Perusing the topics here. I am convinced that this is the right place to get some answers. Although it does appear that my question should be relegated to the "childrens table".

I am curious to understand more about the topic in archeoastronomy that is used to date ancient alignments utilizing calculations based on precession (which I very basically understand as an imaginary circle drawn in space by the earth's axis-line as the earth wobbles).

For example a "temple" in Peru has been dated as being aligned with the summer SOLSTICE(es)[edited] as of 6000 years ago by unnamed experts in the field. Other unnamed experts date it at 28000 years old - to which the first set of "experts" grudgingly concede is "remotely" possible - with an absolute qualification that there was no one here to build it that far back.(Statements of absolute certainty like this and, without proof make me itchy)

First part of question is: Why the huge discrepancy in the dating? Isn't the concept of precession a proven, quantifiable fact? (OK, that was two questions.)

Second part of question is: How do the first set of experts find themselves entrenched in their dating if they must also grudgingly agree that the second dating is remotely possible? (Both groups agree that neither the Inca nor any known predecessors actually built it.)

Anything which can help me to come to grips with this dating rationale would be great.

I am not versed in astronomy whatsoever. I can barely find Sirius on a clear night. And, math is also a foreign concept as well. What I DO understand is Ancient Civilizations. So, I would like to understand this topic more so that I can discuss it with others in the field - without sounding like a complete idiot.

[Edit to clarify before more protests stack up] I acknowledge that I did not recall and therefore omitted the fact that non-solar stellar alignments AT THE TIME of summer solstice are the basis for this alignment-dating. At least that is the theory I am discussing. (There are others) One window frames the Sun at sunrise while another frames some star. The third window... I cannot locate a theory of what it was supposed to be for. I just wanted to understand the basis for DRIFT over time. And, the accepted answer does that very nicely. Thank you everyone

  • 2
    $\begingroup$ Could you add references on the facts you are talking about? Or you just eared them somewhere? $\endgroup$
    – J. Chomel
    Oct 21, 2017 at 5:30
  • 1
    $\begingroup$ Not clear what "aligned to the equinox" means, since the Equinox is a date, not a direction. The location of the sunrise at the equinox is due East, and that doesn't vary with precession. Could it be a star with a helical rising that occurred on the Equinox 6000 years ago. $\endgroup$
    – James K
    Oct 21, 2017 at 6:45
  • $\begingroup$ My bad... I meant SOLSTICE alignment not EQUINOX alignment. Editing the question. I now see the confusion. $\endgroup$
    – Orian
    Oct 21, 2017 at 7:01
  • $\begingroup$ It is an example only from discussions among archaeologists, anthropologists as well as unpublished collaborations on the subject. To be clear; I am not asking for a rebuttal on the actual dating. It was just an example to explain why I am asking the question. Any reply which explains the rationale and any astronomical method behind dating based on angles of the solstitial sunrise is great. perhaps I should have just asked.. "What is the astronomical rationale behind dating astronomically-aligned structures using precessional calculations?" I hope that clears it up. $\endgroup$
    – Orian
    Oct 21, 2017 at 7:05
  • 1
    $\begingroup$ "Aligned with the solstice" has the same problems as "aligned with the equinox", actually. The reality is that the location and timing of equinoctial or solstitial sunrise and sunset can change slightly due to changes in the Earth's axial tilt -- but that's a 41,000 year cycle unrelated to precession. Changes due to precession involve changes in the positions of stars. $\endgroup$ Oct 22, 2017 at 10:18

1 Answer 1


The answer is fairly simple. It takes about 22,000 years for the Earth to perform 1 spin or move 360 degrees along it's axial precession. The 22,000 year cycle means that the temple was lined up at 6,000 years ago and at 28,000 years ago, as well as 50,000 and 72,000.

In other words, if we build a statue that points directly at the North Star, In a couple thousand years it will visibly no longer be pointing at the North star, but in 22,000 years, after a full circle, it will be pointing at the North star again. More specifically the North Star will apparently move in a circle that takes 22,000 years to complete, moving away from and back towards the celestial north pole.

Any skepticism on the 28,000 year date wouldn't have anything to do with alignment as the alignment works just as well with either date, and it works for any period a full circle in the past, like 50,000 years ago. The skepticism is likely related to doubt that it's that old.

Similar to the hour hand on a clock that takes 12 hours for the hour hand to return to where it started, the Earth takes about 22,000 years to return to the direction the North Pole is pointing. That means, no time between 6,000 and 28,000 years ago will line up. Those are the two most recent alignment periods. It moves in a circle in 3D, more like the picture below than like a clock.

enter image description here

As an astronomical note, I want to point out that 22,000 is somewhat inaccurate. The Earth's axial precession in and of itself is about 25,771 years, but that's only part of the movement. There's also a larger, slower apsidal precession which adds to or subtracts from the 25,771 year wobble. These together affect the alignment of the Earth to the mostly fixed stars, and that period varies, but it was about 22,000 years between 6,000 and 28,000 years ago. It can take as little as 20,800 years or as many as 29,000. Average is about 23,000.

Many Native American structures lined up quite well to astronomical alignments, so dating ancient structures by measuring when it lined up a reasonable approach. As a related, somewhat similar example, the Mayan 2012 prophecy is also based on an astronomical alignment and they were, as I recall, a year or two late, but it was still a very good estimate given when they made it. Understanding these alignments requires seeing things in 3 dimensions and that can be a little confusing to try to picture.

  • $\begingroup$ Thanks. This jibes with some of the more technical (read: difficult to understand) papers I have tried to read. If I understand this correctly, you would have to do the precessional math as well as the apsidal math to be able to come up with a reliable estimate for any position of a "fixed" body on the horizon at sunrise. i.e. Precessional math alone is not sufficient. Is that a reasonable summary statement? $\endgroup$
    – Orian
    Oct 21, 2017 at 8:16
  • $\begingroup$ @Orian yes, that's correct. The precession comes to 25,771 years, sometimes abbreviated to 26,000. Apsidal usually takes away a couple thousand years from that, but can add to it too depending on how they line up. $\endgroup$
    – userLTK
    Oct 21, 2017 at 9:28
  • $\begingroup$ Milankovich cycles... Interesting work and nice touch on the answer. $\endgroup$
    – Orian
    Oct 21, 2017 at 10:09
  • 2
    $\begingroup$ Just to clarify ... If a statute is built today that points at the north celestial pole, it will always be pointing at the north celestial pole. What star is close to that point changes because of precession. (Today the star closest to the pole is called Polaris. 5000 years ago the star named Thuban was at the celestial pole.) You could argue that the "north star" is whatever star is closest to the pole, so our statute would always point "in the direction of" the "north" star - the accuracy depends on how close the North Star is to the north celestial pole. :-) $\endgroup$
    – JohnHoltz
    Oct 21, 2017 at 13:53
  • 1
    $\begingroup$ The Mesoamerican (or "Mayan", though other cultures used it, too) Long Count Calendar has nothing to do with astronomical alignments. It's simply the accumulation of days in multiples of 18 or 20. The 2012 date was the end the 13th b'ak'tun, when the calendar roledl over from to $\endgroup$ Oct 22, 2017 at 9:30

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .