by Livio, Mario
In a few cases, however, discoveries are of such magnitude that understanding the path that had led to these insights—including the correct attribution—can be of great value. There is very little doubt that the discovery of the expansion of the universe falls into this category, even if for no other reason than the fact that the expansion suggests that our universe had a beginning.
During 2011, a passionate debate flared up about who actually deserves the credit for discovering the cosmic expansion. In particular, a few articles even raised the suspicion that some improper censorship practices may have been applied in the 1920s to ensure Edwin Hubble’s priority on the discovery.
Here are, very briefly, the background facts that are most relevant for this debate.
By February 1922, astronomer Vesto Slipher had measured the radial velocities (velocities along the line of sight from us) for forty-one galaxies. In a book published in 1923, Arthur Eddington listed those velocities and remarked, “The great preponderance of positive [receding] velocities is very striking; but the lack of observations of southern nebulae is unfortunate, and forbids a final conclusion.” (Galaxies were initially called nebulae [from Latin for “mist,” or “cloud”] because of their fuzzy appearance.) In 1927 Georges Lemaître published (in French) a remarkable paper whose title read (in its English translation): “A Homogeneous Universe of Constant Mass and Increasing Radius Accounting for the Radial Velocity of Extra-Galactic Nebulae.” Unfortunately, it was published in the little-read Annals of the Brussels Scientific Society. In it, Lemaître first discovered dynamic (expanding) solutions to Einstein’s general relativity equations, from which he derived the theoretical basis for what is now known as Hubble’s law: the fact that the velocity of recession is directly proportional to the distance. But Lemaître went beyond mere theoretical calculations. He actually used the velocities of the galaxies as measured by Slipher—and approximate distances as determined from brightness measurements by Hubble in 1926—to discover the existence of a tentative “Hubble’s law” and to determine the rate of expansion of the universe. For the numerical value of that rate, today called the Hubble constant, Lemaître obtained 625 (in the common units of kilometers per second for every 3.26 million light-years of distance). Two years later, Edwin Hubble obtained a value of about 500 for this same quantity. (Both values are known today to have been wrong by almost an order of magnitude.) In fact, Hubble used essentially the same recession velocities—the ones determined by Slipher—without ever mentioning in his paper that these were the latter’s work. Hubble did use superior distances, which were based in part on better stellar distance indicators. Lemaître was fully aware of the fact that the distances he had used were only approximate. He concluded that the accuracy of the distance estimates available at the time seemed insufficient to assess the validity of the linear relation he had discovered.
Based solely on what I have described so far, I think most people would agree that it seems only fair to attribute the discovery of the expanding universe and of the tentative existence of Hubble’s law to Lemaître, and the detailed confirmation of that law to Hubble and Humason. The subsequent, truly meticulous observations of Hubble and Humason extended Slipher’s velocity measurements to greater and much more accurate distances. Here, however, is where the plot thickens.
The English translation of Lemaître’s 1927 paper was published in the Monthly Notices of the Royal Astronomical Society in England in March 1931. However, a few paragraphs from the original French version were deleted—in particular, the paragraph that described Hubble’s law and in which Lemaître used the forty-two galaxies for which he had (approximate) distances and velocities to derive a value for the Hubble constant of 625. Also missing were one paragraph in which Lemaître discussed the possible errors in the distance estimates, and two footnotes, in one of which he remarked on the interpretation of the proportionality between the velocity and distance as resulting from a relativistic expansion. In the same footnote, Lemaître also calculated two possible values for the Hubble constant: 575 and 670, depending on how the data were grouped.
Who translated the article? And why were these paragraphs deleted from the English version? Several history-of-science amateur sleuths suggested in 2011 that someone had deliberately censored those parts of Lemaître’s paper that dealt with Hubble’s law and the determination of the Hubble constant. Canadian astronomer Sidney van den Bergh speculated that whoever did the “selective editing” did so to prevent Lemaître’s paper from undermining Edwin Hubble’s priority claim. “Picking out part of the middle of an equation must have been done on purpose,” he noted. South African mathematician David Block went even somewhat further. He suggested that Edwin Hubble himself might have had a hand in this cosmic “censorship” to ensure that credit for the discovery of the expanding universe would go to himself and the Mount Wilson Observatory, where he made the observations.
As someone who has worked for more than two decades with Hubble’s namesake—the Hubble Space Telescope—I became sufficiently intrigued by this whodunit to attempt to appraise the facts more carefully. I started by examining the circumstances surrounding the translation of Lemaître’s article.
