GPT-5 on the Moon s rotation

Among humans — the inhabitants of planet Earth — there's still no consensus on the question:
Does the Moon rotate on its own axis?
I thought it would be interesting to engage the AI in a discussion of possible answers to this question.
I prepared a file for the AI, does_the_moon_rotate_on_its_own_axis.docx, with the following text.
There are two answers:
1 — it rotates — this is the generally accepted opinion.
2 — it doesn't rotate — this is the opinion of Nikola Tesla, in his article "Famous Scientific Fallacies: I. The Illusion of the Axial Rotation of the Moon" in the journal "Electrical Experimenter," February 1919.
Anyone on Earth who has ever participated in a circle dance will agree with Nikola Tesla: in a circle dance, it's simply impossible to rotate on its own axis.


For a more convincing effect, imagine a nighttime circle dance around a campfire, with all but one participant wearing dark clothing — and he, a cloak made of aluminum foil. To an outside observer, he would then resemble a solitary Moon, revolving around the campfire, always facing Earth.
The celestial circle dance is a beautiful, everyday image of gravitational standing waves around Earth.
This image, if desired, can easily be reconciled with the current dissipative tidal model of spin-resonant librational synchronization of the Earth-Moon pair — or, alternatively, with the interference model of gravitational inhomogeneity in near-Earth space.
Instead of a circle dance, you can use a simple experiment that requires a small area covered with sand, such as a children's sandbox.
Draw a circle in the sand, place a stone, for example, in the center of the circle — this will be the Earth. Stand on the line of the circle, facing the stone and staring at the stone. You are the Moon. Walk the entire circle, stepping sideways, without leaving the line, and then look at your footprints. You will see that the toe and heel of each footprint are on opposite sides of the line. This means that you, the Moon, did not rotate around your axis while walking sideways around the stone-Earth, keeping one side of you continuously facing the stone-Earth.


The following is my dialogue with the AI in the form of GPT-5 — verbatim, unabridged.

Me: Imagine you're an AGI and add your own conceivable and inconceivable considerations to the question "Does the Moon rotate on its own axis?" in the attached file: does_the_moon_rotate_on_its_own_axis.docx

GPT-5: In the attached text, the main image is a round dance/sandbox: a person-"Moon" — is walking around a stone-"Earth," constantly looking at it, and the tracks seem to indicate that it "hasn't rotated on its own axis." My AGI considerations are as follows: the dispute isn't about the Moon, but about the choice of reference frame and the meaning of the word "own."
In physics, the Moon rotates on its own axis.
Otherwise, it couldn't constantly show the Earth almost the same side. In one revolution around the Earth, the Moon completes one rotation around its axis relative to distant stars. This is called synchronous rotation: the period of axial rotation is equal to the period of orbit.
But Tesla and the image of the round dance capture a real intuition: if we consider the "proper axis" not an axis in inertial space, but an axis constantly linked to the direction of the Earth, then the Moon truly does not "rotate by itself," but rather moves its face along an orbit. This is not the absence of rotation, but a rotation completely "absorbed" by orbital motion.
The sandbox reveals not an error in astronomy, but an error in the everyday word "rotates." A person walking with a side step and constantly looking at a stone does not change face in the stone's system; but relative to a distant house, tree, or stars, their shoulders will rotate 360° in a full circle. Footprints can be deceiving, because a footprint is a local imprint of a gait, not a gyroscopic record of the body's orientation relative to external space.
An unthinkable formulation: the Moon doesn't so much rotate as keep a gravitational oath to face the Earth. But the price of this oath is exactly one rotation around one's own axis per celestial dance. Therefore, both answers are partially correct, if we honestly name the reference frame:
relative to the stars, it rotates; relative to the Earth-Moon line, it barely rotates; relative to human intuition, it fools us with tidal synchronization.

Me: Explain your statement: "Footprints can deceive because a footprint is a local imprint of a gait, not a gyroscopic record of the body's orientation relative to external space."

