Stars move through the sky very slowly, which is not noticeable in a human lifespan. I’m aware of proper motion of Barnard's Star and things of the sort but I’d like to obtain a noticeable record of changes from 1000s of years ago.

There are old star maps I’ve seen online but it is hard to even understand what’s going on. Are there any sources that can explain it simply to me or even show side by side example of old/ancient and new/modern?

  • 1
    $\begingroup$ Hi! If you can supply some links to sites that you've found that were hard to understand, that would be very helpful. $\endgroup$ Commented Aug 11, 2021 at 14:07
  • 2
    $\begingroup$ The Babylonians collected over 7 centuries of data in their astronomical diaries. $\endgroup$
    – PM 2Ring
    Commented Aug 11, 2021 at 20:25
  • 2
    $\begingroup$ Look up Anthony Avenis books for more on ancient astronomy. I really like his writings. He was on StarTalk , too. www.anthonyfaveni.com $\endgroup$
    – Raphael
    Commented Aug 12, 2021 at 14:24
  • $\begingroup$ Your question is excellent. You asked if we "can see" those two words are the definition of observe. You also used the term "drastic changes" this would mean noticeable also part of your question. "Noticable" is a comparative word meaning you want evidence that is large enough to prove a point as opposed to changes which are so small as to exceed the accuracy of the measuring device or process. It's like buying a digital weight scale whose minimum unit is 1 gram but you are looking for hundredths of a gram differences. That process produces no evidence let alone proof. Stephen's answer to your $\endgroup$
    – Robert1954
    Commented Oct 17, 2021 at 16:03

3 Answers 3


In practice, you're probably not going to get anything useful from ancient star maps, for several reasons:

  1. Very few of them actually survive from more than a few hundred years ago.
  2. Maps (and visual diagrams in general) are hard to accurately reproduce if you don't having printing, so a diagram from a document originally produced, say, 2000 years ago may be rather distorted if the oldest copy you have was made only 1000 years ago, and was itself most likely a copy of a copy of a copy...
  3. Even a genuine 2000-year-old map probably wouldn't have the accuracy for you to tell whether an apparent shift in the relative positions of stars was due to actual proper motion or just due to sloppiness in making the original map.

What you can potentially use are ancient catalogs with numerical coordinates of stars. This is what Edmund Halley did in the early 1700s, when he compared modern measurements of the positions of stars against their coordinates from the catalogs of Hipparchus, Timocharus, and Ptolemy, and argued that Sirius, Aldebaran, and Arcturus were all shifted by various offsets (15 to 31 arc minutes) compared to their positions 1800 years ago or so. (He also pointed out that in Tycho Brahe's catalog, finished in 1598, Sirius was about 2 arc minutes north of where it was now, indicating the stellar motion had been continuous.) See Aitken 1942 for a reproduction of Halley's 1718 publication.

The cumulative proper motions of Sirius, Arcturus, and Aldebaran over the last 2000 years amount to something like half to a full diameter of the moon, so if you had an accurate map from 2000 years ago, you could probably notice the difference (though it would be relatively subtle). But, again, I don't think such a map exists.

  • 9
    $\begingroup$ And of course, Hipparchus had access to data spanning several centuries (probably from Babylonian sources), which enabled him to discover the precession of the equinoxes. $\endgroup$
    – PM 2Ring
    Commented Aug 11, 2021 at 20:23
  • 2
    $\begingroup$ We actually have a "map" (a statue, really) representing the stars as seen in, roughly, 125BCE: it is the Farnese Atlas, it is based on Hipparchus' catalogue, and a study by Schaefer (journals.sagepub.com/doi/10.1177/002182860503600202) used the stars' positions in the atlas as a reference to give a date to which they refer. $\endgroup$ Commented Aug 14, 2021 at 15:30

The proper motion of most stars is extremely small, measured in milli-arcseconds per year, where an arcsecond is 1/3600 of a degree, and of course milli means a thousandth of that. One degree is 3,600 arcseconds, or 3,600,000 milli-arcseconds.

Except Barnard's Star. https://www.kqed.org/quest/41313/do-constellations-change-over-time

Ancient maps small insight. https://courses.lumenlearning.com/astronomy/chapter/the-sky-above/

I think, the most work that have been done since ancient times, is matching names and different culture's maps under IAU.

