The date of Christmas is a matter many find confusing, and yet the adopted anniversary is easy to understand if you follow through the history, astronomy and human biology that are involved.
Why is the Nativity commemorated on December 25th, when it is clear Jesus was not actually born on that date? And how can a year be termed “Before Christ” if he was born in that year? This is a topic that crops up often in the media and online discussions as the holiday (‘holy day’) season approaches… and yet I have never seen a correct explanation being given.
The answers are a mixture of science (astronomy and physiology), religious observances and historical contingencies, but if one follows it through step by step it’s all quite simple.
Before going further I should make clear that I have no religious beliefs myself, no axe to grind. Twenty years ago I was writing a book about the history and astronomical basis of the calendar (or, better, the distinct calendars used by different cultures, and how they evolved). In doing so I learned that many aspects of the various dating systems which have been employed – indeed are still employed – are quite complex, whereas others are relatively straightforward. I’d argue that Christmas comes under the latter heading.
First, let us think about the calendar in use in that era two millennia ago, when the Romans ruled in Egypt, Judaea and Palestine. For some centuries the calendar of the Roman Republic had contained twelve months ending with December, February and January, in that order. February was regarded as being unlucky, and occasionally a leap-month (called Mercedonius or Intercalaris) was inserted into it, for a variety of reasons, often political rather than seasonal.
March (named for Mars, the god of war) was the first month of the year, in that the new consuls were appointed from March 15th (the Ides of March; vide Julius Caesar) and the campaigning season began with the vernal equinox ten days later; that is, the weather was improving after winter and the legions could be marched to the battlefronts. As the Roman Republic expanded, by the mid-second century BC it became clear that the consuls needed to be appointed and come into power earlier, so as to allow time for the army to be organised for the fighting ahead, and to execute the more-distant marches to the borders of the Roman domain as they were pushed outwards.
And so the calendar was re-organised from 153 BC. September through December have names indicating them to be the seventh through tenth months, but now they were shifted to become the ninth through twelfth whilst retaining their misleading labels to the present. Have you never thought how peculiar that is?
February and January needed to have their order reversed, because the latter is named for Janus, the god who looks in both directions (i.e. to the year before and the year following). Previously this was fulfilled by January being the last month of the year; now it became the first. With that, January 1st became the start of the year, the inauguration day (dies solennis) for the consuls. That is the origin of our present custom of the year beginning with that date. As will become apparent, the date of Christmas Day also hinges on this.
Do not imagine, though, that January 1st has persistently been the date of New Year in the tradition we have inherited across the millennia. Many alternative dates have been used, and it was quite confusing for people in the Middle Ages to move between European states, which variously used January 1st, March 1st, March 25th, Easter (which moves by up to five weeks), September 1st, September 24th and Christmas as New Year. Not only that, but they did not necessarily use the same year count. For example, in mediaeval Pisa they counted from March 25th in 1 BC, whereas in nearby Florence the radix of their calendar was March 25th in AD 1.
Don’t laugh. This still affects many people. If you look at adverts in New Zealand newspapers with regard to shifting UK pensions to NZ, you will find the date April 6th cropping up. This is because that’s the day on which the personal income tax year begins in the UK. It represents an eleven-day adjustment made in 1752, when New Year was shifted from March 25th (indeed in the Florentine style, counting from that date in AD 1) back to January 1st, and the leap-year cycle and Easter calculation used were also brought into agreement with the Catholic Gregorian calendar, which had been instigated in 1582. New Zealand inherits its calendar from that British Act of Parliament.
Such changes in date can catch out the unwary. For example a book might tell you that Queen Elizabeth I died in 1603; in fact she died on March 24th in 1602, the last day of that year on the calendar in force in Britain at that time. I began my book by writing about how George Washington’s birthday got moved both in date and year by the British calendar reform; in the U.S. the public holiday called Washington’s Birthday is never celebrated on either of the dates that he would have recognised, but instead perennially occurs between those dates.
Now let us get back to discussing Christmas Day. The essential point to be remembered from the above is that there has been a long-term (more than 2,000 year) custom of marking the start of a New Year as January 1st. We can declare other types of year (regnal years, tax years, school and university years, sports years), but January 1st is nowadays generally used across cultures as the international standard, notwithstanding local or religious calendars.
Next, let us dispose of the confusion between the actual date of the Nativity and the traditional date (i.e. December 25th in the year we now label as 1 BC). There have been many proposals for the actual date. The one I favour is that the Star of Bethlehem was actually a comet observed in 5 BC, and the combination of that celestial phenomenon with what we know about the Hebrew calendar coupled with the biblical accounts and historical constraints (such as King Herod dying in 4 BC) point towards a birthdate in the middle of April of that year.
