Collection

zero Useful+1

zero

Light bending

Physical terms

synonym Gravitational deflection of light (gravitational deflection of light) generally refers to bending of light

The light is passing strongly gravitational field It will happen nearby bend , this is General relativity One of the important predictions of.

The verification of the ray bending prediction of general relativity is often dramatized, simplified and exaggerated to the audience and readers through the media directly facing the public and some scientific and cultural books. For example, in an artistic representation Einstein In his lifetime of French film Einstein, there was a scene where Einstein held a black photographic negative in Berlin, Germany, one day in the autumn of 1919 Plonk Said: (Main idea) What a real light bending, what a beautiful verification! The statement in some science books, such as "Eddington led a delegation to South Africa to see the solar eclipse, and really saw it", is too rough and misleading.

Whether the theoretical prediction has been confirmed by observation or not is directly related to whether the theory should be accepted as a correct theory. Therefore, the author believes that after the prediction of light bending made by general relativity, it is worthwhile to make a careful examination of the real history of the prediction. Moreover, on the basis of this examination, the author will General relativity In what sense and when can it become a correct theory. This discussion should also have some reference significance for how to treat the correctness of other theories in the history of science.

Chinese name: Light distortion

Cause: Spatial fluctuation
Category: Physical terms

catalog

Prediction and confirmation

Announce

edit

Brief clarification

There are three historical facts that are easy to confuse about the prediction and confirmation of light bending. Before narrating and verifying the true history of the prediction of light bending, briefly clarify them respectively.

First, light bending is not a prediction unique to general relativity. As early as 1704, he held the theory of light particles Newton It is proposed that a large mass object may bend the light like bending the tracks of other particles with mass. French astrobiologist a century later Laplace A similar view was put forward independently. 1804 Germany Munich Johann von Soldner (1766-1833) of the Observatory Newtonian mechanics , regarding light particles as particles with mass, it is predicted that 0.875 will occur when light passes the edge of the sun Angular second Deflection of. But in the eighteenth and nineteenth centuries Wave theory of light Gradually gaining the upper hand, the predictions of Newton, Sodner and others were not taken seriously.

In 1911 University of Prague Professor Einstein began to calculate the bending of the sun to light in his framework of general relativity, when he calculated that the starlight at the edge of the sun would deflect 0.87 angular seconds during the solar eclipse. Back in 1912 Zurich Einstein discovered that space is curved. By 1915, Einstein, who had worked at the Prussian Academy of Sciences in Berlin, had revised the deflection of starlight at the edge of the sun to 1.74 Angular second 。

Secondly, observation is needed to test not only whether the light is bent, but also how much the light is bent, so as to judge which theory is better consistent with the observation data. The key factor here is Observation accuracy 。 Even if the observation results negate the prediction of Newton's theory, it does not mean that it supports General relativity The prophecy of. only Observations The observation results can only be said to support general relativity if they are consistent with Einstein's prediction within the allowable error range. In the early 1960s, there was a new Gravitational theory Brans Dicke Theory also predicts that starlight will be deflected by the sun by 8% less than that predicted by general relativity. In order to distinguish between general relativity and Brance Dick theory, which is more consistent with the observation results, higher requirements are put forward for the observation accuracy.

Third, the effect of light bending cannot be seen directly by the eyes in the telescope or on the photographic film. The amount of light deflection needs to be obtained through a series of observations, measurements and reduction. The best opportunity to test the degree of bending of light passing through the vicinity of a massive object is to take photos of the sky near the sun during a total solar eclipse. Take a number of photographic negatives during the total solar eclipse, and then wait for a certain time (preferably half a year) before the sun is far away from the area where the solar eclipse occurs. The degree of deflection of starlight can be determined by measuring the front and rear sets of negatives.

Verification history of prophecy

It should also be pointed out that even in the Total solar eclipse It is also impossible to see stars close to the edge of the sun. Taking an observation in 1973 as an example, most of the stars photographed are concentrated at the distance of 5 to 9 solar radii from the sun center (see Figure 1), so the star deflection at the sun edge must be extrapolated according to the reduced curve. The one or two stars closest to the sun often strongly affect the final result.

