Did we ever actually see the Earth revolving around the Sun?

Also, is the geocentric model completely disproved or was it sidelined because the heliocentric model made things easier to understand?

(BTW I know Earth revolves around the Sun and am just asking out of curiosity.)


  • 2
    $\begingroup$ Please remember that comments should be civil and constructive and that extended conversations should be moved to chat. $\endgroup$ Commented May 30, 2014 at 12:34

4 Answers 4


What you're asking, basically, is whether there are any proofs for the heliocentric model of the Solar System.

A literal naked-eye observation of the Earth revolving around the Sun would be rather difficult, since human beings have never gone to another planet yet, and have only been to the Moon briefly, decades ago.

Here are several proofs; some of them are historically relevant also.

Kepler's laws of planetary motion

This became one of the earliest proofs, as soon as Newton figured out the law of universal gravitation, and the "fluxions" (what we would call today differential equations). When you assume a heliocentric model, and the inverse square law for gravity, then Kepler's laws in a heliocentric model come out of the equations naturally, as soon as you do the math.

This is like saying: "if it's heliocentric, and knowing that the law of gravity is correct, then Kepler's laws should be such-and-such". And then: "oh, but the theoretical calculations for Kepler's laws match the actual observations with great precision. Therefore, our hypothesis (heliocentric, inverse square law) must be correct."

enter image description here

It was the earliest strong indication that the heliocentric model is in natural accord with the basic laws of science, whereas the geocentric view was becoming more and more contrived as evidence accumulated.

Tycho Brahe in late 1500s provided the enormous mass of observations of planetary motion. Johannes Kepler in early 1600s used Brahe's observations to come up with his laws empyrically (and also arguing for the heliocentric model). Isaac Newton in late 1600s said "yes, Kepler is right, because of mathematics and the law of gravity, and here's the proof from calculus".


Stellar parallax

An early argument against heliocentrism was that, if the Earth was really revolving around the Sun, then very distant objects, such as the stars, would appear to be ever-so-slightly bobbing back and forth around their average positions. Since that's not the case, it was argued, therefore the Earth must be fixed.

You can see this argument in historical archives, proposed by theologians in the late 1600s, in favor of the geocentric model, and against the growing consensus then of "natural philosophers" (what we would call today scientists) that the heliocentric model was correct.

enter image description here

In reality, parallax does exist, it's just very small. It was measured experimentally in the 1800s, and was then quickly used to determine, for the first time, the distance to the nearest stars.


Aberration of starlight

The direction where we see a distant star also changes when Earth's speed vector changes during its revolution around the Sun. This is different from parallax; it's more akin to the way raindrops on the side windows of a car leave diagonal traces when the car starts moving (even though the raindrops fall vertically).

enter image description here

It's essentially a relativistic phenomenon (when applied to light), but it can be partially explained in a classic framework. It was actually observed before parallax in the late 1600s (the heyday of Newton), but it went unexplained until early 1700s.


Orbital mechanics of interplanetary probes

Landing a probe on Mars or Venus would simply not work if you assumed a geocentric model. A geocentric description of the Solar System might remain valid in a purely kinematic perspective (just the geometry of motion) as long as you remain on Earth. But the illusion breaks down quickly as soon as you start to consider dynamics (see Kepler's laws), and/or when you try to actually leave Earth (space probes).

Let me reinforce this point, since several answers and comments got it wrong: the geocentric and the heliocentric models are not completely interchangeable, or a matter of relativity. You could build an "explanatory" geocentric model, and it would be "correct", purely in a kinematic fashion (the geometry of motion), and only as seen from Earth. But the model breaks down as soon as you consider dynamics (forces and masses); it would also reveal itself as incorrect even from a kinematic perspective as soon as you leave Earth.

This is not just an artifice to simplify calculations. The dynamic calculations are wrong in a geocentric model.

In order to compute the very high precision trajectory of the space vehicle carrying the Curiosity rover and successfully place it on Mars, you must operate from a heliocentric perspective. The dynamics are all wrong otherwise. You would not miss the target only a little, in a geocentric approach, you would miss it by a lot - the vehicle would not even go in the general direction of Mars.


