Easter 0033 and 1999

The years 0033 and 1999 share an interesting characteristic, the first day of the week after Passover begins is about two weeks after the first day of spring in both years. Because of this coincidence, Easter day 1999 is kind of special. It is not that this only happens once every 1000 years, in fact something like this happens at least every 19 years, because the phases of the moon have a cycle of approximately 19 years with respect to the advent of spring. But not everyone of these years is the second to last in a two thousand year sequence!

We can go back in time, figuratively speaking, and take a closer look at the year 0033. Over almost twenty centuries, one might expect that our view of the year 0033 has grown progressively more hazy. But the scriptures, history, and the calendar itself have preserved a relatively clear picture of at least some aspects of this long passed time. And, with modern advances in astronomy and calculating machines, we can enhance that picture.

Easter day has a kind of lunar relationship with the vernal equinox, the official beginning of spring. Because of this relationship, the date of the vernal equinox has been revised and a number of days have been dropped out of the count of calendar days over the centuries. The modern calendar, the Gregorian calendar, is basically the same as the calendar in use in the year 0033, the Julian calendar. But the Julian calendar placed the vernal equinox about a calendar day after its actual advent, and made a small error in its estimate of the length of the tropical year, which is the time that passes from equinox to equinox (about 365.2422 days in this century (Note 1)). The Council of Nicaea made a correction in 325 A.D., and the Gregorian calendar made a more significant one in 1582. So, these days the official date of the vernal equinox remains close to its occurrence and its actual date doesn't slowly change over the centuries, leaving the date of Easter drifting toward the summer solstice.

The Julian Calendar fixed the date of the vernal equinox at March 25th (its actual date was about March 24th at the time, 46 B.C.), and assumed that the length of the tropical year, the time from equinox to equinox, was 365.25 days. The Council of Nicaea fixed the date at March 21st, the date of its actual occurrence in 325 A.D., but left the correction of the small error in the estimate of the length of the tropical year to the Gregorian Reform, which was over twelve centuries later. Pope Gregory XIII dropped ten days out of the year 1582, in which, the actual date of the equinox was March 11th. And the Gregorian calendar omits three leap days every four centuries, thus adjusting the estimate of the length of the tropical year.

Knowing about the Gregorian Reform, we can now begin to prepare for our figurative journey back in time. From the year 0033 to the year 1582 the calendar faithfully ticked off 365 days every common year and 366 days every leap year, or 365.25 X 4.0 days every four years, 1461 calendar days. Without the Gregorian Reform, the number of calendar days from any date in the year 0033 to the same date in 2001 is simply (2001 - 0033) X 365.25, or 718812 days. Now we drop the 10 days left out in 1582 and have 718802 days, and we subtract another 3 days left out in the years 1700, 1800, and 1900 and have 718799 calendar days from 0033 to 2001 for any date. We begin our journey in 2001 for the ease of calculation. But by simply subtracting 366+365 from 718799 we have the number of calendar days from 0033 to 1999, which is 718068 calendar days from any date in 0033 to the same date in 1999.

Another way to calculate the time that has elapsed over the centuries is by thinking about time in terms of the tropical year. In these days estimates of the length of the tropical year are made to tiny fractions of a second (31,556,925.9747 seconds in 1999, according to The Astronomical Almanac for the year 1999). For our purposes we will round to the nearest second, 31556926 seconds, or about 525948.77 minutes, or 8765.8128 hours, or 365.2422 days.

Now the length of the tropical year does vary from year to year and decreases over the centuries, but these variations, at most, amount to seconds. However, over about two thousand years, even seconds can add up to significant error. So, let's let 31556930 be the average length of the tropical year over the last twenty centuries; then we can simply multiply 31556930 seconds, or 365.242245 days, by 1999 - 0033 to give us the time that has passed (or will have passed) over the intervening 1966 years, about 718066.25 days from the vernal equinox of 0033 to the vernal equinox of 1999 (Note 2).

There is a difference of almost two days between the 718068 calendar days and 718066.25 tropical year days. This difference is about what would be expected given the lack of any correction for the error in the Julian calendar's estimate of the length of the tropical year for the years 0033-325, from the first Easter to the Council of Nicaea.

Easter's relationship to the vernal equinox comes through its relationship with Passover. The bible tells us that Jesus said he would "be raised the third day." (Luke 9:22, KJV) This was also the seventeenth day of the first month of the Hebrew year, a passover feast day. Leviticus, chapter 23, verses 5 and 6 command the following: ?In the fourteenth day of the first month at even is the LORD?s passover. And on the fifteenth day of the same month is the feast of unleaven bread unto the LORD: seven days ye must eat unleavened bread.? The ?third day? of Luke 9:22 was also the third day of the seven day feast of unleavened bread, which follows the evening of the Passover supper.