An important figure in the history of the verification of the ray bending prediction of general relativity was the British physicist Arthur Eddington (1882-1944). In 1915, when Einstein gave the final result of light bending of stars on the edge of the sun the First World War All parties are engaged. Eddington, who was in a hostile country, learned about Einstein's theory through the Dutch, and was very interested in testing the prediction of general relativity about bending light. After World War I, Eddington persuaded the British government to subsidize the observation of light bending during the total solar eclipse on May 29, 1919. The British organized two observation expeditions for the eclipse, one to Sobra in northern Brazil; The other team went to the Gulf of Guinea in Africa Principe Principe. Eddington joined the latter team, but his luck was poor. The weather conditions in Principe were not very good when the total solar eclipse occurred. In November 1919, the results of the two observation teams were reduced to 1.98 ″ ± 0.12 ″; The result of the Principe team is 1.61 ″ ± 0.30 ″. On November 6, 1919, the British announced that the light was deflected according to Einstein's prediction.

But this announcement was hasty, because the final results calculated by the two observation teams were doubted by later researchers. Astronomers understand that there are many factors that lead to errors in the final results in such a complex observation as the detection of light bending. (The encyclopedia netizen added: The light is curved in the air with different densities. We know the principle in mirages and roads with strong light, so the light passing around the sun can also be a factor of error due to the bending of the air with different densities.) One of the factors that has a great impact is the change of temperature, which is caused by the change of temperature Atmospheric disturbance The model of is changed, the telescope focusing system is changed, and the size of the photographic film is affected by Thermal expansion and cold contraction And changes, which lead to systematic error Greatly increased. Eddington and others clearly recognized the impact of temperature change on instrument accuracy. They said in their report that the temperature difference less than 10 ° F can be ignored. However, the night temperature of Sobular is 75 ° F, the day temperature is 97 ° F, and the temperature difference between day and night is 22 ° F (the indoor temperature or the temperature difference after some treatment will be reduced). Later, the researchers took the influence of temperature change into consideration and recalculated the negative film of Sobular. The maximum deflection of light can reach 2.16 ″ ± 0.14 ″.

The imaging quality of the negative also affects the final result. In July 1919, 26 comparative negatives were taken in Sobular, 19 of which were made by Royal Greenwich Observatory Of Astrograph Shooting, there was a problem with the focusing system of this instrument specially used for astrophotography observation. The film quality was poor. Another 4-inch telescope took 7 films with good imaging quality. The light deflection calculated from the first 19 negatives is 0.93 ″ (Selected Astronomical Works, p.460), but the light deflection calculated from the last 7 negatives is far greater than Einstein's predicted value. The last published value is the average of all 26 negatives, only the first 19 negatives weighting The value is smaller. In 1929, German researchers checked the British observation results and found that if one of the stars was removed, such as a star with poor imaging, the final results would be greatly changed (Total Eclipse, pp. 200-201).

Later, in 1922, 1929, 1936, 1947 and 1952, when the eclipse occurred, astronomers from all over the world organized observations to examine the bending of light, and the published results were in good agreement with the predictions of general relativity, while others were seriously inconsistent. However, in the early 1960s, astronomers began to believe that the sun did deflect starlight, and believed that Einstein's predicted deflection was closer to observation than Newton's predicted deflection. However, the prediction of general relativity still deviates from the observation results, and Einstein's theory may need to be revised.

The total solar eclipse on June 30, 1973 was the second longest total solar eclipse in the 20th century, and the sun was in the background of the Milky Way sky with the most dense stars when the total solar eclipse occurred, which was very conducive to the inspection of light deflection. Americans Mauritania The Shingaiti Desert Oasis has built a special insulated hut for observation, and Observation accuracy Careful preparations have been made, such as keeping the darkroom and film solution at 20 ° C, monitoring the temperature changes of all parts of the whole instrument, etc. After taking photos of the solar eclipse, the observation team sealed the hut, sealed the stop pin on the telescope with cement, and returned to take a comparison film at the beginning of November. After analyzing all observations with a well-designed calculation program, it is found that the deflection of starlight at the edge of the sun is 1.66 ″ ± 0.18 ″ (Total Solar Eclipse, page 206). This result once again confirmed that the prediction of general relativity is more consistent with the observation than that of Newtonian mechanics, but it is difficult to exclude the previously proposed Brance Dicke theory.