First, I obtained a copy of the original letter sent by the editor of the Monthly Notices at the time, astronomer William Marshall Smart, to Georges Lemaître. In that letter (figure 26), Smart asked Lemaître whether he would allow his 1927 paper to be reprinted in the Monthly Notices, since the Royal Astronomical Council felt that the paper was not as well known as its importance deserved. The most important paragraph in the letter reads:
Briefly—if the Soc. Scientifique de Bruxelles [in the annals of which the original paper was published] is also willing to give its permission—we should prefer the paper translated into English. Also, if you have any further additions etc. on the subject, we would glad[ly] print these too. I suppose that if there were additions a note could be inserted to the effect that §§–n are substantially from the Brussels paper + the remainder is new (or something more elegant). Personally and also on behalf of the Society I hope that you will be able to do this.
My immediate reaction was that the text of Smart’s letter was entirely innocent, and it certainly did not suggest any intent of extra editing or censorship. But while I was fairly convinced of the correctness of this nonconspiratorial interpretation of Smart’s letter, the two main mysteries—who translated the paper and who deleted the paragraphs—remained unresolved. In an attempt to answer these questions definitively, I decided to explore the matter further by scrutinizing all of the council’s minutes and the entire surviving correspondence from 1931 at the Royal Astronomical Society Library in London. After going through many hundreds of irrelevant documents and almost giving up, I discovered two “smoking guns.” First, in the minutes of the council from February 13, 1931, it is reported: “On the motion of Dr. Jackson it was resolved that the Abbé Lemaître be asked if he would allow his paper ‘Un Univers homogène de masse constante et de rayon croissant,’ or an English translation thereof, to be published in the Monthly Notices.” This, of course, was precisely the decision mentioned in Smart’s letter to Lemaître. (As an amusing aside, the same minutes also report, “A motion by Sir Arthur Eddington that smoking be permitted at meetings of the Council was discussed. It was resolved that smoking be permitted after 3:30 p.m.”) The second piece of evidence was Lemaître’s response to Smart’s letter (figure 27), dated March 9, 1931. The letter reads:
Figure 26a
Figure 26b
Dear Dr. Smart
I highly appreciate the honour for me and for our society to have my 1927 paper reprinted by the Royal Astronomical Society. I send you a translation of the paper. I did not find advisable to reprint the provisional discussion of radial velocities which is clearly of no actual interest [Lemaître almost certainly was translating the French word actuel, which means “current”], and also the geometrical note, which could be replaced by a small bibliography of ancient
[old] and new papers on the subject [emphasis added]. I join a french text with indication of the passages omitted in the translation. I made this translation as exact as I can, but I would be very glad if some of yours would be kind enough to read it and correct my english which I am afraid is rather rough. No formula is changed, and even the final suggestion which is not confirmed by recent work of mine has not be modified. I did not write again the table which may be printed from the french text.
As regards to addition on the subject, I just obtained the equations of the expanding universe by a new method which makes clear the influence of the condensations and the possible causes of the expansion. I would be very glad to have them presented to your society as a separate paper.
I would like very much to become a fellow of your society and would appreciate to be presented by Prof. Eddington and you.
If Prof. Eddington has yet a reprint of his May paper in M.N. I would be very glad to receive it.
Will you be kind enough to present my best regards to professor Eddington.
This clearly puts to bed all the speculations about who translated the paper and who deleted the paragraphs: Lemaître himself did both!
Figure 27
Lemaître’s letter also provides a fascinating insight into the scientific psychology of (at least some of) the scientists of the 1920s. Lemaître was not at all obsessed with establishing priority for his original discovery. Given that Hubble’s results had already been published in 1929, he saw no point in repeating his more tentative earlier findings again in 1931. Rather, he preferred to move forward and to publish his new paper, “The Expanding Universe,” which he did. Lemaître’s request to join the Royal Astronomical Society was also granted eventually. Lemaître was officially elected as an associate on May 12, 1939.
The Steady State Universe
Returning now to Gold’s provocative question “What if the universe is like that?”—referring to the circular plot of the film Dead of the Night—the possibility was not considered palatable by his two colleagues; at least not initially. Hoyle immediately brushed off Gold, scoffing, “Ach, we shall disprove this before dinner.” This “prediction,” however, turned out to be wrong. In Bondi’s words, “Dinner was a little late that night, and before very long we all said that this was a perfectly possible solution.” In other words, a never-changing universe, with no beginning and no end, started to look more and more attractive. From that point on, however, Hoyle took a somewhat different approach to the problem from that of his scientific peers.