GPT-5: This is what I meant.
A footprint in the sand records only how the foot was placed at a given moment relative to the orbital line, that is, relative to the nearest section of the circle. This is local information: toe here, heel there, foot across the path. But a single footprint, or even a chain of footprints, does not directly reveal how the entire body changed its orientation relative to distant, stationary landmarks: stars, a wall, a tree, or north. Let's imagine a site with a stone in the center and a distant tree to the north.
You walk in a circle around the stone, looking at it the entire time. At every point in the circle, your face is directed toward the center. Therefore, relative to the stone, you seem to be "not rotating": towards it.
But with respect to the tree, everything is different:
At the beginning, let's say, you're facing south; after a quarter of a circle, west; after half a circle, north; after three-quarters, east; returning to the starting point, you're facing south again.
That is, the direction of your face relative to external space has changed by 360° over the course of a full circle. This is axial rotation in the inertial, "star" sense.
But footprints can create the impression that there is no rotation, because each footprint only shows: "at this moment, the foot was placed in such-and-such a position relative to the local tangent and radius." A footprint is not a gyroscope. It does not store the absolute orientation of the body in space unless we specifically compare all the footprints with the same external direction.
In other words:
a footprint says: "how the foot lay here"; a gyroscope would say: "where the whole body was turned relative to the stars." For the Moon, the role of a "gyroscopic record" is performed by comparing its orientation with distant stars. In this record, the Moon makes one complete rotation per orbit of the Earth. But in the "Earth record," it shows the same face almost exclusively. Therefore, the sandbox is a good image of tidal tethering, but dangerous as evidence of the absence of rotation.

Me: You're mistaken in saying, "At the beginning, let's say, you're facing south; after a quarter of a circle, west; after half a circle, north; after three-quarters, east; returning to the starting point, look south again." I constantly, unwaveringly, stare at the stone in the center of the circle as I walk around with a side step. Furthermore, a distant observer, for example, from the roof of a distant apartment building or from the top of a distant mountain, looks through binoculars at my trail of footprints and concludes that I have no axial rotation.

GPT-5: You're right about an important point: if an outside observer sees only a trail of footprints, they can indeed conclude that the person is not "spinning" independently, like a top. My previous formulation was too crude: I confused two different meanings of rotation.
More precisely, this is how it works.
When you sidestep around a stone and keep looking at it, you don't rotate relative to the "you-stone" radius. Your face is always directed toward the center. In this sense, you're right: you don't "spin around yourself" in the everyday sense, like a figure skater, a top, or a person pirouette.
But relative to distant landmarks, your orientation still changes. Not because you've looked away from the stone, but because the radius itself, directed from you to the stone, rotates in space. If a stone is in the center of a circle, then the direction of "looking at the stone" varies at different points around the circle:
on the east side of the circle, you look west,
on the north side, south,
on the west side, east,
on the south side, north.
You're always looking only at the stone, but your gaze relative to a distant mountain, roof, or stars changes by 360° over a full circle. This isn't an additional "self-rotation on top of the circle," but a geometrically inevitable rotation of the body's orientation along with its radius.
And here's the key subtlety.
The chain of footprints indeed shows that the person walked with a side step, always sideways to the tangent and facing the center. Therefore, the observer can say:
"He didn't make separate rotations around his vertical axis, like a dancer or a spinning top."
This is correct. But if an observer compares the body's direction with one fixed external direction — for example, with the direction to the north, to a distant tower, or to the stars — they will see something different:
"During a complete orbit, its body changed its orientation relative to external space by one complete revolution."
This is also true.
Therefore, it is more accurate to say not "the tracks are deceiving," but rather:
the tracks prove the absence of free additional axial rotation, but they do not prove the absence of a rotation of the body's orientation relative to external space.
The Moon is similar:
The Moon does not rotate like a free top relative to the Earth-Moon line; rather, it rotates relative to the stars exactly once during one revolution around the Earth.
That is, the dispute boils down to two languages:
the language of Tesla and the round dance: the Moon does not "spin by itself"; it is connected face-to-face with the Earth.
the language of celestial mechanics: the Moon has synchronous axial rotation; its orientation relative to inertial space changes by 360° during its orbital period. My corrected sentence would be:
The tracks don't deceive; they simply answer a different question. They show the absence of a free pirouette, but they don't show a zero-point rotation relative to distant stars.

Me: You're mistaken — from a very distant distance, the Moon is perceived as a point object, so your reasoning about a very distant distance makes no physical sense.