Table 1 Optical observation results of light bending during multiple solar eclipses

date

place

Results and errors（ Angular second ）

May 29, 1919

Sobral

1.98±0.16

Principe

1.61±0.40

September 21, 1922

Australia

1.77±0.40

1.42－2.16

1.72±0.15

1.82±0.20

May 9, 1929

Sumatra

2.24±0.10

June 19, 1936

USSR

2.73±0.31

Japan

1.28－2.13

May 20, 1947

Brazil

2.01±0.27

February 25, 1952

Sudan

1.70±0.10

June 30, 1973

Mauritania

1.66±0.18

The accuracy of optical observation seems to have reached the limit, and people thought of testing the prediction of general relativity by observing the deflection of the sun to radio waves. Such observations have been carried out since about 1970. From 1974 to 1975, A. B. Fomalont and Sheila Mack (R. A. Sramek) observed the deflection of the sun to three radio sources using the very long baseline interferometry, and finally (1976) obtained that the microwave deflection of the radio source at the edge of the sun was 1.761 "± 0.016". Astronomers finally confirmed the prediction of general relativity with an error of less than 1%. By 1991, using the technology of cooperative observation of multiple observatories, they confirmed the prediction of general relativity on light bending with an accuracy of 1/10000. Only then what is observed is no longer visible light but invisible radio waves.

Table 2 Radio Observation Results of Solar Deflection of Radio Waves

year

place

Ratio of observed values to predicted values of general relativity

one thousand nine hundred and seventy

Owens Valley

1.01±0.11

one thousand nine hundred and seventy

Goldstone

1.04±0.15

one thousand nine hundred and seventy-one

(American) National RAO

0.90±0.05

one thousand nine hundred and seventy-one

Mullard RAO

1.07±0.17

one thousand nine hundred and seventy-three

Cambridge

1.04±0.08

one thousand nine hundred and seventy-four

Westerbork

0.96±0.05

one thousand nine hundred and seventy-four

Haystack/National

0.99±0.03

one thousand nine hundred and seventy-five

(American) National RAO

1.015±0.011

Westerbork

1.04±0.03

one thousand nine hundred and seventy-six

(American) National RAO

1.007±0.009

one thousand nine hundred and eighty-four

VBLI

1.004±0.002

one thousand nine hundred and ninety-one

VBLI

1.0001±0.0001

Cognitive process

Announce

edit

According to the aforementioned verification history of light bending, it seems that there is such a question: can we only say that Einstein's general theory of relativity became "correct" until 1973 or even 1991? In order to dispel this doubt, the author believes that it is necessary to talk about the correctness of general relativity at three levels,

In the eyes of the public

The first level is the correctness of general relativity in the eyes of the general public.

On November 6, 1919 Royal Astronomical Society At the conference held jointly with the Royal Society, astronomer Royle announced: "The starlight really follows Einstein Gravitational theory The prediction of "is deflected". The next day, Britain, always cautious《 The Times 》(Times) Prominent headline articles appeared: "Revolution in Science", two subtitles were "New Theory of the Universe", "The Bankruptcy of Newton's Concept" (Pais, p. 306-307). On December 14, 1919, the cover of the weekly Berliner Illustrate Zeitung published a picture of Einstein, accompanied by the title: "A new great man in world history: Albert Einstein His research marks a new revolution in our concept of nature, comparable to Copernicus Kepler , Newton's shoulder to shoulder. " （Pais, p.308）

from General relativity It is suggested that the so-called verification in 1919 was not qualified to a considerable extent according to the inspection of people's prediction of light bending in more than half a century. But there is no doubt that Einstein won great honor for the announcement of this verification. Under the publicity of the media, Einstein quickly became a legend and a hero admired by thousands of people. Einstein visited Britain for the first time in 1921 and stayed at Lord Haldane's residence in London. Haldane's daughter fainted when she saw the famous guest coming to her house.

Hero's behavior is always associated with right, justice and other attributes. In the era when few people in the world could understand general relativity, it was obvious that most readers of the Times, Berlin Pictorial and other media had accepted general relativity as a correct theory. In fact, today's media and most historians and philosophers of science also regard the solar eclipse observation in 1919 as an observation that confirmed Einstein's theory.

Einstein himself

The second level is the correctness of general relativity in Einstein's eyes.

