was initiated. Note also that some historians disagree with the presumed year of Christ’s birth. By correlating historical Roman events such as the rule of Herod with Biblical text, they conclude that it’s more likely Christ was born between 6 BC and 4 BC.

** Unfortunately, the Gregorian calendar is not perfect either. Compared to the true mean tropical year, our new calendar is still off by one day every 3236 years. Not perfect, but good enough for government work.

About the Author