One of the time periods of the Long Count of the Maya is the baktun, which lasts 144,000 days. In some texts of the Maya a new period of the Long Count is said to begin after 13 baktuns (exactly 1,872,000 days). Such a new period in the Long Count of the Maya will probably begin on 21 December 2012, when the Long Count returns to 0.0.0.0.0 (see the Historical Calendar Page), and some people expect to see special things in the sky or on Earth then that would not otherwise be expected.
The interest for 21 December 2012 seems to have been spurred by publications by John Major Jenkins, who (if I understand correctly) is of the opinion that the Maya designed their Long Count in order that the coming special date 0.0.0.0.0 corresponds to a southern solstice (when the winter begins in the northern hemisphere) when the Sun stands in the Milky Way in the sky. Mr. Jenkins reached this opinion based on extensive research of Maya texts and other traces. See //alignment2012.com for his opinion.
As far as I know, no text by the Maya has been found in which they explain (unambiguously) how they chose the day on which their Long Count should start (or on which the special date of 0.0.0.0.0 should return), so we cannot be sure why the Maya designed their calendar in that way. We can form opinions about this, based on other knowledge we have of the Maya culture and of other calendars, and depending on how much weight we assign to each fact and each opinion. Different people can form different opinions about this.
The basic arguments of many stories about 21 December 2012 seem to be as follows:
My claim is that nothing will happen on 21 December 2012 that is tied to the conjunction and that is of interest to the physical sciences. I explain my arguments below.
The conjunction is not clearly limited to a single day, but covers a period of many years.
The Milky Way is a luminous band in the sky that can be seen only (sometimes) at night, outside, from dark locations far from city lights and other lights. The Milky Way has a fixed location amidst the stars in the sky. The annual path of the Sun between the stars in the sky crosses the Milky Way in two locations, near the constellations of the Archer and the Twins. So, the Sun passes through the Milky Way in the sky twice a year.
The southern solstice (the beginning of winter in the northern hemisphere, and of summer in the southern hemisphere) always falls around 21 December in the Gregorian calendar (around 11:13 UTC on 21 December 2012). The spot where the Sun then is between the stars slowly shifts between the stars, because of the precession of the equinoxes. In about 26,000 years that spot moves once around the whole sky (compared to the stars), roughly along the ecliptic (the annual path of the Sun between the stars in the sky), so the southern solstitial point moves through the Milky Way twice each 26,000 years.
The Milky Way has no very clear boundaries, but is on average about 12 degrees wide in the sky, and the solstitial point takes about 12°/360°*26000 = about 900 years to cover such a distance.
The Milky Way has no clear central line either, so there is uncertainty about when the solstitial point crosses that central line. Different groups of people can each use reasonable definitions for the central line that yet deviate from one another. If we estimate (for example) that the uncertainty about the "best" central line of the Milky Way is half a degree (which is only a small fraction of the width of the Milky Way), then the corresponding uncertainty in the date at which the solstitial point crosses the central line is 0.5°/360°*26000 = about 36 years.
The conjunction of the southern solstitial point and the Milky Way is therefore quite rare (it happens only once each about 13,000 years) but also lasts many years.
If you accept the central line that the IAU has defined for the Milky Way (see question 480), then (based on some experimenting with planetarium program Redshift 5) the southern solstitial point crossed the central line of the Milky Way already back in 1999.
It is quite likely but not entirely certain that the Long Count date of 0.0.0.0.0 corresponds to 21 December 2012.
The exact correspondence between the Long Count and modern calendars was lost when the Spanish conquistadors destroyed many Maya documents in the 16th century. In the course of time, the beginning of the Long Count has been proposed to correspond to dates in our calendars that varied by as much as 1000 years.
It seems unlikely to me that the Maya would be (almost?) the only people to define a calendar based on a date in their distant future.
It seems very unlikely to me that the Maya (or any other people) would design a calendar such that a certain round date in that calendar would correspond to some astronomical conjunction more than 2000 years into their future. All calendars that I know that have a certain specific day as a reference point have for that reference point a date in the past of the calendar makers that is important to them, such as the birth date or the date of the beginning of the reign of an important leader, or of the founding of an important city or of their country, or the (assumed) beginning of the world.
