Does gravity really "attract"?
October 23, 2010 3:29 PM   Subscribe

Is my conception of gravity deeply flawed, or is this bad terminology?

My understanding was that gravity is simply a way to describe how "sharp" the curve of local spacetime is, and thus how fast you move (fall) towards the thing which is curving said spacetime. It's not something like a magnet.

I've dug around askmefi and google, and cannot find anything about this; every description talks about gravity "attracting" things. What's the deal, here?
posted by curious nu to Science & Nature (20 answers total) 8 users marked this as a favorite
It's both. Gravity follows the inverse square law, so (in Newtonian terms) you can absolutely think of gravity as mass attracting mass.

Under Einstein, mass distorts space-time: a weight placed on a stretched rubber sheet is the classic metaphor. So an object in motion across the sheet towards the mass will fall or curve towards the distortion.

Newtonian physics gets you most of the way: it's more than enough to land man on the moon. Physics AE (After Einstein) is used under extreme conditions of mass or accuracy: GPS satellites, small oddities in the orbit of Mercury, black holes, etc.

Both are "correct" to a certain degree of approximation; Einstein is just more so, under a greater range of conditions (i.e. it works for both objects with mass and those without, such as the photon). Therefore you can think of gravity in either form. I suspect "attraction" is just easier for most people to relate to.

Of course, this drives us towards the question of what mediates gravity: the elusive graviton, theories of quantum gravity and strings, etc - but that's another question entirely.
posted by Bora Horza Gobuchul at 3:56 PM on October 23, 2010 [3 favorites]

it's considered one of the four fundamental forces, but in relativity theory is described as you do.

The issue is "gravity" is just the name given to the fact that mass goes toward mass. Why this happens is a theory of how forces work, or why, and we haven't totally worked that out yet. The relativity theory of gravity is one description, but some particle theorists also talk about "gravitons", for instance. Which doesn't mean those two explanations couldn't be reconciled...

I think if you google "gravity" and "relativity" you would get the sort of description you're thinking of.
posted by mdn at 3:59 PM on October 23, 2010

Actually, you get Maxwell's equations with a few more dimensions and the same approach.

My understanding was that gravity is simply a way to describe how "sharp" the curve of local spacetime is

That, or an attractive force at a distance between two objects. That general relativity is a better way to describe the actions of gravity doesn't mean than in many cases it isn't convenient to think of it in the simpler way.
posted by a robot made out of meat at 4:05 PM on October 23, 2010

Also, your conception of gravity is with high probability deeply flawed. I just accept that this is true of most topics in my day-to-day interactions with research physicists.
posted by a robot made out of meat at 4:08 PM on October 23, 2010 [1 favorite]

our understanding of gravity has always been incomplete, in particular it could never be reconciled with quantum mechanics. as a result we are exploring hypotheses whereby gravity is not a fundamental force, but rather a side effect.
posted by paradroid at 4:08 PM on October 23, 2010

My husband the physicist says:

Newtonian physics is all about forces, hence we talk about things in terms of attraction. So mass "pulls".
In General Relativity there are no forces: everything is just trying to travel the shortest distance between two points. If space time is curved (by mass), the shortest distance between two points can be a curve.

So basically your understanding is correct. But to the casual observer with an apple falling on his/her head, something traveling the shortest distance over curved space time can look like it's a matter of attraction, and that works as an explanation most of the time.
posted by lollusc at 5:26 PM on October 23, 2010

(He then follows up that explanation by saying, "Um, well, actually that's a pretty flawed understanding of how things work, but it's very beguiling, and an explanation that works for most people, so I guess we'll leave it at that." Bah, physicists.

(Then he left me with the question of why shining a beam of light sideways works, and the light doesn't fall down towards the earth. And he won't tell me the answer.)
posted by lollusc at 5:29 PM on October 23, 2010 [1 favorite]

Lollusc, this is a derail, but...

Beams of light are deflected by gravity. That was one of the major predictions that came out of General Relativity. But the effect is really small, and it was only confirmed by observing eclipses of the Sun. The positions of stars very near the sun in the sky were off by exactly what the theory said they would be.

Now as with all the discussions above, there are two ways of looking at that. 1. According to GR, the light is being deflected because space is distorted.