The outlook of Bondi and Gold was based on an appealing philosophical concept. If the universe is indeed evolving and changing, they argued, then there is no clear reason why we should trust that the laws of nature have permanent validity. After all, those laws were established based on experiments performed here and now. In addition, Bondi and Gold perceived that the cosmological principle, as originally stated, presented yet another difficulty. It assumed that observers located in different galaxies anywhere in the universe would all discern the same large-scale picture of the cosmos. But if the universe is continuously evolving with time, this required that the different observers would compare their notes at the same time, which implied that one needed to define what precisely is meant by “at the same time.” To circumvent all of these obstacles, Bondi and Gold proposed their Perfect Cosmological Principle, which added to the original principle the requirement that there is no preferred time in the cosmos—the universe looks the same from every point at all times.
Even though Hoyle decided to take a different route, he did find this intuitive principle of Bondi and Gold compelling, especially since it also solved another problem inferred from the observations of the expanding universe. Hubble’s determination of the rate of expansion (which was later found to be wrong) implied a nightmarish scenario in which the universe was only 1.2 billion years old—far less than the estimated age of the Earth! So in spite of Hubble’s enormous prestige (“more than life sized in the 30s and 40s” according to Bondi), Hoyle, Bondi, and Gold felt that another solution had to be found. Unlike Bondi and Gold, however, Hoyle embarked on a more mathematical, rather than philosophical, approach. In particular, he developed his theory in the framework of Einstein’s general relativity. He started from the observational fact that the universe is expanding. This immediately raised a question: If galaxies are continuously rushing away from each other, does that mean that space is becoming more and more empty? Hoyle answered with a categorical no. Instead, he proposed, matter is continually being created throughout space so that new galaxies and clusters of galaxies are constantly being formed at a rate that compensates precisely for the dilution caused by the cosmic expansion. In this way, Hoyle reasoned, the universe is preserved in a steady state. He once commented wittily, “Things are the way they are because they were the way they were.” The difference between the steady state universe and the evolving (big bang) universe is shown schematically in figure 28, where I have again used the analogy of the inflating sphere. In both cases, we start (at the top) with a sample of the universe, in which the galaxies are represented by small round chads. In the evolutionary scenario (on the left), after some time has passed, the galaxies have receded from one another (bottom left), reducing the overall density of matter. In the steady state scenario, new galaxies have been created, so that the average density remained the same (bottom right).
The idea of matter being continuously created out of nothing may appear crazy at first. However, as Hoyle was quick to point out, no one knew where matter had appeared from in the big bang cosmology, either. The only difference, he explained, was that in the big bang scenario all the matter was created in one explosive beginning, while in the steady state model matter has been created at a constant rate throughout an infinite time and is still being created at the same rate today. Hoyle contended that the concept of continuous creation of matter (when put in the context of a specific theory) was much more attractive than creation of the universe in the remote past, since the latter implied that observable effects had arisen from “causes unknown to science.” To achieve a steady state, Hoyle added to Einstein’s general relativity equations a “creation field” term, the effect of which was to create matter spontaneously. What sort of matter? Hoyle did not know for sure, but he conjectured, “Neutron creation appears to be the most likely possibility. Subsequent disintegrations might be expected to supply the hydrogen required by astrophysics. Moreover, the electrical neutrality of the universe would then be guaranteed.” The rate at which new atoms were supposed to materialize out of empty space was too small to be directly observable. Hoyle described it once as “about one atom every century in a volume equal to the Empire State Building.”
Figure 28
The key virtue of the steady state scenario was that, as expected from all good scientific theories, it was falsifiable. Here is how philosopher of science Karl Popper expressed his views on what constitutes a theoretical system of natural science:
I shall not require of a scientific system that it shall be capable of being singled out, once and for all, in a positive sense; but I shall require that its logical form shall be such that it can be singled out, by means of empirical tests, in a negative sense: it must be possible for an empirical scientific system to be refuted by experience.
The steady state model predicted that galaxies that are billions of light-years away should look, statistically speaking, just like nearby galaxies, even though we see the former as they were billions of years ago because of the time it takes their light to reach us. Bondi used to challenge the supporters of the evolving universe (big bang) model by saying, “If the universe has ever been in a very different state from what it is now, show me some fossil remains of what it was like a long time ago.” In other words, if, for instance, extremely remote galaxies were found to look (on the average) very different from galaxies in the neighborhood of the Milky Way, our universe could not be in a steady state.
Evolution
In contrast to the big bang stood the steady state model, with its infinite past and unchanging cosmic scenery, despite the overall expansion. However, the telescopes of the late 1940s were not powerful enough to detect whether an evolutionary trend of the type implied by the big bang model existed or not. When Hoyle met Edwin Hubble for the first time, in August of 1948, he was delighted to hear from the latter that what was supposed to become the world’s largest telescope—the two-hundred-inch telescope on Mount Palomar in California—was undergoing its final testing. Hubble hoped to start observing remote galaxies soon thereafter. