Terra revolving round Sol question

I'll get out of my depth before long here, but no: rotating the coordinate system should show up primarily in the metric.

Would not the metric also be imbedded in the [latex]T_\mu_\nu[/latex] term? The left side of the equation seems to deal with space curvature while the right side seems to define the forces on any x. Perhaps this is an oversimplification of what is going on.
 
No they wouldn't.
Yes they would.

When they look carefully at the planets, they see that the planets influence each other at a distance through a force that decreases with the square of their respective distances and their masses addition to[/I] a force that increases proportionally to the distance from the universal axis.
When they look carefully at the planets our Newtonian mechanics would notice they were spinning, and would work out that the Earth was spinning too. They'd entertain the rotating universe for about five minutes flat. If they decided to humour the idea, they'd draw a comparison with a flat galactic rotation curve, and some of them would propose dark matter to explain it. Others would say why invent a mystery substance? and point to what Newton said about density varying by degrees, and how it matched what the upstart Einstein said about inhomogeneous space. Then they'd look at the plot of gravitational potential and say that on a large scale the force of gravity increases like it increases from the centre of the Earth. Then somebody would slap them and remind them that the Earth was spinning.

So that's why gravity can't be the mystery force.
There is no mystery force, the Earth is spinning that's all, like just about everything else.

No, not quite. We're analyzing our rotating universe in a rotating frame of reference. That's not an inertial frame of reference.
Fine. But see the http://en.wikipedia.org/wiki/Coriolis_effect]Coriolis effect[/url] and look at the picture. Somebody referred to it somewhere, maybe you. We're slap in the middle of a fictitious rotation which ought to cause things to move through the universe in curved lines, only they don't, presumably because of some mystery force. It's all getting out of hand.
 
Or does the mass fall up to you?
If you were the same mass as "the mass", you'd fall towards each other. In practice you are so small compared to the Earth that's the Earth's motion towards you is not detectable.

When two things concurrently fall down from opposite sides of the mass towards its centre is the mass actually falling up to both of them? Does that mean the mass expands?
If you had three big masses like pearls on a string the outer two fall inwards and the middle one gets stretched.

When two things move toward a third thing at different speeds how does that third thing alternatively move toward the two things concurrently at different speeds?
It doesn't. You take some reference point between all three masses, then when you let them fall towards each other you express their speeds with reference to your point.
 
Would not the metric also be imbedded in the [latex]T_\mu_\nu[/latex] term? The left side of the equation seems to deal with space curvature while the right side seems to define the forces on any x. Perhaps this is an oversimplification of what is going on.


All three of the tensors that appear in Einstein's field equations (Ricci, metric, and stress-energy) are covariant tensors of rank 2, so the components of all three tensors transform covariantly. Hence

[latex]
\[
T^\prime_{\alpha\beta} =
\frac{\partial x^{\mu}}{\partial x^{\prime \alpha}}
\frac{\partial x^{\nu}}{\partial x^{\prime \beta}}
T_{\mu\nu}
\]
[/latex]​


For the Ricci tensor, change every T in that equation to R. For the metric tensor, change every T to g.

That equation applies to every pair of coordinate systems that's allowed by the definition of a differentiable manifold.

That covariant transformation law is equation (11) in Einstein's Die Grundlage der allgemeinen Relativitätstheorie. By denying the applicability of that cooordinate transformation, Farsight is arguing with Einstein.
 
Would not the metric also be imbedded in the [latex]T_\mu_\nu[/latex] term? The left side of the equation seems to deal with space curvature while the right side seems to define the forces on any x. Perhaps this is an oversimplification of what is going on.

Clinger's response is correct, but you specifically asked where the centripetal and Coriolis forces come in. My intuition (which is not super-confident) is tat they have to come in from the left-hand side of the equation.

Consider a flat, empty Universe, with lambda=0, no mass, no gravity waves. Describe it in ordinary spherical coordinates, it should be that T=0 everywhere. Now change to a rotating coordinate system; Clinger's transformation says that T=0 in the new coordinates. But there are nonzero pseudoforces in these coordinates; they must have come in from something on the right hand side.
 
Yes they would.

When they look carefully at the planets our Newtonian mechanics would notice they were spinning, and would work out that the Earth was spinning too. They'd entertain the rotating universe for about five minutes flat.

[...]

...Then somebody would slap them and remind them that the Earth was spinning.

[...]

There is no mystery force, the Earth is spinning that's all, like just about everything else.
:confused:
Er, do you even remember what we are discussing, or why?

