Experimental confirmation of General Relativity

[davieddy]

newscientist.com
Gravity probe measures Earth's dent in spacetime.

[mfgoode]

Davieddy thats a good link.

from the same article I quote:

"The predictions of general relativity fall well within the probe's precision, though that might change when the team announces far more accurate results. "It's not a done deal that general relativity is safe," says Bencze, who points out that the team's job is to make the best measurements, rather than confirm general relativity. "If Einstein is right, good for him," says Bencze. "If not, too bad."

Mally

[jinydu]

Not much word on the actual results so far on the official website:

einstein.stanford.edu

[jasong]

Duh... I'm such a dummy I didn't even realize this thread was referring to the Einstein@home experiment.

[mfgoode]

Much of Einstein's crucial theories were more based on 'mind experiments' than from a laboratory.

The principle that light bends in a gravitational field was a pure thought experiment and he worked at it to give the equations later which are still not satisfactorily resolved today.

By these he was able to predict the odd motion of Mercury and the deviation of light by the eclipse of a star which was verified by astronomers a few years later.

Mally

[ET_]

Originally Posted by mfgoode

Much of Einstein's crucial theories were more based on 'mind experiments' than from a laboratory.

The principle that light bends in a gravitational field was a pure thought experiment and he worked at it to give the equations later which are still not satisfactorily resolved today.

By these he was able to predict the odd motion of Mercury and the deviation of light by the eclipse of a star which was verified by astronomers a few years later.

Mally

Wasn't Michaelson & Moorley experiment, in late 1800, the first one done to demonstrate the deviation of light? Although their instrument had not the required precision...

Luigi

[Uncwilly]

Originally Posted by mfgoode

By these he was able to predict the odd motion of Mercury and the deviation of light by the eclipse of a star which was verified by astronomers a few years later.

The motion of Mercury was known, it was one of the points that showed a weakness in Newton. Al's theories provided an explaination.
There are questions about the 1919 eclipse data. It was a solar eclipse. There are some that have stated that, the resolutions involved were not great enough to demonstrate the effect. The 'seeing' conditions of the solar atmosphere could have contributed to the observation.

[ewmayer]

Originally Posted by ET_

Wasn't Michaelson & Moorley experiment, in late 1800, the first one done to demonstrate the deviation of light? Although their instrument had not the required precision...

I think you may be confusing two things here:

(1) The Michelson-Morley experiment, which showed to extraordinary precision that the speed of light in vacuo was constant - thus disproving the "ether" hypothesis for transmission of light, and validating one of the key predictions of *special* relativity. (Note that the MME was actualley performed in 1887, when Einstein was a schoolboy; it was only after the publication of SR in 1905 that its import was fully appreciated, and together with the fact that the Nobels were first awarded in 1901, Michelson had to wait until 1907 to get his. I'd say this is also an example of the natural human bias against "negative" experimental results.)

(2) The 1919 eclipse expedition(s) whose goal was to show whether the bending of light by a massive object followed the predictions of *general* relativity, in that the deviation should differ precisely by factor of 2x from that of the Newtonian theory (which also predicts a deflection - many people seem to misunderstand this).

Originally Posted by Uncwilly

The motion of Mercury was known, it was one of the points that showed a weakness in Newton. Al's theories provided an explaination.
There are questions about the 1919 eclipse data. It was a solar eclipse. There are some that have stated that, the resolutions involved were not great enough to demonstrate the effect. The 'seeing' conditions of the solar atmosphere could have contributed to the observation.

It is possible that there may have been some optimistic "fudging" of the data (or more likely, the error bars around the data) going on. Ah yes, the Wikipedia article on experimental tests of GR says as much:

"The early accuracy, however, was poor and is described further in the article on predictive power. Dyson et al. quoted an optimistically low uncertainty in their measurement, which is thought to be plagued by systematic error and possibly confirmation bias. In 1801 J. Soldner had pointed out that Newtonian gravity predicts that starlight will bend around a massive object, but the predicted effect is only half the value predicted by general relativity as calculated by Einstein in his 1911 paper. The results of Soldner were revived by the Nobel laureate Philipp Lenard in an attempt to discredit Einstein.^{[url=][4][/url]} Eddington had been aware in 1919 of the alternative predictions but had rejected eclipse data consistent with the Newtonian predictions. Considerable uncertainty remained in these measurements for almost fifty years, until observations started being made at radio frequencies. It was not until the late 1960s that it was definitively shown that the amount of deflection was the full value predicted by general relativity, and not half that number."

