Need for GR: Historical Indications

[greypilgrim]

Hi.

There were theoretical (find transformations under which Maxwell's equations remain invariant) and experimental (speed of light is constant, Michelson-Morley) indications that made the development of SR inevitable.

But what about GR? Was there a "need" for this theory or was Einstein just curious what would happen if he applied the equivalence principle exhaustively?

From the theoretical side, I assume Einstein was not happy with the fact that Newton's law of gravitation is nonlocal. Experimentally, I can only think of the perihelion precession of Mercury that was known before the theory and couldn't be explained otherwise. Einstein's other two classical tests, light deflection and gravitational redshift, hadn't been observed before.

But those two indications don't seem to be enough for me to convince someone that a new theory as radical as GR is needed. So was there other evidence?

 

[PeterDonis]

From the theoretical side, I assume Einstein was not happy with the fact that Newton's law of gravitation is nonlocal.

It was more than just "not happy". As soon as SR was established, it was obvious that Newtonian gravity as it stands is inconsistent with it. So the natural question was, how do we make a theory of gravity that is consistent with SR. That was Einstein's primary impetus towards GR.

those two indications don't seem to be enough for me to convince someone that a new theory as radical as GR is needed.

Initially it didn't convince Einstein either; he tried a simpler way of constructing a theory of gravity as an inverse square force with a light-speed time delay, similar to the way electrodynamics can be formulated in terms of retarded potentials. Unfortunately this didn't work: the simple theory he constructed this way grossly contradicted observation--for example, it didn't predict stable orbits for planets.

It was only after these abortive attempts that Einstein had what he called "the happiest thought of my life"--as he put it, "if a person falls freely, they will not feel their own weight". This led to the equivalence principle, which turned out to be the key idea that started him on the road to what we now know as GR.

 

[the_emi_guy]

Specifically, the things that fell out of SR that required Newtonian gravity to be revisited were:

1 - Gravitational fields are caused by mass, but SR revealed the matter-energy equivalence. Thus energy, in all of its forms, should be sources of gravity.

2 - Gravitation fields contain gravitational potential energy, but SR revealed that energy cannot travel faster than c. This conflicts with Newtonian action-at-a-distance.

3 - Gravitation fields contain gravitation potential energy, thus the presence of gravitation field is a source of field itself. This is responsible for the non-linear (& difficult to solve) nature of the Einstein field equations. This is why we cannot simply emulate electrostatics (inverse square law & fields traveling a c). To find the net field of a collection of charges, we can consider them one at a time and apply linear superposition. We can do this because charges create fields, but the fields themselves are not sources of field.

 

[MeJennifer]

But what about GR? Was there a "need" for this theory or was Einstein just curious what would happen if he applied the equivalence principle exhaustively?

Historically speaking there was no need.

Only after the experiment in Principe did it prove a solution to a scientific problem.

 

[the_emi_guy]

Historically speaking there was no need.

Only after the experiment in Principe did it prove a solution to a scientific problem.

Perhaps you have a more conservative interpretation of "need".

As has been pointed out, as soon as SR was established, it was clear that Newtonian gravity was inconsistent with it.

In physics, inconsistencies between theories need resolution.

 

[MeJennifer]

As has been pointed out, as soon as SR was established...

Perhaps you have a more conservative interpretation of "established".

And by the way PPN is inconsistent only in the most extreme gravitational experiments.

 

[Buzz Bloom]

Only after the experiment in Principe did it prove a solution to a scientific problem.

Hi @MeJennifer:

Can you provide some details about the "experiment in Principe"? Citing a specific reference would be very helpful.

Regards,
Buzz

 

[MeJennifer]

Hi @MeJennifer:

Can you provide some details about the "experiment in Principe"? Citing a specific reference would be very helpful.

Sure here are two interesting links:

http://cds.cern.ch/record/489163/files/0102462.pdf
http://astrogeo.oxfordjournals.org/content/50/4/4.12.full

 

[The Bill]

Historically speaking there was no need.

Only after the experiment in Principe did it prove a solution to a scientific problem.

The OP mentioned the anomalous(according to Newtonian gravity) precession of the perihelion of Mercury, which had been recognized as a problem since the 1850s.

Personally, I feel that Einstein's stated motivations of extending the principle of relativity to include something like Mach's principle, and of finding relativistic field equations for gravity is enough of a need.

 

[MeJennifer]

... to include something like Mach's principle, ...

In that respect he failed miserably.
Take for instance Gödel's rotating universe.

 

[The Bill]

In that respect he failed miserably.
Take for instance Gödel's rotating universe.

Whether he failed or succeeded in this is irrelevant to the fact that Mach's principle helped motivate him to develop the Einstein field equations.

 

[Buzz Bloom]

Sure here are two interesting links:

Hi $MeJennifer:

Thank you much for the two links. I of course knew about the astronomical observations in 1919 to test whether GR was correct about the gravitational bending of light rays. I had forgotten where these observations took place, and that Principe was the name of the location.
The following quote is from Wikipedia:

This eclipse was photographed from the expedition of Sir Arthur Eddington to the island of Principe (off the west coast of Africa).​

I looked at the map of the island on Google maps, and I find it a bit confusing. Apparently the name of the island country is São Tomé and Príncipe, which consists of two main islands and some very small others. The large island is São Tomé, and the other much smaller one is Ilhéu das Rolas. I could find nothing on the maps specifically named Principe. Is there a place on the island specifically named Principe, and is that where the observations took place?

Regards,
Buzz

 

[jtbell]

Google Maps (for me, at least) and Wikipedia both label the second-largest island as "Ilha do Principe" = "Island of Principe".