2. A different way of looking at it is that photons have mass (because they have energy and mass and energy are the same thing) and they're being affected by Gravity as a force.
posted by Chocolate Pickle at 5:56 PM on October 23, 2010

The term for that is "gravitational lensing" and astronomers have since found other examples of it in the sky, especially when they look towards certain galactic superclusters.
posted by Chocolate Pickle at 6:10 PM on October 23, 2010

Best answer: OK. Both the "attractive force" and the "curved space time" ideas are both useful, because they are both models of gravity. The Newtonian idea of an attractive force is completely legitimate to use for things on human scales. The relativistic curved space time is more useful on astronomical scales.

What this really gets into is what it means to model a phenomenon. If you ask an engineer what gravity is, she'll say it's a force that attracts objects downwards. If you ask a theoretical particle physicist what gravity is, he'll say it's the manifestation of a spin-2 scalar field, whose precise form hasn't been worked out yet. An astronomer will cite general relativity. They're all right, and all wrong at the same time.

The lesson is, just use the model that works in your situation!
posted by auto-correct at 6:39 PM on October 23, 2010 [3 favorites]

Even more confusing is the fact that everything tells us that gravity should behave as a wave and transport energy as gravitational radiation. But because it's so weak we are having a heck of a time detecting it. There are a number of Earth-based experiments but I think most of them are still barely picking up anything out of the noise. A planned space-based detector (LISA) ought to really deliver solid proof though.
posted by Rhomboid at 7:23 PM on October 23, 2010

Best answer: A surprising number of these posts are claiming that both Newtonian physics and general relativity are true, or at least valid, because they each make accurate predictions in their respective domains, even though they contradict each other in their basic assumptions about how the world works. This is, more or less, instrumentalism, roughly the idea that scientific theories should be evaluated only in terms of their ability to make accurate predictions. The opposing view is realism, roughly the idea that science should aim to describe objective reality—so no two contradictory theories could be valid at the same time. I'm pretty sure that most scientists are realists.

A realist (like me) would say that, given that general relativity has completely superseded Newtonian physics' description of gravity, we use Newtonian concepts like "pull" for convenience, but ultimately the underlying physical reality is that gravity is a "fictitious force."
posted by abcde at 9:20 PM on October 23, 2010

OP, abcd's links are certainly worth looking at, but I disagree with his opinion.

To be a realist in regards to gravity is to say that gravity obeys GR. As in, GR is the rule that the Universe must follow. I consider this to be a poor way of thinking, since it ignores thousands of years of scientific theories being superseded by more complete theories.

I think the better way of thinking is to treat GR as a mathematical model of how matter interacts on large scales. It is also a good conceptual model. That doesn't make it true.

In my field, particle physics, this way of thinking is a given. No one assumes the current theories to be TRUTH, but rather a stepping stone towards better understanding of the universe.

I think Kuhn's description of paradigms in The Structure of Scientific Revolutions provides a great way of looking at what a scientific theory really is.
posted by auto-correct at 11:50 PM on October 23, 2010

The curved-spacetime aspect of gravity has additional effects besides attracting mass and bending light; it actually creates extra space. For example, since the sun curves spacetime substantially, if you were to sketch out a cube measuring 2 million km on a side, centered on the sun, that cube will contain more than the 2³ million³ km³ = 8 × 10¹⁸ cubic kilometers it would cover in flat Euclidean space.

Following the rubber sheet and weight analogy that BHG mentions, you can see that the surface area of the cone-like region of the rubber sheet right around where the weight is stretching it, like this, is greater than the area of a flat square with the same 2D measurements.

This is the aspect of general relativity that was confirmed by Einstein's prediction of the precession of the perihelion of Mercury's orbit. (Mercury's orbit is a sort of corkscrew shape rather than an ellipse because of the extra space that close to the sun.)
posted by XMLicious at 12:03 AM on October 24, 2010 [1 favorite]

Best answer: auto-correct: Yeah, I decided to skirt the whole issue of, "what about if/when GR itself turns out to be wrong?" My line of thinking was this: If either model is true, it has to be GR, because it predicts all of the results of Newtonian physics, plus a bunch of others that Newtonian physics gets wrong. Just because GR will likely be superseded in the future doesn't mean it's on equal footing with Newtonian physics, whose model of reality is already known to be wrong.