When observed in a telescope, Venus has phases like the Moon, and also grows and then shrinks in size, synchronized with its phases (it's large as a thin crescent, it's small when it's gibbous). In a geocentric model, the size changes could be explained by an elliptic orbit of Venus around Earth, but the phase changes synchronized with that are harder to explain. Both phenomena become trivial to explain in a heliocentric model.

It should be noted that this does not necessarily prove the heliocentric model, just the fact that Venus is orbiting the Sun, not the Earth. So it's an argument against pure (or strict) geocentrism.

Jupiter, when observed in a telescope, clearly has its own satellites. This was an early blow against a strict geocentric model, which assumed that everything must orbit the Earth. It opened the door to the idea that orbits could be centered on other celestial bodies too, and to the idea that things in orbit around larger objects could have their own smaller satellites (and therefore the Earth could orbit the Sun without losing the Moon).

The list could continue (and the full list is very long) but these arguments should suffice. You don't necessarily have to see something with your own eyes in order to know with certainty that it's there. In the case of Earth's revolution around the Sun, it was simply a matter of an overwhelming amount of evidence piling up in favor of it.

Geocentrism simply doesn't make any sense whatsoever in modern science and space exploration.

  • $\begingroup$ Can someone provide an example of a dynamics calculation that goes wrong in a geocentric model? $\endgroup$
    – avh4
    Commented Jul 6, 2014 at 19:15
  • $\begingroup$ Well, epi-cycles were introduced to correct for the retrograde motions of the plants in a geocentric model; they never really worked super well, and even if one could make them work mathematically, they represent really complicating and unphysical orbits... Not the best argument, but I thought it should be included here. $\endgroup$
    – astromax
    Commented Jul 24, 2014 at 2:21
  • 1
    $\begingroup$ Well, at a simple level, in a geocentric model you're assuming that the frame of the earth is an inertial frame, so using classical dynamics F = ma. But in that frame F doesn't equal ma, there are centrifugal and coriolis forces. The heliocentric model describes that frame as non-inertial, accounting for the extra terms in the dynamics. The "it's all just choice of frame of reference" claim has to face the issue that the only frame of reference in which those terms disappear from your dynamics equations is that of the Sun. So what does that mean, physically? :-) $\endgroup$ Commented Aug 11, 2016 at 11:19
  • 1
    $\begingroup$ Mind you, general relativity means you'll still miss Mercury using heliocentric Newtonian dynamics. And maybe other planets, I'm not sure. But that's a much smaller error than treating the earth as a Newtonian inertial frame, and therefore failing to go even in the general direction of the planet you want. $\endgroup$ Commented Aug 11, 2016 at 11:26
  • $\begingroup$ You can get the phases+size change of Venus from a geocentric model, but it requires stacking up epicycles into a complicated trajectory that is functionally equivalent to a Venus-orbiting-the-Sun model. $\endgroup$
    – Mark
    Commented Mar 19, 2020 at 3:10

Well, I'd argue with our satellites that are currently orbiting various planets in our solar system (Venus Express, Mars Reconnaisance Orbiter, Cassini, etc.).
Those have directional antennas to transmit the big chunks of data (omnidirectional antennas for low-gain communication like wake-up calls in case of a safety-mode event). So if those would not take into account the revolution of Earth around the Sun, we'd not have any scientific data.

I think there won't be any more 'direct' and 'visible' proof than this. Ofc if the person you want to convince about the truth of the Copernican system is more open to physical arguments (and doesn't run away at the sight of a bit math) you can start by laying out the gravitational potential for the restricted 3-body-problem see also, explain how literally none of our Space missions would have succeeded if the system were indeed Ptolemeian (lack of centrifugal forces around the sun -> everything shoots in a different direction).

I could go on here forever, all the dynamics which is explained nicely by Newtons law of Gravity (until a certain point..) would collapse.

Concerning your question whether the Ptolemeian model is completely disproved: It is as disproven as it gets. If your theory makes wrong predictions, it's disproven. You can add more and more virtual spheres of higher and higher order (modify the theory) in your imagined medieval sky, but at a certain point it becomes ridiculous. Even before the space age there was so much evidence pointing towards a Heliocentric viewpoint (Phases of Venus, Comets around the sun, discovery of Neptune and Uranus under heliocentric assumption...) Also here I could go on forever, but if you're interested in this, best read one of the books that are free on Project Gutenberg. They're legaly there as their copyright expired long ago and I found it very funny to read how already in the 1850's so much evidence for the Heliocentric theory was piled up.

  • 1
    $\begingroup$ The communication with the satellites doesn't prove (or disprove) anything. Using the heliocentric model makes the math easier to determine where to point the directional antenna, but the same result can be obtained using the geocentric model. $\endgroup$
    – LDC3
    Commented May 26, 2014 at 17:25
  • $\begingroup$ Communication cutoffs wouldn't happen according to the geocentric system. And the simple fact that we put them there using heliocentric calculations. $\endgroup$ Commented May 26, 2014 at 17:43
  • $\begingroup$ Sure they would. They go into the shadow of a planet, have a CPU error caused by a gamma ray requiring reboot, etc. And yes, we used the heliocentric model for the calculations, but like I said, the same results can be obtained using the geocentric model. It just makes the math complex. $\endgroup$
    – LDC3
    Commented May 26, 2014 at 17:48
  • 1
    $\begingroup$ First, there is no centrifugal force; it is imaginary. Second, the model is responsible for the forces acting on the satellites. And yes, both models can predict when a planetary shadow will occur. I didn't say you made the math more complex, I said the geocentric model made the math more complex than the heliocentric model. Please don't say I said something that I didn't say. $\endgroup$
    – LDC3
    Commented May 26, 2014 at 18:04
  • 2
    $\begingroup$ @LDC3 - the geocentric and the heliocentric models are equivalent only from a kinematic perspective, and even then only as long as you remain on Earth. But from a dynamic perspective, which you need when you're launching interplanetary probes, they are very different. You could never land Curiosity on Mars if you operated from a geocentric model. $\endgroup$ Commented May 28, 2014 at 19:07

According to Newton, any binary system revolves around the center of mass. For the Sun-Earth system, the center of mass is inside the Sun. I believe most people will then say that the Earth revolves around the Sun and not the other way around.

  • 1
    $\begingroup$ Actually the centre of mass is frequently not inside the Sun. $\endgroup$
    – ProfRob
    Commented Mar 5, 2018 at 0:15
  • 1
    $\begingroup$ @RobJeffries If the solar system was just the sun and the earth, then the center of mass is close to the center of the sun. "Thus if the average radius of the Earth's orbit is 93 million miles (150 million km) then the radius of the Sun's counter orbit is about 280 miles (450 km). The center of mass of the Sun-Earth system is well within the body of the Sun." applet-magic.com/centermass.htm If you include all the mass in the solar system, then you are correct. $\endgroup$
    – LDC3
    Commented Mar 6, 2018 at 4:19

The answer by Florin Andrei comprehensively explains scientific and Earth based observation support for the heliocentric model.

If we include interplanetary probes in the discussion, then yes, we have observed the Earth [orbiting] the Sun.

In order to receive commands from controllers on Earth the probes track (read calculate and observe) the location of Earth in order to point their radio dishes. Without accurate pointing, they are not able to receive the radio signal from Earth. Accurate pointing is also required to transmit results to Earth.

And some of those results have been photos of the Earth, in space right where the heliocentric model says it should be.

While not showing the Earth orbiting the Sun, there is a movie showing the Moon orbiting Earth here and here

  • $\begingroup$ In a similar vein, we also have a couple of probes crawling around the surface of Mars at the moment, and every indication so far is that there is no argument in favor of geocentric models that wouldn't work just as well to justify a Marti-ocentric model. $\endgroup$
    – David H
    Commented Feb 18, 2015 at 9:41

You must log in to answer this question.

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