The idea that the Long Count was designed to have 0.0.0.0.0 on 21 December 2012 was invented when someone noticed not too long ago that the next beginning of a new period of 13 baktuns (on 21 December 2012) coincides with a solstice. Munro Edmonson writes [Edmonson, p. 119]:
There was, however, nothing arbitrary about the fixing of the end of the Long Count era. Victoria Bricker has pointed out to me that 188.8.131.52.0 4 Ahau 3 Kankin corresponds to an astronomically correct winter solstice: December 21, 2012 A.D. (Julian day number 2456283). Thus there appears to be a strong likelihood that the eral calendar, like the year calendar, was motivated by a long-range astronomical prediction, one that made a correct solsticial forecast 2,367 years into the future in 355 B.C.
(The mentioned 184.108.40.206.0 corresponds to the 0.0.0.0.0 that I mention elsewhere.) Absent from this description (and from its neighborhood in the book) is any indication about this from texts of the inventors or users of the Long Count. That we think that it fits so nicely doesn't say anything about what the inventors had in mind when they designed the Long Count.
I don't think that the Maya had the knowledge to be able to predict the date of such a conjunction over 2000 years into their future to an accuracy and precision of one day.
The Maya could of course pick a certain day over two thousand years into their future, just like anyone can do, but they did not have the knowledge to be very likely to predict the correct day for the conjunction.
To be likely to calculate the correct day, the Maya would have had to be able to do the following:
Define the central line of the Milky Way very accurately, and hence also measure the positions of stars very accurately.
If that central line is not clearly defined, then you cannot determine either when the solstitial point crosses that central line. The solstitial point shifts by only about 360/26000 = 1/70 degree per year, so, to be able to predict the exact year in which the conjunction was suposed to happen, the Maya would have to define the central line of the Milky Way at an accuracy of better than 1/70th of a degree. That angular distance is at about the limit of what the human eye can distinguish. However, I have never heard of a star map from the Maya that shows the central line of the Milky Way or even the positions of stars with that kind of accuracy.
Accurately predict the motion of the solstitial point.
If you can determine from observations the time of a solstice or equinox accurately to one day (which seems reasonable if you have no modern equipment), then, to be able to make a prediction with a likely accuracy of one day, you need to have observations spanning as many years into the past as the number of years into the future for which you want to make a prediction.
To correctly predict the date of a southern solstice 2367 years into the future in the 4th century BC, the Central Americans must then have had accurate records of observations of solstices and equinoxes from the preceding 2400 years, but no indications of such records for so many years have been found.
The conjunction has no astronomical or physical scientific significance.
The positions of the planets on 21 December 2012 are not remarkable. Here are the geocentric ecliptic longitude \( λ \) and the elongation \( E \) of the Sun and all planets on 21 December 2012, measured in degrees:
For example, Mercury is then 15.5 degrees west of the Sun, and Jupiter 159.4 degrees east of the Sun. The planets aren't especially close together, and I see nothing special in their configuration on that date.
Here is a picture (made using xplns) of the location of the Sun, Moon, and planets in the sky on 21 December 2012. The round white spot (below the name "Pluto") is the Sun, and the half round spot between Jupiter and Uranus is the Moon. The white lines indicate the approximate boundary of the Milky Way. I see nothing special here, either.
Besides the basic arguments that I mentioned above, I also encounter other claims related to 21 December 2012 or to astronomical knowledge of the Maya. I note a few of them below, with my response.
The Sun (and hence also the ecliptic) does not approach the center of the Milky Way to closer than about 5 degrees (which is ten times the apparent size of the Sun and the Moon in the sky), and that happens not on 21 December but around 18 December, and not just in 2012 but in every year.
It is of course possible that the Maya recognized some other point as the center of the Milky Way than we do today, but I don't know of any Maya text that clearly defines that point, so there is no evidence that the Maya regarded 2012 as a special year in this regard.
This is total nonsense. Sunrise does not happen at the same moment everywhere on Earth, and the conjunction is of things outside of the Earth and does not depend on where you are on Earth, and not either on whether it is sunrise there at the time.
It may be that sunrise was a special time of day for the Maya and that such special conjunctions should be celebrated preferably at sunrise, but that does not say anything about the conjunction itself.