Also, even if GR is shown to be wrong too, it has to be a much more accurate approximation of reality that Newtonian physics is. I'm reminded of Asimov's famous essay, appropriately titled The Relativity of Wrong, in which he points out, "When people thought the earth was spherical, they were wrong. But if you think that thinking the earth is spherical is just as wrong as thinking the earth is flat, then your view is wronger than both of them put together." Acting as if the earth is flat may still make it way easier to calculate distances when making short trips. But that doesn't mean it's not completely wrong in terms of the physical reality, while the idea that the earth is a sphere is almost entirely right in the same terms.

Say curious nu had asked, "is the Earth round or flat?" The right answer is not, "well, you can think about it in a variety of ways—they're both useful models, but it really depends on your situation, and we don't have an ultimate answer." Instead, it's "round—pretty much. But in day-to-day life, people still talk about it as if it were flat, because it's so much more convenient."

Now, whether or not general relativity is as close to the truth as the spherical-earth theory was, we don't know (yet). But we do know now that the Newtonian view of gravity is like the flat earth theory—very convenient at times, but wrong, wrong, wrong.
posted by abcde at 2:22 AM on October 24, 2010 [1 favorite]

"Beams of light are deflected by gravity. That was one of the major predictions that came out of General Relativity. But the effect is really small, and it was only confirmed by observing eclipses of the Sun. The positions of stars very near the sun in the sky were off by exactly what the theory said they would be.

Now as with all the discussions above, there are two ways of looking at that. 1. According to GR, the light is being deflected because space is distorted.

2. A different way of looking at it is that photons have mass (because they have energy and mass and energy are the same thing) and they're being affected by Gravity as a force."

It's a bit more complicated than that. The Newtonian perspective is ambiguous about how light should be deflected. If you think of light as massless, it shouldn't be. But if you consider how much something is deflected as its mass tends to zero, you do get a deflection but it's the wrong size. Although the effect is really small, it's twice as big in GR as you'd expect from Newton's (non-zero) prediction.
I'd say your number 2 isn't really accurate - photons don't have mass, and don't get deflected in the same way massive objects do.

I'd also point out, separately from the above, that the rubber-sheet analogy tends not to work brilliantly as an explanatory tool, as people tend to think of things falling towards the source of the distortion. But on a real rubber sheet that only happens because gravity is pulling something downwards towards it. Real rubber sheets tend to be embedded in the gravitational potential of the Earth. The point is more about the distortion of an object's path by the curved geometry, not that things naturally tend to head towards the bottom of a slope.
posted by edd at 5:14 AM on October 24, 2010

IIRC the difference between the Newtonian and relativistic prediction for the displacement of a beam of light by a gravity is because the latter includes the effect of time dilation due to the gravitational field.

(Do you need GR, or is SR enough for the latter? It's too long since my physics degree...)
posted by pharm at 1:33 PM on October 24, 2010

Just to put a footnote on my other posts: I talked to my physics-major friend about all this, and he reminded me that any successful theory of quantum gravity will likely involve gravitons, restoring gravity to its status as an actual force. So that weakens my point just a bit—er, maybe we'll revert back to flat-earth theory after all...

Nonetheless, until there's a widely-accepted theory of quantum gravity that overturns relativity, it still feels more logical to me to use the terminology of the only well-tested theory of gravity we have.
posted by abcde at 9:56 AM on October 25, 2010

Edd, photons may well not have mass, but they do have momentum. A beam of light can push an object in a vacuum. Some have suggested using it for interstellar travel.
posted by Chocolate Pickle at 6:16 PM on October 25, 2010

Chocolate Pickle: I'd never suggest otherwise. My point is that while one might normally think of gravity as the attraction between two masses, that doesn't make it right to think of the photon as having a mass due to its energy and that that is why it is deflected by a gravitational field. I'd argue that instead you need a better understanding of gravity - like GR - rather than making an analogy between energy and mass that isn't exactly correct.
posted by edd at 5:40 AM on October 26, 2010

« Older Help me get my deposit back by suggesting the BEST...   |   Which liquids can I drink? Newer »
This thread is closed to new comments.