How did Einstein view the predictions and observations of his theory? As early as 1914, before Einstein had calculated the correct light deflection value, he had already said in a letter to Besso with great confidence: "No matter whether the eclipse observation is successful or not, I have no doubt about the correctness of the whole theoretical system." (Pais, p.303)

Another story is also widely spread. It said that when the news that the prediction of light bending was confirmed by the British eclipse observation came, Einstein was in class, and a student asked him, what would he do if his prediction was proved wrong? Einstein replied, "Then I will feel sorry for my dear God. After all, my theory is correct." (Pais, p.30)

Einstein had his own views on the prediction and observation verification of general relativity. In 1930, Einstein wrote: "I think the main significance of general relativity is not to predict some weak observation effects, but its theoretical basis and structural simplicity." (Pais, p.273) In Einstein's view, it is the inherent simplicity of general relativity that guarantees its "correctness". The confirmation in 1919 did bring honor to Einstein, but that was outside of scientific theory; The confirmation in 1919 may have made more people "believe" that general relativity is "correct", but to a large extent, this kind of confirmation only played a role of "persuasion".

From the perspective of the history of science, the progress models of precise mathematical science do have such laws and characteristics: they are often refined by using the most profound mathematical knowledge available at that time theoretical model Their "correctness" is largely guaranteed by their internal simplicity and unity. Although they are bound to give testable predictions, such as Copernicus heliocentric theory Predicted the annual parallax of stars, Einstein General relativity Predicted the bending of light, Hawking Of black hole The theory predicts hawking radiation However, it is not necessary to wait until these predictions are confirmed. Those theories should and have been regarded as scientific theories.

Scientists and researchers

The third level is the correctness of general relativity in the eyes of scientists and relevant researchers.

As we all know, Einstein won the Nobel Prize in physics in 1921 because of his theory of light quanta. Royal Swedish Academy of Sciences When summarizing Einstein's main physics work in the awarding speech, Alenius, chairman of the Nobel Prize Committee for Physics, mentioned that "Einstein's third research is about the quantum theory founded by Planck in 1900 Nobel Prize 。” Arenius certainly mentioned Einstein's work on relativity in his speech, but he described relativity as "fundamentally related to epistemology", "the famous philosopher Bergson criticized this theory in Paris", and "the astrophysics community also has doubts about this theory, because the relevant conclusions are currently being strictly tested." Apparently in this The nobel prize in physics In the eyes of the chairman of the committee, the so-called verification by the British two years ago did not seem to have happened.

The so-called suspicion of astrophysics can be seen from the following examples. It was held in Washington in 1920 History of Astronomy Or an important conference in the history of cosmology, whose main purpose is to Sapril (Harlow Shapley) and Curtis (Heber Curtis) provided a place to debate their respective views on the structure of the universe. This meeting is called "Great Debate" in the history of science. C.G. Abbot, the organizer of the "Great Debate", refused to consider relativity as a possible conference topic. He said, "I pray to God that scientific progress will send relativity to Fourth dimensional space Somewhere outside, it will never come back from there to torment us. "

Although, this Abbott can be regarded as an extreme example against general relativity. But the historical fact of the history of science is that in the professional field, general relativity has gone through a more tortuous road than special relativity. For a long time after the general theory of relativity was proposed, physicists were not interested in it. just as Stephen Weinberg It was pointed out that all modern physics that studied matter at the most basic level at that time relied to a large extent on two pillars: special relativity and quantum mechanics. In other words, general relativity is different from special relativity, and it was not necessary for the main research topics at that time, such as material theory and radiation theory.

In addition to some scientists who are not interested in general relativity, others who are interested in it are conducting more rigorous and precise tests on general relativity. As far as the prediction of light bending is concerned, 12 optical observation tests were carried out from 1919 to 1973; In addition, from 1970 to 1991, 12 radio observation tests were carried out.

In Einstein's view, it seems that there is no need for these tests. As early as 1914, his theory was already guaranteed to be correct by its inherent simplicity; In the eyes of the general public, the 1919 test is enough to prove that general relativity is correct. So what is the significance of testing the bending of light in the decades after 1919?

The author believes that it is very important for the theory to be accepted by the scientific community to verify a theory through observation. For example, Einstein, who proposed the theory, believed that the correctness of the theory was intrinsically guaranteed. For more people, they have no ability to judge the correctness of the theory on the basis of a deep understanding of the theory, so they can only use the "prediction confirmation" model, which is also effective in other occasions, to judge the correctness of the theory. If that Abbott can live to 1991, as long as he uses the scientific thinking and scientific methods commonly used in the scientific community to treat problems, he will also admit General relativity It is correct within the accuracy range of 1/10000.