Why the apparent four-year error? The likely explanation seems to be that when tables of dates for Easter were being compiled in the sixth century by a monk in Rome named Dionysius Exiguus, a recorded statement that Jesus was born in the 28th year of the reign of Augustus Caesar was interpreted as implying a year count from when the first Emperor took that appellation in 27 BC, rather than reckoning from 31 BC when he assumed power under his original name of Octavian, after defeating Antony and Cleopatra at the Battle of Actium.
The next question people ask is often this: even if it (December 25th in 1 BC) is not the real date, but a traditional designation, how can we possibly label the year of Jesus’s birth as 1 BC (Before Christ)? The question is based on a cultural misunderstanding. In Orthodox Judaism an infant boy’s life may be taken to begin not at birth, but with his circumcision and naming on the eighth day post-partum:
And when eight days were accomplished for the circumcising of the child,
his name was called Jesus… (Luke 2:21).
The week-long delay is to allow the infant’s blood-clotting capabilities to develop – the first example of where human physiology enters the calculations. This ceremony therefore occurs on the same day-of-the-week as the child’s birth, the octave day. (Note that our seven-day cycle derives from the Jewish septenary practice reinforced by the astrology of the Babylonians, who used a planetary week based on the seven ‘planets’ then known: Sun, Moon, Mercury, Venus, Mars, Jupiter and Saturn, these giving us the names of our days-of-the-week; in contrast the Romans used an eight-day week, the seven-day cycle eventually being legalised by Constantine the Great in the early fourth century.)
If you count forward to the eighth day after December 25th, of course you come to January 1st. The traditional lifetime of Jesus begins with his circumcision and naming on that date… In AD 1. Various Christian denominations still celebrate that date as the Feast of the Circumcision though others have either re-named it or dropped it altogether. In an accompanying graphic I show how the event was inscribed in the Anglican Book of Common Prayer for 1599.
The front cover of the Booke of Common Prayer for 1599 along with some of the text that refers specifically
to the Feast of the Circumcision, an important juncture in the annual liturgical round and the event which
defines the start of our dating era (i.e. the Anno Domini system).
To a chronologist, counting years as starting on January 1st is often termed ‘the style of the circumcision’ (stylus circumcisionis), although the phrase ‘the style of the common people’ (stylus communis) is also used in this connection, perhaps for reasons of sensitivity.
The proper way to reckon the relevant (traditional) dates is to count backwards, then. What might be regarded in the Judaic convention as the start of the life of Jesus Christ was January 1st in AD 1. The Nativity was on December 25th, a week before. That was also the traditional date of the (northern) winter solstice, about which I will say more below. Before we look at that, let’s consider an interesting coincidence between astronomy and human biology, and its influence on calendars. This is the second way that physiology enters into things.
The human gestation interval is often stated as being nine months, or forty weeks. Of course there are substantial variations, and it also depends upon whether one counts from the last menstruation, or from ovulation, or fertilisation shortly thereafter, but these are modern realisations. In the past, nine months was the benchmark.
If we count back nine months from the winter solstice on December 25th, we come to the vernal equinox on its traditional date of March 25th. This is why that date has frequently been used as the radix for calendars, the vernal/spring equinox being the natural start of the year, the end of winter (in the northern hemisphere) and the start of spring. The equinox is still used by some; for example the Jalali calendar used in Iran and elsewhere marks New Year (Nowruz) on the day which the vernal equinox occurs. In terms of astronomy, this is more accurate than the system we use (the Gregorian calendar). So much for Western science.
It seems that Dionysius Exiguus based his chronological calculations on the dates for the vernal equinox and winter solstice in 1 BC, and therefore his reference points (using modern scientific terminology) for the arrival of Jesus on Earth. March 25th is nowadays celebrated as Annunciation Day, or the Feast of the Incarnation, when the angel Gabriel is said to have come and told the Virgin Mary that she would bear the Son of God; it is also the Feast Day of St Mary, or Lady Day. Obviously it is an important juncture for Christianity. To Dionysius this was the pivotal point: the Annunciation, rather than the Nativity.
Much is often made in discussions with regard to Christmas Day having been set so as to avoid coinciding with Saturnalia, the Roman mid-winter festival which seems to have consisted of much debauchery and inbibing, running from about December 17th through to the 23rd. This terminated with the winter solstice, though the dates for that astronomical event jumped around due to the vagaries of the calendar in the Roman Republic. Even after Julius Caesar regulated the calendar from 46 BC there was still some slippage because after his assassination the Romans mistakenly inserted leap days every three years instead of four; that’s why we cannot be sure of the day-of-the-week for January 1st in AD 1, proper adherence to the Julian calendar not starting until either AD 8 or AD 12, and we don’t know which of those is correct.