The Sun is now a few dozen lightyears north of the plane of the Milky Way. Not everybody agrees exactly how many dozen, which means that not everyone agrees where the plane of the Milky Way is near the Sun. See question 337.
If we are now a few dozen lightyears north of the plane of the Milky Way, then it is clear that we cannot pass through that plane in 2012, because to get to that plane would take a few dozen years even at the speed of light, and the Sun moves much slower than light does.
From J. Laskar et al., 1993: Astronomy and Astrophysics, volume 270, p. 522 I find that the (instantaneous) period of the precession of the equinoxes is on average about 25,678 Julian years but that it varies between about 24,820 and 26,550 years during the coming 500,000 years (ignoring the unknown influence of ice ages). Between the years −2000 and +3000 the period of the precession decreases at a rate of 0.100 year per year, and the period was 25,946 years in the year 0. (This linear approximation yields errors of at most 6 years compared to the full method, for years between −2000 and +3000.)
The period of the precession of the equinoxes is about equal to five periods of 13 baktuns, and 5 * 13 = 65 baktuns are exactly 9,360,000 days, or approximately (but not exactly) 25,626 Julian years.
Some people claim that the Maya (or their predecessors - for convenience, I refer to them all as Maya) knew about the precession of the equinoxes and tried to follow the precession using their Long Count, and that the period of the precession of the equinoxes was exactly 65 baktuns long according to the Maya. However, I have seen no indications that the Maya knew the precession, except for this rough correspondence between the period of the precession and 65 baktuns, and that correspondence could very well be a coincidence. (See at the bottom of this page for a discussion of coincidence.)
To detect precession, one needs to measure the position of the Sun between the stars in the sky very accurately and then compare those positions that were measured a century or more apart, because precession is very slow. 100 years of precession causes the phenomena of a particular star at a particular time of day to occur about one day sooner (compared to the solstices and equinoxes), or on the same day about 5 minutes earlier. However, I've not heard of any Maya texts that record measurements of the positions of one or more stars with sufficient accuracy to be able to make these kinds of comparisons.
Moreover, the period of the precession when the Long Count was invented was not equal to 65 baktuns = 25,626 Julian years, and the period of the precession is not constant anyway. In the year −500 the period of the precession was about 25,998 years and in the meantime it has declined to about 25,744 years. Only around the year 3176 will the period of the precession be equal to 65 baktuns.
The earliest recorded date in the Long Count that I know of is from the year −31, so the Long Count and the date of the end of the period of 13 baktuns on 21 December 2012 were already fixed then. If the Long Count is indeed based on the period of the precession, then the Maya would have had to measure the period of the precession already in or before the year −31 (and found 65 baktuns for it). The period of the precession was then about 25,951 years, which is 1.3 percent less than 65 baktuns, so the Maya would then have measured the period of the precession with an error of 1.3 percent.
Such seemingly small differences can have great consequences in the long run. According to the data of Laskar the vernal equinox (and hence also the southern solstitial point which is exactly 90 degrees removed from the vernal equinox) shifts over 28.46 degrees between the years −31 and 2012, but if the period of the precession were equal to 65 baktuns then the shift would have been 360/25626.28*(2012 - (−31)) = 28.70 degrees. The Maya would then have had to aim for a shift of 28.70 degrees for 2012, but in 2012 we've only had 28.46 degrees of that shift. It takes an additional 18 years (until the year 2030) before the shift has increased to 28.70 degrees.
How much extra time is needed to compensate for the difference between the true precession and the precession according to a period of 65 baktuns depends on when exactly the Long Count was fixed, because the period of the precession is not constant. The longer ago the Long Count was fixed, the more compensation is needed. If the Long Count was already fixed by the year −300, then 21 extra years would be needed.
The periods in the calendars of the Maya are always the same, and no extra periods are ever inserted or removed. Therefore, the calendars of the Maya cannot keep accurately in step with astronomical periods.
Most calendars attempt to follow one or more astronomically determined periods, such as the seasons or the phases of the Moon, but those periods are not a whole number of days long and are not equal either to some fixed ratio of whole numbers. That means that a calendar that always has the exact same periods that are always a whole number of days long cannot run in step with any astronomical period. For example, a calendar with always 365 days in a year (such as one of the calendars of the Maya) runs about (but not exactly) 1/4 day more out of step with the seasons each year, compared to the first year, so in the course of time the beginning of every season runs through all months of such a calendar.