Let me remind you:
NASA said:
...imagine that the earth were standing still and that the rest of the universe were rotating around it: would its equator still bulge? Newton would have said "No".
So in this thought experiment the earth is not rotating, but universe is rotating around it. That is given.
All I wanted to point out is that Newton would not have said "No", in the case that this rotating universe is anything like the one we have.
 
Clinger's response is correct, but you specifically asked where the centripetal and Coriolis forces come in. My intuition (which is not super-confident) is tat they have to come in from the left-hand side of the equation.

Consider a flat, empty Universe, with lambda=0, no mass, no gravity waves. Describe it in ordinary spherical coordinates, it should be that T=0 everywhere. Now change to a rotating coordinate system; Clinger's transformation says that T=0 in the new coordinates. But there are nonzero pseudoforces in these coordinates; they must have come in from something on the right hand side.

In the case you describe (flat spacetime), both the left and right-hand side of the equation are zero in all coordinate systems, as are both R_{\mu \nu} and R separately. The only thing that isn't zero is the metric.

To see where the "fictitious forces" come in, you need a different equation - the geodesic equation, the equation that describes the free motion of point particles. That contains a Christoffel connection, which is a non-tensorial object with three indices. It's zero in flat space in Minkowski coordinates, but it's not zero in rotating or spherical coordinates (which is possible because it isn't a tensor).
 
Last edited:
Clinger's response is correct, but you specifically asked where the centripetal and Coriolis forces come in. My intuition (which is not super-confident) is tat they have to come in from the left-hand side of the equation.

Consider a flat, empty Universe, with lambda=0, no mass, no gravity waves. Describe it in ordinary spherical coordinates, it should be that T=0 everywhere. Now change to a rotating coordinate system; Clinger's transformation says that T=0 in the new coordinates. But there are nonzero pseudoforces in these coordinates; they must have come in from something on the right hand side.
T could have non-zero components even if T = 0. It was my naive thinking that these non-zero components of the T tensor would represent these so called ficticious forces. I have been reviewing the Stanford University (by Leonard Susskind, consisting of twelve 1.5 hour segments) lectures on GR. I can tell that we will get to this question soon enough. I kind of jumped the gun with my question here.

By the way, this Internet access is a real joy for someone like me. Imagine lectures by someone like Susskind at no cost in my home. When there is a point I don't get (like recently, some questions about covariant differentiation), I can pause, do a little digging and go back to the lecture. Fantastic!
 
Clinger's response is correct, but you specifically asked where the centripetal and Coriolis forces come in. My intuition (which is not super-confident) is tat they have to come in from the left-hand side of the equation.
Einstein's field equation can be read in two different ways, both legitimate.

If you regard the equation as a coordinate-independent relationship between four tensors (counting the Ricci scalar as a tensor), then its meaning is independent of coordinate systems, and it has nothing to do with the fictional forces.

If you regard the equation as a relationship between the coordinate-dependent numerical components of the three non-scalar tensors, then the values of those components incorporate the fictional forces.

Consider a flat, empty Universe, with lambda=0, no mass, no gravity waves. Describe it in ordinary spherical coordinates, it should be that T=0 everywhere. Now change to a rotating coordinate system; Clinger's transformation says that T=0 in the new coordinates. But there are nonzero pseudoforces in these coordinates; they must have come in from something on the right hand side.
If spacetime is flat and the cosmological constant is zero, then the left hand side of Einstein's field equation is zero, so the right hand side must be zero also. The coordinate-dependent pseudo-forces show up only in the Christoffel symbols and in the coordinate-dependent components of the metric. For flat spacetime, the coordinate-dependent components of the metric don't affect the equation because they're multiplied by the Ricci scalar, which is zero for flat spacetime.

Note that flat spacetime and T=0 is a special case. If T is nonzero, then the numerical components of all non-scalar tensors will may depend on your coordinate system.

ETA: sol invictus gave a simpler explanation, three minutes before I posted.
 
Last edited:
No they wouldn't. When they look carefully at the planets, they see that the planets influence each other at a distance through a force that decreases with the square of their respective distances and their masses in addition to a force that increases proportionally to the distance from the universal axis.
You are trying to talk to someone completely ignorant of the actual way that Newton established his theory of universal gravity. Farsight does not understand the importance of the inverse square relationship between the force of gravity and the distance because it plays a role in the theory and the evidence. He only knows a few buzz words.
 
You may well be correct.

I think I should just join Perpetual Student in studying up on modern physics. I've always thought it's fascinating, but I never got around to learning it.
 