The situation persists, though, that the traditional date of the winter solstice is December 25th, that is also the date of Christmas Day for reasons explained above, and yet the actual (astronomically-defined) instant of the winter solstice varies across a range of 53 hours, most often occurring nowadays on December 21st or 22nd. Much of that wobble of 53 hours is due to the fact that the leap-year cycle chosen for the Gregorian calendar is a second-best solution, but there is still an offset between the traditional date (the 25th) and the actual date (21st/22nd) which averages more than three days. How come?
The answer is that when the Gregorian reform occurred in 1582, making up for the one-leap-year-every-four-years rule of the Julian calendar being slightly too often, the decision was made to knock just ten days out of the calendar. The effect of that was to return the dates of the equinoxes and solstices to those occurring at the time of the Council of Nicaea in AD 325 (when the first doctrine of the Church, the Nicene Creed, was laid down under the eye of Constantine), rather than in the era of Jesus Christ, which would have required around 13 days to be dropped.
Of course in the sixteenth century the astronomical movements involved were not understood, and I could make a case for 12, 13, 14 or 15 days all having merits in terms of a retrospective calendar correction, but let’s leave that aside. Apart from anything else we know now that the intervals between the equinoxes and the solstices are not constant, and that negates any idea of setting up a ‘perfect’ calendar.
The date of the vernal equinox was also affected by deleting just ten days from the year 1582, obviously. While March 25th remains the traditional date for the equinox, and therefore the Feast of the Annunciation/St Mary’s Day and so on, in fact the instant of that equinox occurs (again varying across a 53-hour range) on March 19th, 20th or 21st. For purposes of calculating Easter, the Catholic Church effectively declared the whole of March 21st to be the ecclesiastical equinox, whilst the astronomically-defined equinox is usually earlier than that; in fact, the next true equinox on March 21st (defined in terms of Coordinated Universal Time, UTC) will not occur until the year 2103.
Parts of the liturgical calendar entries for March and December in 1599 from the Booke of Common Prayer.
These show that the occurrences of the vernal equinox and the winter solstice were well before their
traditional dates of March 25th and December 25th, the Feasts of the Annunciation and of Christmas,
both of which being preceded by a fast day. At that time the Church of England was still using the Julian
calendar, whereas the Roman Catholic Church had adopted the Gregorian system from 1582.
The Gregorian reform pushed the dates for the equinoxes and solstices later by 10 days, but not the full
12 or 13 days required if one wants to agree with the times of those events in the era of Jesus Christ.
This distinction between the equinox defined by the clerics and the equinox defined by astronomers has at various times led to people questioning what is going on, because they think that the mnemonic for Easter – that it’s the first Sunday after the first full moon after the equinox – is literally the case. It’s not. The ‘equinox’ in the mnemonic is not the astronomical juncture; the ‘moon’ there is not the moon in the sky, but rather the fictitious ecclesiastical moon; and ‘Sunday’ must depend on which prime meridian you choose for timekeeping in this regard – Jerusalem, perhaps?
Near the beginning of this long blog I wrote that how Christmas Day was fixed on December 25th is relatively straightforward. You might not agree, having ploughed through my description above. So let me summarise:
- The traditional date of the Nativity is well-recognised to be some (likely four) years after the actual birthdate of Jesus Christ.
- In the era in question January 1st was regarded as being the start of each year.
- The traditional date of the vernal equinox was and is March 25th, and that of the winter solstice December 25th.
- The human gestation period corresponds with the interval between those two.
- In the Judaic tradition a male child is circumcised and named a week after being born, and that might be regarded as being the start of his life.
- It was the work of Dionysius Exiguus that later led to the adoption of the BC/AD scheme, and he was responsible for the four-year error (5 BC vs. 1 BC) when in the early 6th century he was calculating new Easter tables; he regarded the Annunciation occurring on the day of the vernal equinox (March 25th) as a fundamental consideration.
- Stepping forwards nine months from the Annunciation on March 25th one arrives at a date for the Nativity/Christmas Day of December 25th, the traditional winter solstice.
- Step forward another week and one arrives at January 1st in AD 1 for the circumcision and naming of Jesus, and therefore the start of the era reckoning which we still follow today.
Phew! That was a long story to tell. Thankfully this blog post is a Christmas special, and so I did not run out of space.