Julian calendar

• 1. Introduction

  ›

  The Julian calendar gains one full day against the sun every 128 years, a slow and silent error built into its very design. It was proposed in 46 BC by Julius Caesar and took effect on the 1st of January 45 BC, by his edict. For more than 1,600 years it ruled the Roman Empire and most of the Western world, until 1582. That year is when a pope decided enough drift was enough. But the calendar Caesar built did not simply vanish. It still governs religious life in parts of the Eastern Orthodox Church, in parts of Oriental Orthodoxy, and in the Berber calendar of the Maghreb. How did a Roman reform survive into the present? Why did a system meant to track the seasons forever slowly fall behind them? And what happened in the year before it began, a stretch so chaotic that history remembers it as the last year of confusion?

• 2. The Last Year Of Confusion

  ›

  46 BC ran for 445 days, the longest year in Roman history. Caesar lengthened it to compensate for the intercalations that had been missed during his time as pontifex. The year had already been stretched from 355 to 378 days by a regular intercalary month inserted in February. Then, probably shortly after his return from the African campaign in late Quintilis, Caesar added 67 more days. He did this by inserting two extraordinary intercalary months between November and December. In letters written at the time, Cicero called these months Intercalaris Prior and Intercalaris Posterior. There is no basis for the claim that they were named Undecimber and Duodecimber. Those terms arose in the 18th century, over a thousand years after Rome fell. The number 67 may have compensated for three omitted intercalary months, since 67 equals 22 plus 23 plus 22. The arrangement also made the distance from the 1st of March 46 BC, the old Roman New Year, to the 1st of January 45 BC exactly 365 days. Because 46 BC closed out a series of irregular years, it earned its grim nickname. The new calendar began operating only after this realignment was complete, in 45 BC.

• 3. A Calendar Open To Abuse

  ›

  The old Roman year held 12 months and 355 days. Between February and March, a 27- or 28-day intercalary month called the Mensis Intercalaris was sometimes inserted. It was formed by adding 22 or 23 days after the first 23 days of February. The last five days of February, which counted down toward March, became the last five days of Intercalaris. If managed correctly, this could have kept the Roman year roughly aligned with the seasons. The problem was who held the power to manage it. The pontifices who controlled intercalation were often politicians. A Roman magistrate's term matched a calendar year. So a pontifex could lengthen a year when he or an ally held office, and refuse to lengthen one when his opponents were in power. Caesar's reform aimed to end this manipulation permanently, by building a calendar that stayed aligned to the sun with no human hand needed. The fix proved its worth quickly. Varro used the new calendar in 37 BC to set fixed dates for the start of the four seasons, something that would have been impossible only eight years earlier.

• 4. Borrowing From Egypt And Greece

  ›

  Caesar probably first met a drifting calendar in Egypt, where a fixed year of 365 days slid by one day against the sun every four years. An attempt to add an extra day every fourth year had failed there in 238 BC, in the Decree of Canopus. Caesar landed in the Nile delta in October 48 BC and was soon drawn into the Ptolemaic dynastic war, especially after Cleopatra arranged to be introduced to him in Alexandria. After imposing peace, a banquet was held to mark the event. The poet Lucan described Caesar speaking with a wise man named Acoreus during the feast, declaring his intention to build a calendar more perfect than that of Eudoxus. Caesar returned to Rome in 46 BC. According to Plutarch, he summoned the best philosophers and mathematicians of his time to solve the calendar problem. Pliny credits the astronomer Sosigenes of Alexandria as his principal aide and the reform's chief designer. Sosigenes may also have written the astronomical almanac Caesar published to ease the change. Macrobius adds that a certain Marcus Flavius assisted as well. The final calendar combined three traditions: the old Roman months, the fixed length of the Egyptian year, and the days of Greek astronomy. Older systems had each fallen short. The octaeteris, a cycle of eight lunar years popularized by Cleostratus, runs 1.53 days longer than eight mean Julian years. The 19-year cycle of Meton, at 6,940 days, ran six hours longer than the mean Julian year, which became the basis for the 76-year cycle of Callippus.

