Physics Stack Exchange is a question and answer site for active researchers, academics and students of physics. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Einstein said that gravity can be looked at as curvature in space- time and not as a force that is acting between bodies. (Actually what Einstein said was that gravity was curvature in space-time and not a force, but the question what gravity really is, is a philosophic question, not a physical one)

share|cite|improve this question
The spacetime curvature is the modern explanation for the force. But the force is still there. The force, as defined by Newton, is what one may read out of the acceleration of massive bodies via $F=ma$. Because the apples still accelerate, there's still a force even though we know that the reason is a curved spacetime. – Luboš Motl Aug 10 '12 at 7:20
@Luboš Motl how about unit of force how can you show unit of force N is spacetime curvature – german Aug 10 '12 at 7:24
You can look like that (in terms of space distortion) at other fundamental forces too. What is unique to gravity is that it acts on all bodies so you can say that the geometry of the gravity force is actually the geometry of the REAL physical space. Further insights are in the field of solid state physics. – Asphir Dom Aug 10 '12 at 7:51
True story, my girlfriend was in a bus which went into the back of another car and she hit her arm on the seat in front. She wasn't best pleased when I told her that she couldn't have really hurt her arm as the force was only fictitious (intertial). Gravity may not be a quantum field (or it may be, I should say probably is, I got in trouble before for not accepting that the graviton is all but discovered) I'm not sure that makes it any less of a force especially using the classical definition as pointed out by @Lubos Motl. – Bowler Aug 10 '12 at 8:23
@german, Curvature corresponds to "tidal force" (tidal acceleration), not force. The unit of curvature isn't the Newton, it is acceleration per meter ($s^{-2}$). – Nick Nov 24 '12 at 22:36

Websters specifically defines force as the gravitational interaction (definition 4b). We all were taught in high school that gravity was a force.

Given the lack of consensus among the authorities, a more edifying, less controversial, and equally true statement might be:

In general relativity, gravity is a fictitious force.

In classical mechanics, fictitious forces are not considered "real" forces. However, nobody, not even relativists, goes around claiming "the Coriolis force is not a force".

The issue of gravity being a force or not has nothing to do with general relativity. If you believe that inertial forces are forces, then gravity is a force. If you believe that inertial forces are not forces, then gravity is not a force.

share|cite|improve this answer

In GR, there are always two points of view--- local and global. In the local point of view, you look in a neighborhood of a point, and make a free-falling frame, and then motion is entirely in straight lines at constant velocity so that you don't see gravity. In this way of looking at it, gravity is not a "force", meaning it doesn't make a generally covariant contribution to the local curvature of the particle space-time paths.

In the global point of view, you see an incoming particle from infinity deflected by a field, and you say a force has been acting if the particle is deflected. In this point of view, every deflection is a force by definition.

The global point of view is the way in which gravity is treated in quantum field theory or string theory. The local point of view is the insight due to Einstein, and it is no surprise he would emphasize it in his public remarks.

The answer is "it depends on your philosophical definition of force, whether you take a local view or a global view." I prefer the global view, since it is more quantum, so I say gravity is a force, but I don't disagree with people who take the other view, since it is also valuable.

share|cite|improve this answer

Well, if we're talking about what Einstein said, then the way Einstein defined gravitational field and gravitational force in GTR is that it is given by the connection, with its components by the Christoffel symbols: $$\Gamma^{\alpha}_{\mu\nu} = \frac{1}{2}g^{\alpha\beta}\left[g_{\mu\beta,\alpha}+g_{\nu\alpha,\beta}-g_{\mu\nu,\beta}\right]$$ where commas denote partial derivatives and the metric $g_{\mu\nu}$ plays the role of gravitational potential.

But this is quite different from Newtonian gravitational force.

In Newtonian mechanics, you have 'real' forces and 'inertial' (aka "fictitious") forces, the difference being that you can make inertial forces disappear by adopting an inertial frame. For example, Newton's laws in a uniformly rotating reference frames introduce centrifugal and Coriolis forces that are proportional to the mass of the object acted upon and can be removed changing to an inertial, and hence non-rotating, frame.

In other words, inertial forces are the "fault" of choosing a non-inertial frame of reference.

By the above definition, gravity is an inertial force. Similarly to the Newtonian case, it can be made to disappear by changing the reference frame--but there is also a big difference: in the Newtonian framework, inertial frames are global, and so inertial forces disappear everywhere. In GTR, that's no longer the case: there are only local inertial frames in general, and so you can only make it disappear locally.

Caution: modern treatments of general relativity do not adopt this definition. Many of them (e.g., Misner, Thorne, and Wheeler) intentionally do not identify either 'gravity' or 'gravitational field' with any particular mathematical object, not the connection, not the curvature, nor anything else. But then (for MTW) it is not technically correct to say that gravity is spacetime curvature either, but rather refers "in a vague, collective sort of way" to all of these geometrical constructs.

share|cite|improve this answer

In the framework of GR, gravity is indeed not a force as it's a consequence of Newton's first law instead of the second one.