  • 2
    $\begingroup$ Barnard's Star is, of course, too faint to be seen with the naked eye and would not appear on any ancient star maps. There are a few stars with high proper motion that are visible to the naked eye (Groombridge 1830, 61 Cygni, Epsilon Indi), but they're either quite faint or aren't often visible to the Northern-Hemisphere civilizations that would have been likely to make detailed star maps. $\endgroup$ Commented Aug 12, 2021 at 17:32
  • 1
    $\begingroup$ @supercat, your logic is flawed. A star takes many thousands of years to reach its full brightness, and many thousands of years to fade away (unless it goes nova). So there won't be any stars that have changed noticeably in the last 2000 years, unless they exploded with a bang. $\endgroup$
    – TonyK
    Commented Aug 12, 2021 at 23:00
  • 1
    $\begingroup$ @TonyK: Fair point, though of course some stars do go supernova. Further, during the time a star is forming or fading out, the brightness would change by many orders of magnitude, so it would seem plausible that there might be stars which are easily naked-eye-visible today but weren't visible thousands of years ago, or that were naked eye visible back then but are today dimmer than other stars that weren't visible back then. I think I overestimated the probability of such "changes", but if an ancient civilization's star map has a star that isn't easily visible today... $\endgroup$
    – supercat
    Commented Aug 12, 2021 at 23:12
  • 1
    $\begingroup$ @supercat Historical novae (classical novae and supernovae) have always been described as completely "new" stars (nova = "new" in Latin; "guest stars" in traditional Chinese astronomy), not as a previously known star that got brighter. $\endgroup$ Commented Aug 13, 2021 at 9:40
  • 1
    $\begingroup$ @PeterErwin: Certainly there must have been numerous times throughout the ages when a star went from being previously-unnoticed to barely noticeable, and there would likely have been no way to distinguish at the time whether the star had previously not been there, had been present but dimmer, or had always been present at the same brightness but it had gone unnoticed until someone with particularly good eyes happened to notice it. If a star which was first cataloged in 1600 were today brighter than many other stars that were cataloged centures before, however, ... $\endgroup$
    – supercat
    Commented Aug 13, 2021 at 18:49

Are there any old/ancient star maps that we can compare to today’s and see drastic differences?

Yes, we can! Well, depending on some definitions (e.g. “star”, “map”, “drastic”).

The universe is not static (see Big Bang), everything is moving, space is expanding, stars are born, live, age, and die (see stellar evolution). Thus, we do have changes. And we are able to observe them – observation led to the theories given above. “Observe” does not necessarily mean “see with the naked eyes”, though. (It can even mean “deduce from observing a large number of objects of the same type”.) The changes generally happen at an astronomical time scale, see the answer of Lariliss. Therefore, generally the changes in one human lifetime or even several of those are too small to be seen (but see below). (If “map” means “astronomical catalogue”, position changes are seen therein.)

“old/ancient star maps”, as I understand it, are not about the position of stars in the universe! First, it is not about stars at all. “A star is an astronomical object consisting of a luminous spheroid of plasma held together by its own gravity” (https://en.wikipedia.org/wiki/Star). That is not what an ancient star map would record. That ancient star map would record visible (luminous/light reflecting/…) “dots” in the night sky. This excludes our Sun, although it is a star (by the recent definition), but includes for example supernovae (former stars), novae, comets, asteroids, nebula, other planets in our solar system, our Moon, to name a few. Changes in the observation of those objects can be seen in much shorter time spans. The (super-)novae are explicitly named such, i.e. nova meaning new (star, although it is not a new star but a bright dot in the night sky where before there was no dot visible). When “map” means “depiction of luminous dots in the night sky”, then there would be maps without the (super-)nova, and later maps with it, maps showing a comet and maps not showing it. The Moon changes from “new” (i.e. completely dark) to full to new in one lunar month. The other planets circle the Sun not synchronized to the Earth circling the Sun, thus changing their position in the sky. Comets come and go, and some come again (see https://en.wikipedia.org/wiki/List_of_periodic_comets). Thus there are observable changes.

Another point is that the “old/ancient star maps” would not record the positions of those dots in the universe, but the observed, apparent positions on the sky. There are drastic changes of the apparent positions due to changes of the position of the observer. Movement of the observer does not mean traveling from north to south pole (which would also lead to quite different observations, see Northern and Southern celestial hemisphere) but sitting down at one point of the surface of the Earth and letting the Earth do all the motion. So, what does the Earth do to the unsuspecting observer of the sky? https://courses.lumenlearning.com/astronomy/chapter/the-sky-above/ gives some good explanations (and figures!):

Earth rotates around its own axis. This leads to (apparent!) rise and down of Sun, Moon, “stars” (meaning dots in the night sky). “As Earth rotates about its axis, the sky appears to turn in the opposite direction around those celestial poles (Figure 3)” at https://courses.lumenlearning.com/astronomy/chapter/the-sky-above/. Photographs of the night sky at different times of one night show quite different positions of the “stars”.

Earth rotates around the Sun (OK, around the common centre of gravity of Sun and Earth, which lays in the Sun but not in the centre of the Sun, happy now?) in one sidereal year. When positioned at the Earth’ poles, that does not drastically change your direction of view. But when positioned at the equator, you would look in quite the opposite direction (very simple model, not drawn to scale): View from earth with 1/2 year difference: opposite directions

And in the opposite direction there are different “stars”, i.e. during the year the “stars” seem to move, see Figure 5: Constellations on the Ecliptic of https://courses.lumenlearning.com/astronomy/chapter/the-sky-above/. Photographs of the night sky at different times of the year show different stars. (Please do not stand at the poles just do disprove this, otherwise I will point out the parallax.)