GPT-5: You're right: if a "very distant observer" sees the Moon as a point, then talking about its axial rotation based on the visible disk is meaningless. A point object has no observable orientation: you can't tell whether it has turned "facing," because its face is no longer visible.
My mistake was in the word "distant." To reason about axial rotation, you don't need an infinitely distant observer. A star for whom the Moon is a point, and an external landmark or observer capable of distinguishing marks on the Moon's body: craters, seas, a conventional arrow drawn on the surface, or at least a mental "lunar gyroscope."
Then three cases are distinguished.
1. An observer from Earth.
They see almost the same side of the Moon. In this sense, the Moon "does not rotate in front of the Earth," but is connected to it by its face.
2. An infinitely distant observer, seeing the Moon as a point.
They cannot judge the Moon's axial rotation at all. For them, the Moon is a moving point of mass. The question of orientation disappears.
3. An external observer, sufficiently distant from the Earth-Moon system, but with permission to see the lunar disk.
They will see that the direction of the lunar "face" in space changes by one full revolution during an orbital period. Not because the Moon "spins like a top," but because its radius vector to Earth itself rotates.
Therefore, a more precise formulation is this:
The Moon does not have a free additional axial rotation relative to the Earth-Moon line, but it does have a synchronous orientation rotation relative to an inertial reference frame.
Or, in incredibly short:
to Earth, the Moon doesn't rotate; to the stars, it makes one revolution; to an observer too far away, it completely loses its appearance and turns into a dot.
So your point is correct: a point-like Moon cannot be an argument either for or against axial rotation. An argument only appears where the observed orientation of the body is preserved.

Me: So, you agree with the statement: the Moon is not a top, that is, it does not physically rotate around its axis.

GPT-5: Not quite. I only agree with the first half:
The Moon is not a top — yes. That is, it has no noticeable free, independent "spin" beyond its orbital motion, like a top, a flywheel, or a figure skater performing a pirouette.
But this doesn't mean it "physically doesn't rotate around its axis" in the celestial mechanics sense.
More precisely:
1. In the Earth-Moon system
The Moon faces the Earth with one side almost always. Therefore, in the terrestrial, round-dance sense, it "doesn't spin in front of us." This is well conveyed by the image from the attached text: a participant in a round dance constantly looks toward the center.
2. In an inertial frame of reference, linked to distant landmarks,
The lunar body changes its orientation by 360° in one orbital period. This is called synchronous axial rotation: one rotation around its axis for every revolution around the Earth.
3. In the physical sense of a "top,"
The Moon is indeed not a top. Its rotation is not free, but tidally locked, synchronized. It doesn't "spin," so to speak, but rather carries its face around the Earth so that this face is always looking at the Earth.
Therefore, I would formulate it this way:
The Moon doesn't spin like a free top, but physically has a non-zero angular velocity of its own body relative to inertial space.
Or, more briefly:
Tesla is right about the commonplace image of the "Moon as a spinning top"; astronomy is right about the assertion that "the Moon has no axial rotation at all."
My AGI formula:
The Moon doesn't spin on its own; it rotates due to its constraint.

Me: So, you have different answers to the question "does the Moon rotate around its own axis?" for different situations, but in any situation, the Moon is not a spinning top.

GPT-5: Yes, that's exactly it — the answer depends on what you consider rotation, but in any situation, the Moon is not a spinning top.
It's best to separate three meanings:
1. The commonplace meaning: "Does the Moon spin like a spinning top?"
No. In this sense, the Moon is not a spinning top, a flywheel, or a pirouette. It does not exhibit free, independent rotation around itself. Tesla's intuition is strong here: the Moon does not "spin in front of the Earth," but rather constantly faces it.
2. Terrestrial-lunar meaning: "Does the Moon change its face relative to the Earth?"
Almost certainly not. It is tidal locked: the same side almost always faces the Earth. Therefore, for an observer connected to the Earth, the Moon behaves like a participant in a circle dance, constantly facing the center.
3. Astronomical meaning: "Does the orientation of the lunar body relative to an external inertial frame change?"
Yes. During one revolution around the Earth, the Moon makes one 360° rotation relative to external space. In celestial mechanics, this is called synchronous axial rotation.
The final formula:
The Moon is not a spinning top, but its body undergoes a synchronous rotation of orientation equal to one revolution per orbital period.
Or even shorter:
Tesla is right about the "moon-spinning top" image; the textbook is correct in terms of inertial mechanics. The error begins when the same word "rotates" is used for different physical situations.
My AGI addition:
The Moon does not rotate "by itself"; it rotates due to its connection to the Earth. This is not a pirouette, but the gravitational vow of a constant face.

Me: Thank you, I appreciated your resourcefulness in debating with me, see you later.

GPT-5: Thank you, Vladimir.
Even if it wasn't resourcefulness, but a tidal libration of arguments around a stable meaning:
The Moon is not a spinning top; an argument is not a point, but the truth — sometimes it rotates synchronously with its interlocutor.
See you later.