Each point in space-time comes with its own velocity space attached, and you need the parallel transport (and thus a connection aka gravity field) to be able to even define what you mean when you say a body moves without acceleration.

In the more general setting of arbitrary second-order systems (ie if we forget about Newton's laws), the space of acceleration fields carries an affine structure. A connection is one way to choose a zero point and make it into a vector space so you can have the notion of addition of forces (or rather acceleration fields). From this point of view, gravity would indeed be a force like any other, but special insofar as it gets chosen as the one that is called zero.

share|cite|improve this answer
This is a question of local vs. global again. – Ron Maimon Aug 10 '12 at 19:20

If gravity were a force, then there wouldn't be gravitational time dilation.

So let's assume gravity is a force, that pulls everything downward. We have a tower with one observer at the bottom and top.

The observer at the top drops two balls waiting $t$ between the two drops. The bottom observer would measure the very same time interval $t$ between the two falls.

But in reality there is difference between the two times, the bottom observer measures a smaller amount of time due to dilation. This effect is confirmed by many experiments. To have time dilation we need an accelerating frame of reference.

The reason of the time dilation is that the plane of simultaneity of an observer sweeps past other observers in a different rate than the rate of his clock.

In the following chart you can see an accelerating observer's worldline highlighted with blue (accelerating with constant proper acceleration). The radial lines are its planes of simultaneity at 0.2s, 0.4s, ... on his clock. The other hyperbolas are worldlines of points that remain rest at the frame of this observer, they are also accelerating but with a different rate. The red dots are the events when the clocks of each points hit 1s.

Rindler chart

You can see when the blue observer's clock hit the 1s, at the same moment the clocks at the points on the right are passed 1 second long ago, while clocks at the left are lagging behind. No curvature needed to get dilation, just accelerate.

So to sum up, when you stand on the Earth, you are actually in an accelerating reference frame that accelerates upward, and gravitation is just a fictitious force, the same force you feel in a car or train, when it accelerates.

Then why Earth is not falling apart, if things are accelerating upward on it? Because the space-time is curved. It's curved so inertial observers falling towards the center of the Earth. But we who are "hovering" in this field are accelerating upwards in this curved coordinate system.

share|cite|improve this answer
I don't follow your logic here. If you believe in the equivalence principle, then you get gravitational time dilation. But I don't see how that connects logically to the question of whether gravity is a force. – Ben Crowell Jul 18 '13 at 14:34
@BenCrowell my logic is about the force-field vs. curvature thing. Both of them satisfy the equivalence principle. You cannot feel if a mysterious force move all particles in your body. Just as you cannot feel it when you are in free fall. If gravity is a force field and you stand on the ground, you are not accelerating, as the forces cancel each other. The same happens with the observer at top of the tower. No relative motion, clocks are in sync. But in reality clocks are not in sync. So you must be in an accelerating frame and gravity can only be a fictitious force. – Calmarius Jul 18 '13 at 15:08

Einstien is right about one thing, gravity is not a force as defined by F=ma, but gravity is a force if you define force as resulting from energy.

Energy is hidden in the equation F=ma twice. Once in the Force and once in the acceleration. That is how energy is expressed in this equation. If motion is involved, energy is involved.

So is Einstein right about space-time curvature causing gravity? I don't know, but if it is space-time curvature, then the space-time curvature must be able to create energy.

"Force" is the result of energy acting on mass. "Mass" is defined by the weight of mass in gravity. Gravity is energy or a source of energy.

F=ma has an energy input which is "a" and an energy output "F"

If energy comes out of the equation, energy must go in, energy must be on both sides.

Mass is the medium used to calculate the energy in terms of acceleration, and it is the acceleration of gravity that is used for calculating "mass."

So energy in from gravity is expressed as constant acceleration. The product of energy and mass combined is what gives mass weight. The energy stored as weight can be transferred to another form of energy by necessary means. But gravity appears to be able to put energy into mass.

So if Einstein hasn't addressed the energy of gravity, he will have had a hard time understanding it. What ever the source of gravity, gravity is acceleration and not force. Force is mass by acceleration where as gravity is just acceleration.

The thing about that is all mass accelerates at the same rate, which generates different force on all things at all times with enormous variations in force resulting.

How can gravity be constant and yet apply unlimited numbers of force at any given moment? Gravity isn't a force, it is acceleration that generates force.

Same behavior is observed in electro-magnetic fields and explains many behaviors of gravity. If gravity's field is different, it is still related as it also explains gyroscopic effects. As you spin a metal mass, the centrifugal force creates a difference in charge from the outside and the inside of the spinning metal. By becoming charged, the metal aligns with the "field of gravity." Could be something different, but mass in gravity behaves a lot like mass does in magnetic fields.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.