The Roman calendar (continued)

The need for a leap-day

Within just a few decades, it was realised that a more accurate length for the year was 365¼ days rather than 365 days, so Mercedinus was allowed to have an extra day every four years:

Jan

Mar

Apr

May

Jun

Qui

Sex

Sep

Oct

Nov

Dec

Feb

Mer

Rem

28

31

29

31

29

31

29

29

31

29

29

29

[0,22,0,23]

5

Spring equinox: Mar. 1-24 (var.); Mar. 13 (ave.)
Winter solstice: Dec. 6-29 (var.); Dec. 18 (ave.)

The problem with superstition

This calendar would have worked well if it had not been for the fact that the Romans of this time were very superstitious people. In particular, they regarded even numbers as being unlucky! A normal year was 354 days long (an even number), and January had an even number of days (so did February of course, but as it was often split into two parts, each with an odd number of days, that wasn't so bad). This "problem" was rectified by adding an extra day to January. Now both it and the (normal) year would have an odd number of days:

Jan

Mar

Apr

May

Jun

Qui

Sex

Sep

Oct

Nov

Dec

Feb

Mer

Rem

29

31

29

31

29

31

29

29

31

29

29

29

[0,22,0,23]

5

Spring equinox: Jan. 29-Mar. 23 (var.); Mar. 12 (ave.)
Winter solstice: Dec. 5-28 (var.); Dec. 17 (ave.)

A 24-year correction

All of the above was implemented by the Emperor Tarquinius Priscus in 616 BC.The obvious problem now was that the average year was 366¼ days – one day too long – which meant that equinoxes and solstices started slipping one day earlier every year. After 24 years (592 BC), the spring equinox was occurring on January 17th (average), so it was decided to omit one Mercedonian month every 24 years:

Mercedinus

[0,22,0,23,0,22,0,23,0,22,0,23,0,22,0,23,0,22,0,23,0,22,0,0]

This would have given an average year of 365.29 days, which would have been quite reasonable in theory, but in practice there were several problems:

  1. It was the job of a group of priests called the "Pontifices" (or "Pontiffs") to keep track of the calendar, and they were totally ignorant of basic astronomical facts (like how long the average year should be).
  2. It was virtually impossible to remember where they had got to in the 24-year cycle.
  3. Leap years were considered unlucky and were therefore avoided during times of crisis, such as wars.
  4. The priests were quite open to bribery, and were thus happy to make a year either shorter (to bring good luck to the briber) or longer (to bring bad luck to the briber's enemies).

The Decemvirs

The result of all this, in the end, was that by 452 BC, the spring equinox had slipped to early May. In that year, the 10-man Roman Council called the "Decemvirs" decided that by moving February and Mercedinus so that they came immediately before March, the spring equinox would be shifted by 23+23+5=51 days, back to the middle of March, where it should be. They also renamed Mercedinus "Intercalaris", which was a much more meaningful name, and declared that, instead of removing an entire month every 24 years, 7 days would be removed every 8 years. This would give an average year length of 365.375 days:

Jan

Feb

Int

Rem

Mar

Apr

May

Jun

Qui

Sex

Sep

Oct

Nov

Dec

29

23

[0,22,0,23,0,22,0,16]

5

31

29

31

29

31

29

29

31

29

29

Spring equinox: Mar. 4-23 (var.); Mar. 16 (ave.)
Winter solstice: Dec. 9-Jan. 3 (var.); Dec. 21 (ave.)

The "Year of Confusion"

The 8-year cycle was a lot easier to remember and the bribery was ended, but the Romans' superstitions still meant that the occasional leap-year was omitted in times of war or other crises, with the result that by 47 BC, the spring equinox had once again slipped to June 13th (average). Thus it was in 46 BC that Gaius Julius Caesar, under the advice of the Alexandrian astronomer Sosigenes, introduced the 445-day-long "Year of Confusion" in order to bring the spring equinox back into March once again. Each of the seven months with 29 days had an extra day added, and two further months (called Undecember and Duodecember), of 30 days each, were added to the end of the year:

Jan

Feb

Int

Rem

Mar

Apr

May

Jun

Qui

Sex

Sep

Oct

Nov

Dec

Undec

Duodec

30

23

23

5

31

30

31

30

31

30

30

31

30

30

30

30

Spring equinox: May 23
Winter solstice: Duodecember 24

The Julian Calendar

Then, the following year (45 BC), an entirely new calendar was introduced. Extra days were added to January, Sextilis and December, and Intercalaris was replaced by a "leap-day" every 4 years, following February 23rd – at least that was the plan! In actual fact, the priests misunderstood Julius's instructions because of the way the Romans counted inclusively: to them, "every four years" meant "in year 1, then again in year 4", which was, of course, only every three years!

Jan

Feb

Mar

Apr

May

Jun

Qui

Sex

Sep

Oct

Nov

Dec

31

[28,28,29]

31

30

31

30

31

31

30

31

30

31

Spring equinox: Mar. 25
Winter solstice: Dec. 25

Augustus Caesar

Julius never lived to see a single one of his leap-years, as he was murdered in on March 15th in the year 44 BC, and it was his successor, Octavian (or Augustus) Caesar, who noticed that the equinoxes had once again slipped (to March 22nd). In order to correct for all the additional leap-days that had been added, leap years were discontinued from 9 BC until AD 4. In 8 BC, the Roman Senate voted to honour Julius for introducing the new calendar by renaming Quintilis "July". They also voted to reward Augustus for spotting the leap-year problem by renaming Sextilis "August":

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

31

[28,28,28,29]

31

30

31

30

31

31

30

31

30

31

Spring equinox: Mar. 25
Winter solstice: Dec. 25

The spring equinox was now, at last, reasonably fixed at March 25th, and thus it was that the Romans’ winter solstice festival of Saturnalia came to be celebrated nine months later, on December 25th.

Later changes to the calendar

The Julian calendar had an average year-length of 365¼ days, which was pretty accurate, but not completely so – the equinoxes and solstices still slipped by about one day every hundred years, so that by the 4th Century AD, the winter solstice was occurring on December 21st, though the Romans continued to celebrate Saturnalia on December 25th. (The Romans also had an 8-day week! The 7-day week was introduced by the Emperor Constantine as part of his Christian reforms on AD 321).

By 1582, the winter solstice had slipped as far back as December 11th, so Pope Gregory XIII omitted 10 days from the calendar that year to bring the solstice back to where it had been at the time of the Council of Nicaea in AD 325 – December 21st. He also introduced the rule which says that there shouldn't be a leap year if the year is divisible by 100, unless it is also divisible by 400. This made the year length more accurate – in fact, it won't slip by another day until AD 4909!

One historical complication is that different countries introduced the Gregorian calendar at different times (Britain in 1752, Russia in 1917 and China as late as 1949). The Eastern Orthodox Church has still not done so! Another historical complication is that, although Julius Caesar had made January 1st the start of the new year, the Christian church didn't like the wild parties that took place around that time, and in AD 567, the Council of Tours declared that the new year should in future begin on March 25th. Throughout the Middle Ages, various new year dates were used in different countries, including March 1st, December 25th and "the day before Easter".

In England, from the 7th Century to the 12th Century, New Year's Day was on December 25th; from the 12th Century up to 1751, it was on March 25th; and since 1752, it has been on January 1st.

Next page: Back to Saturnalia >>

Christmas in September

Gary's Home Page

Christmas in September"
Contents Page

Gary's Home Page