If you were the same mass as "the mass", you'd fall towards each other. In practice you are so small compared to the Earth that's the Earth's motion towards you is not detectable.

If you had three big masses like pearls on a string the outer two fall inwards and the middle one gets stretched.

It doesn't. You take some reference point between all three masses, then when you let them fall towards each other you express their speeds with reference to your point.
I was referring to “All motion is relative” not “Every action has an equal and opposite.”
 
T could have non-zero components even if T = 0.

Hmm. An equation like {some tensor with indices}=0 means that every component individually is zero. Since tensors transform by getting multiplied by something, if that equation is true in one coordinate system it's true in all.

So unless by "T" you mean the trace of the stress-energy tensor - which can indeed be zero even if the components aren't - I can't make much sense of your comment.
 
Hmm. An equation like {some tensor with indices}=0 means that every component individually is zero. Since tensors transform by getting multiplied by something, if that equation is true in one coordinate system it's true in all.

So unless by "T" you mean the trace of the stress-energy tensor - which can indeed be zero even if the components aren't - I can't make much sense of your comment.

This all too new for me to be making these kinds of leaps, sorry. I have been looking at tensors as a kind of function of vectors and scalars and like any function, assumed the function could take on a zero value with the argument not zero. For example, the dot product of two vectors can be zero with the components non-zero -- isn't the dot product a kind of simple tensor?
 
This all too new for me to be making these kinds of leaps, sorry. I have been looking at tensors as a kind of function of vectors and scalars and like any function, assumed the function could take on a zero value with the argument not zero. For example, the dot product of two vectors can be zero with the components non-zero -- isn't the dot product a kind of simple tensor?


Yes, but it isn't zero. In three-dimensional Cartesian coordinates, the components Aμν of the dot product tensor are
  • Aμμ = 1
  • Aμν = 0 for μ ≠ ν
So you need to distinguish between the tensor being zero (which means the result of applying it to the appropriate number of vectors or differential forms will always be zero) from the situation in which zero can result from applying the tensor to some particular vectors/differential forms.
 
Last edited:
Yes, but it isn't zero. In three-dimensional Cartesian coordinates, the components Aμν of the dot product tensor are
  • Aμμ = 1
  • Aμν = 0 for μ ≠ ν
So you need to distinguish between the tensor being zero (which means the result of applying it to the appropriate number of vectors or differential forms will always be zero) from the situation in which zero can result from applying the tensor to some particular vectors/differential forms.

If two vectors are orthogonal in Cartesian space, their dot product is zero even if both vectors may not have non-zero components. Would that not make the tensor zero with the components not zero?
 
This all too new for me to be making these kinds of leaps, sorry. I have been looking at tensors as a kind of function of vectors and scalars and like any function, assumed the function could take on a zero value with the argument not zero. For example, the dot product of two vectors can be zero with the components non-zero -- isn't the dot product a kind of simple tensor?

If two vectors are orthogonal in Cartesian space, their dot product is zero even if both vectors may not have non-zero components. Would that not make the tensor zero with the components not zero?


That does not make the tensor zero.

Let A be a covariant tensor of rank 2, and let u and v range over (contravariant) vectors.

((∃ u) (∃ v) A (u, v) = 0) does not imply A = 0.

((∀ u) (∀ v) A (u, v) = 0) does imply A = 0.

Do you see the difference?
 
That does not make the tensor zero.

Let A be a covariant tensor of rank 2, and let u and v range over (contravariant) vectors.

((∃ u) (∃ v) A (u, v) = 0) does not imply A = 0.

((∀ u) (∀ v) A (u, v) = 0) does imply A = 0.

Do you see the difference?

:blush: Got it. Thanks.

Addendum I
On second thought not quite:
((∀ u) (∀ v) A (u, v) = 0) does imply A = 0,
but it does not imply both u and v have all zero components or that u and v are both zero.

Addendum II
No that's wrong. I am confusing the vectors with the tensor A, which must have zero components for ((∀ u) (∀ v) A (u, v) = 0) does imply A = 0.
Finally got it.
 
Last edited:
Does the distant shell of rotating mass theory hold up when there's a stack of tops spinning in different directions at different speeds? By "stack" I mean tops on on top of each other so they all share the same axis. I was just in a toy store and saw a product called "Totem Tops" that does this. If two "shells" share the same axis and rotate in different directions why don't their opposing forces cancel each other out?
 

ISF - Join now!

Every member here is approved by hand. No bots, no spam, just people who care about evidence and honest debate.

Membership is free!

Create your free account

Back
Top Bottom