And this is not complicated enough: The rotational axis of the Earth, the obliquity, is tilted and even “oscillates between 22.1 and 24.5 degrees on a 41,000-year cycle“ (ibid.). Thus a change of the tilt moves the position of the horizon against the celestial sphere, and a star, which was just barely visible above the horizon might vanish below it over the years. I remember reading about a “star map” drawn on a wall of a historic building (age: around 1,500 years), where that happened, but I do not remember the source.

And this is not complicated enough: The axis shows Axial precession: “In astronomy, axial precession is a gravity-induced, slow, and continuous change in the orientation of an astronomical body's rotational axis. In particular, it can refer to the gradual shift in the orientation of Earth's axis of rotation in a cycle of approximately 26,000 years.” This again leads to apparent movement of stars, see https://en.wikipedia.org/wiki/Axial_precession#History, shift of the equinox and (!) shift of the zodiac:

“There are two different zodiacs which divide the ecliptic into twelve zodiac signs: the tropical zodiac with twelve sections, each with a 30° arc on the ecliptic, which is astronomically oriented to the equinoxes and solstices, and the sidereal zodiac, which is oriented to the - differently sized - constellations in the area of the ecliptic. When the astrological system was developed in Hellenistic Alexandria, probably from the 3rd century BC onwards, the tropical and sidereal zodiacs still largely coincided, because the stars were equated as indicators of the seasons. Compared to that time, however, the two zodiacs today are about 30° out of phase with each other. So if, for example, the Sun is currently in the zodiac sign of Capricorn at the beginning of January, it is spatially in the constellation of Sagittarius. The reason for this is that the Earth's axis, which is decisive for the seasons, lurches - similar to a spinning top, but very slowly, namely one round in about 25,800 years; this process is known as precession. From the Earth's point of view, the point of Aries moves backwards through the constellations of different sizes at a speed of 1° in about 72 years.” (translated from https://de.wikipedia.org/wiki/Tierkreiszeichen#Tropischer_und_siderischer_Tierkreis).

“the Sun is currently in the zodiac sign of Capricorn at the beginning of January, it is spatially in the constellation of Sagittarius” – that is a drastic, observable difference, isn’t it? 30° is quite big. And “one round in about 25,800 years” is no problem: “A 32,500 year old carved ivory Mammoth tusk could contain the oldest known star chart (resembling the constellation Orion). It has also been suggested that drawing on the wall of the Lascaux caves in France dating from 33,000 to 10,000 years ago could be a graphical representation of the Pleiades, the Summer Triangle, and the Northern Crown.” (https://en.wikipedia.org/wiki/History_of_astronomy#Early_history)

More “is” than “could” was in the 2nd millennium BC: “The oldest accurately dated star chart appeared in ancient Egyptian astronomy in 1534 BC. The earliest known star catalogues were compiled by the ancient Babylonian astronomers of Mesopotamia in the late 2nd millennium BC, during the Kassite Period (ca. 1531–1155 BC).” (https://en.wikipedia.org/wiki/Star_chart) 3,500 years before present allow for some (apparent) movement: 3,500 years/72 years is nearly 50°! Nevertheless it could be argued, that the constellations did not (apparently) move (in space), but are at the same place but at another time in the year, i.e. moved in time instead of space. Fun fact: The stars (this time talking about the luminous spheroids of plasma) are not even at the positions, where we see their light today – they were at those positions when they emitted the light, which reaches us today, and then moved away. For stars visible with the naked eye, farthest away is V762 Cassiopeiae in 16,308 light-years according to https://cosmoknowledge.com/farthest-star-you-can-see-with-the-unaided-eye/.

At https://en.wikipedia.org/wiki/Archaeoastronomy#Recreating_the_ancient_sky we read:

“Not only does the Earth rotate, it wobbles. The Earth's axis takes around 25,800 years to complete one full wobble. The effect to the archaeoastronomer is that stars did not rise over the horizon in the past in the same places as they do today. Nor did the stars rotate around Polaris as they do now. In the case of the Egyptian pyramids, it has been shown they were aligned towards Thuban, a faint star in the constellation of Draco. The effect can be substantial over relatively short lengths of time, historically speaking. For instance a person born on 25 December in Roman times would have been born with the Sun in the constellation Capricorn. In the modern period a person born on the same date would have the Sun in Sagittarius due to the precession of the equinoxes.”

The alignment at Thuban instead of Polaris is not a map, but the pyramids are quite big, thus I think it counts.

And this is not complicated enough: There is also Apsidal precession, Stellar parallax, the Sun rotates around the centre of our galaxy, and our galaxy, the Milky Way, moves in our Local Group, which also moves. There is also the apparent movement of the position of stars, which for real is the deflection of light by a mass, which the light passes (see e.g. Eddington experiment and Gravitational lens).

We have (apparent) change of position of stars (et al.), ancient viewed positions are shown to be different from recent ones. The question in the text below the title of OP’s question is somewhat different, Peter Erwin and Lariliss already answered that and Jacopo Tissino added the >>"map" (a statue, really) representing the stars as seen in, roughly, 125BCE<<. A paper (or papyrus or drawing in a tomb) remains to be found.


You must log in to answer this question.

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