To be able to follow an astronomical period reasonably well, a calendar must occasionally vary the length of a period, or occasionally include an extra period or omit a period. Lunar calendars, for example, have some months of 29 days and some months of 30 days. Solar calendars occasionally insert a bissextile (extra) day, and lunisolar calendars occasionally insert an embolistic (extra) month. In this way, the average length of the periods of those calendars can be a simple ratio that is close to the length of the astronomical period that that calendar tries to follow. The calendars of the Maya do not do this.
Also, astronomical periods are not constant, but slowly vary with time. For example, the length of the year of the seasons varies because of the influence of the gravity of the Sun and the other planets, and (through the precession) because of the influence of the distribution of mass in and on Earth (including the flow of matter under the surface, of water in the oceans, and of the air, and the distribution and melting of ice). The length of the synodical month increases because the Moon slowly recedes from the Earth. We can accurately predict some of these influences, but not others (including the distribution of matter in and on Earth and how that influences the rotation of the Earth).
A calendar that wants to follow an astronomically determined period must occasionally be adjusted because the length of the astronomical periods changes in a not entirely predictable way. Calendars such as those of the Maya that have fixed rules that are not adjusted based on observations cannot keep accurately following an astronomical period forever, even if they did so at the beginning.
So, the calendars of the Maya do not keep in step with any astronomically determined period.
My conclusion is that a new period of 13 baktuns in the Long Count of the Maya probably begins on 21 December 2012, but that we should not expect any special phenomena in the sky that are tied to that beginning.
It seems unlikely to me that the Maya tried to design their Long Count in such a way that the new period would start on a southern solstice in the middle of the Milky Way in their distant future, though it is not impossible.
If the Maya did design their Long Count with that conjunction in mind, then I do not think that they had enough knowledge to be able to predict the correct day of the conjunction accurately to better than a few dozen years, so if any special phenomena could be expected on the exact day of the conjunction, then they would probably not be visible on (or not visible just one) 21 December 2012.
I think that the beginning of a new period in the Long Count of the Maya in 2012 is just as unimportant as the 6000th anniversary of the Biblical creation date (celebrated in 1996), or the planetary conjunction of May 2000, or all kinds of past dates for which the end of the world had been predicted. Someone who did not hear about the prediction in advance would not have noticed anything special on those dates, and I predict that December 2012 will be like that as well.
The only effects that can be expected associated with the conjunction of 21 December 2012 have to do with the attention that people draw to that conjunction and date. If people expect that unusual things will happen on a certain date, then on that date they'll behave differently than usual, and that in itself is already an unusual thing. In this way they can fulfill their own expectations.
I believe that it is but a coincidence that 65 baktuns are approximately equal to the period of the precession of the equinoxes and that it is coincidence that the coming date of 0.0.0.0.0 in the Long Count falls on a southern solstice in the middle of the Milky Way.
You might think that it would have to be a very great coincidence, because if you choose the length of 65 baktuns at random then the chances of it happening to be about equal to the period of the precession of the equinoxes would be very small, and if you fix the date 0.0.0.0.0 at random (without regard to solstices or the Milky Way), then the chances of it happening to fall on a southern solstice in the middle of the Milky Way are also very small.
However, the probability of getting three sixes when casting three dies is also very small, but if someone does cast three sixes then that is not evidence of design or foul play. After all, if you cast the dies, then you must get some result, and if the dies are honest, then any result is equally probable, with the same very small probability, so getting three sixes is then just as likely as getting, for example, a two, and then a four, and then a three. This shows that the improbability of an occurrence is not by itself proof of design or foul play.
The observation that 65 baktuns are roughly equal to the period of the precession of the equinoxes is by itself not evidence that the Maya knew the precession, and the observation that the real 0.0.0.0.0 falls on a southern solstice when the Sun is in the middle of the Milky Way is therefore by itself no proof that the Maya must have designed their Long Count with the solstice and the Milky Way in mind.
Without unambiguous evidence from the Maya themselves we'll never be certain why they designed their Long Count in exactly the way they did.
You can read more about the calendars of the Maya on the Historical Calendars Page and more about planetary conjunctions on the Planetary Conjunctions Page.
languages: [en] [nl]
Last updated: 2020-07-18