• 5. Building The Months

  ›

  Ten days were added to the old 355-day Roman year to make a regular Julian year of 365. Two extra days went to January, Sextilis, and December. One extra day went to April, June, September, and November. February was left untouched in ordinary years and kept its traditional 28 days. The ordinary lengths of all the months were set to the same values they still hold today. The reform did not change how days within a month were counted, which stayed anchored to the Kalends, Nones, and Ides. Macrobius states that the extra days were added immediately before the last day of each month, so as not to disturb the position of established religious ceremonies relative to the Nones and Ides. The inserted days were first marked as dies fasti. The Romans later renamed two months after their leaders. Quintilis became Iulius, our July, in 44 BC, honoring Caesar because it was the month of his birth. Sextilis became Augustus, our August, in 8 BC. A senatus consultum quoted by Macrobius explains that several key events in Augustus's rise, including the fall of Alexandria, fell in that month.

• 6. The Bissextile Puzzle

  ›

  The old intercalary month was abolished and replaced by a single leap day. That day was dated as ante diem bis sextum Kalendas Martias, the sixth doubled day before the Kalends of March. From this came the English word bissextile, and the leap year was called the annus bissextus. Where exactly that day sat in the early calendar remains disputed. The 2nd century jurist Celsus described a 48-hour day split into two halves, calling the intercalated day the posterior half. An inscription from AD 168 records that a.d. V Kal. Mart. was the day after the bissextile day. The 19th century chronologist Ideler argued that Celsus meant posterior in a technical sense, naming the earlier of the two days, while Mommsen and others read it in its ordinary meaning. In time, the bissextile day became the earlier of the two for most purposes. In 238, Censorinus placed it after the Terminalia on the 23rd of February, followed by the last five days of the month. Later writers followed this rule, including Macrobius around 430, Bede in 725, and the medieval computists who calculated Easter. Yet Celsus's definition survived for legal purposes. It entered Justinian's Digest and the English Statute De Anno et Die Bissextili of 1236, which was not formally repealed until 1879.

• 7. The Priests' Mistake

  ›

  For thirty-six years, the priests inserted a leap day every three years instead of every four. Accounts of this error appear in Solinus, Pliny, Ammianus, Suetonius, and Censorinus. Macrobius explains that the priests added the intercalary day at the beginning of each fourth year rather than at its end. By the time the mistake was caught, twelve intercalary days had been inserted instead of the nine actually due. Augustus ordered the correction. Twelve years were allowed to pass with no leap day, absorbing the three extra days the priests had added. In the reconstruction by Scaliger, published in 1583, the triennial leap years ran from 42 BC to 9 BC. After 9 BC came twelve years without leap years, so the days Caesar would have placed in 5 BC, 1 BC, and AD 4 were skipped. After AD 4 the calendar ran as Caesar intended, and the next leap year was AD 8. Scaliger established that the Augustan reform took effect in 8 BC. Because of this confusion, the exact Julian day behind any Roman date before March of 8 BC cannot be known for certain, except for dates used in Egypt in 24 BC, which are secured by astronomy. A papyrus discovered in 1999 records astronomical events of 24 BC in both Egyptian and Roman calendars. From it, Alexander Jones concluded that Egyptian astronomers had used the correct Julian calendar even while Rome's priests stumbled.

• 8. Survival And Replacement

  ›

  Pope Gregory XIII promulgated his revised calendar in 1582, because the Julian year of 365.25 days ran longer than the natural tropical year. By that date, the 21st of March, the base date for calculating Easter, had drifted ten days from the March equinox. That equinox was reckoned to have fallen on the 21st of March in 325, the year of the Council of Nicaea. The Gregorian reform kept the same months and lengths but dropped leap days in century years not divisible by 400, cutting the average year to 365.2425 days. Catholic countries adopted it at once. Protestant countries followed slowly over the next two centuries. Most Orthodox countries kept the Julian calendar for religion but switched their civil calendar to Gregorian in the early twentieth century. Adoption stretched across centuries: Great Britain waited until 1752, and Russia until 1918. This gap is why the Russian Revolution of the 7th of November 1917 by the new calendar is remembered as the October Revolution, since it began on the 25th of October by the old. The Gregorian calendar has now replaced the Julian as the civil calendar everywhere, with Greece the last to do so, in 1923. Still, the old calendar endures. The churches of Jerusalem, Russia, Serbia, Montenegro, Poland, North Macedonia, and Georgia celebrate the Nativity on the 25th of December Julian, which is the 7th of January Gregorian until 2100. The Berbers of the Maghreb still keep the Julian year. And on Foula, a remote island in Shetland, a small settlement still marks its festivities by Caesar's reckoning.