The fourteenth and fifteenth days of the first month of the early Hebrew religious calendar are associated with the first full moon of the spring. The fourteenth and fifteenth days of the new year are also the fourteenth and fifteenth days of a synodic month, which is the cycle of the phases of the moon.

The length of the synodic month is about 29.53 days. Like the length of the tropical year, estimates for its length are very precise, 2551442.9 seconds for the mean length of a synodic month in 1999 (Note 3). Also like the length of the tropical year, the length of the lunar phase cycle varies; but has grown slightly longer over the centuries. But on average there are about 12.37 synodic months in a tropical year. In terms of days, a tropical year is equal to about 12(months) X 29.53 days plus about 10.88 extra days.

Now let's imagine going back in time, and looking at the moon. Let?s first make some assumptions about the average length of the tropical year and synodic month: let's again say that over the centuries the average length of the tropical year has been 31556930 seconds, and let's assume the average length of the synodic month has been 2551442.6 seconds over the last 2000 years. Now let's compute the number of synodic months that constitute the time between the vernal equinox of 0033 and the vernal equinox of 1999. That number is ((1999 - 0033) X (31556930))/2551442.6, or 24316.018 months, or 24316 months and 0.53 days. In other words, if we were sitting in a time machine with a view of the heavens, and began our journey at a new moon (a 00.00 day old moon (Note 4)), we would see 24316 full moons flash by on our way back, and arrive at a nearly new moon (about a 29.00 day old moon).

Now we are ready to take the big hop back to 0033. We start at the vernal equinox of 1999 and hop back 1999 - 0033, or 1966 years (tropical years). Recall now that 1999 - 0033 tropical years is also about 718066.25 days, and 1999 - 0033 calendar years is a little more, 718068 days. So, if we assume that the 1999 vernal equinox is about two hours before sunrise in Jerusalem on March 21st, then we hop to March 21st, 0033 plus (718068 - 718066.25) days, or March 21st, 0033 plus 1.75 days (one day and 18 hours), or about eight hours before the sunrise of the 23rd of March (Note 5). This now being Julian calendar time, we might expect the date to be March 25th, 0033, but remember the Julian calendar erred in its original fixing of the date of the vernal equinox and in its estimate of the actual length of the year. By the year 0033 about 14 and one half hours had been added to the original error. Also, the vernal equinox does not always fall on the same date, but rather varies over a period of 2 days, 5 hours, and eight minutes. In these days the actual advent ranges from March 19th at 14:06 (UT) to March 21st at 19:14 (UT) (see www.astro.uu.nl). The vernal equinox of 1999 was March 21st at 01:46 (UT). A similar range from March 24th, 0033 to March 22, 0033 would allow the equinox of 0033 to fall on March 22nd about sunset in Jerusalem (about 3/22/0033 15:00 UT).

So, here we are, near the vernal equinox 0033. The moon was about 03.29 days old when we left 1999, so at this point it is about 03.29 - 0.53 or 2.76 days old (we having traversed 24316 months and 0.53 days). In about eight hours, at sunrise in Jerusalem, March 23rd, 0033, it will be just over three days old, beginning its fourth day. This is also the fourth day of the first month of the Hebrew calendar. Now let's just ease ahead about 10 days to the fourteenth day of the month. This puts us at about sunrise April 2, 0033. And in about 12 hours the moon will be about 13.59 days old, well into its fourteenth day (Note 6).

Now, all these years, months, and weeks have been counted over the centuries, so we can go to something called a perpetual calendar and look up April 2, 0033 and see what day of the week it is, and it is the fifth day of the week, a Thursday. This is the day of the last supper. And April 5th, 0033 will be the ?third day.? This is somewhat like 1999 in terms of dates as well as the number of days after the equinox these things happened.

Now, in the year 0033, Thursday, April 2nd was about 10 days after the vernal equinox. But in the year 1999 Thursday, April 1st, will be about 11 days after the day of the equinox. The modern Hebrew calendar is not bound strictly to the phases of the moon, so Thursday, April 1st, although the day of the passover supper in 1999, follows a full moon, rather than precedes it. But nevertheless, Easter will be two weeks after the equinox, and it doesn't get much closer than that to the way it was in the year 0033!