But remember, at the time of the eclipse expeditions many top scientists (led by the influential Eddington) were already to some degree convinced by (what is to me at least) the even more astonishing prediction of the anomalous perihelion advance of Mercury, which falls out from GR via a direct exercise in regular perturbation expansions, and which agreed with the (highly accurate) astronomical observations and computations (to remove the effects of the motions of the known planets, which dominate the total perihelion shift - the previously-unexplained remainder is the aforementioned "anomalous" shift) to stunning accuracy - IIRC the agreement was to within less than 1%, and has only gotten better due to improved ways of measuring the actual perihelion shift. That was largely responsible for the "confirmation bias" noted in the above quote.

[ewmayer]

Funny side note on the Michelson-Morley experiment: it was performed at what was then known as the Western Reserve Academy in Cleveland, Ohio - in the 1960s the WRA merged with the Case School of Engineering across the street from it to become Case Western Reserve University, where (among other things) famous computer scientist Don Knuth did his undergraduate work -- there's a note about that in the dedication of his TAOCP books -- and much less famously, where I was on the faculty of engineering from 1993-1999. At one end of the Case Quad, the main public area of the engineering quadrangle, there is an homage to Michelson and Morley in the form of the Michelson-Morley Fountain. Quite the gleaming steel phallic symbol, isn't it? Well, one year while I was there, we had a particularly nasty snowstorm, and sometime during one night some students contrived to roll a pair of giant (roughly 6-feet diameter) snow "balls" right up to the base of the steel spire. Looking at the artistic description on the above fountain page, I wonder how a similar artsy-fartsy description of that little artistic addition might sound?

"At the base of the fountain, located near the site of their experiments, a giant pair of frozen snowballs serves to remind generations of students of the cool boldness of the experimenters as they performed this seminal work for Science."

And of course during the warmer months of the year, the lack of snowballs is more than made up by the continuous gush of water out of the top of the steel spire. Seminal work, indeed.

[mfgoode]

Originally Posted by ET_

Wasn't Michaelson & Moorley experiment, in late 1800, the first one done to demonstrate the deviation of light? Although their instrument had not the required precision...

Luigi

ewmayer has just beaten me to the post. Nevertheless... thought transference?



Not quite Luigi.

The MM experiment was designed to see if light was deviated when measured
along the earths motion and perpendicular to it over equal distances.

Classical resolution of the two speeds predicted a difference in the speeds or in times.

This was based on the principle that two boats moving along a moving river, and one across it, for equal distances would cover the distances in unequal times by a factor called Beta to get back to the same starting point

M&m used a very accurate instrument( the interferometer) which could detect the change as there would be a phase difference in the light beams and this could be ascertained. Surprisingly there was no difference!

Based on this experiment Al concluded that distances along the velocity would shrink by the same factor Beta. Lorentz had already theoretically calculated this in experiments on electromagnetism by Maxwell and Faraday.

And so we have Special Relativity in a nutshell ! Lengths contract in the direction of motion. the velocity of light is a natural constant. Adding a velocity or subtracting it makes no difference to its speed !

No it had nothing to do with the Earths gravitational field.

It was the ascending /descending lift under acceleration that gave Al the idea of light bending in a gravitational field.

It was Galileo who postulated that light has an infinite speed as his crude experiments showed no time difference.

In browsing the net I came across an interesting article that might interest your self and Uncwilly.

"Today, according to the US National Bureau of Standards, the speed of light is = 299792.4574 +/- 0.0011 km/s. According to the British National Physical Laboratory, the speed of light = 299792.4590 +/- 0.0008 km/s (making an average with the US standard = 299792.458 km/s).

But 1400 years ago, and without any scientific equipment at all, the speed of light was already known to be 299792.5 Km/s. It is stated in the Koran (Quran, the book of Islam) that light travels in one day the same distance that the moon travels in 12000 lunar orbits. And since velocity = distance / time, a simple calculation reveals the speed of light to be 299792.5 Km/s. This is mathematically verifiable in less than 10 minutes"

You are right Uncwilly I should have said 'explained' not 'predicted' the motion of Mercury

Mally

-

[mfgoode]

Originally Posted by ewmayer

Funny side note on the Michelson-Morley
experiment: Michelson-Morley Fountain.
~ ~
Quite the gleaming steel phallic symbol, isn't it? Well, one year while I was there, we had a particularly nasty snowstorm, and sometime during one night some students contrived to roll a pair of giant (roughly 6-feet diameter) snow "balls" right up to the base of the steel spire. Looking at the artistic description on the above fountain page, I wonder how a similar artsy-fartsy description of that little artistic addition might sound?

"At the base of the fountain, located near the site of their experiments, a giant pair of frozen snowballs serves to remind generations of students of the cool boldness of the experimenters as they performed this seminal work for Science."

Very Interesting and amusing Ernst

Mally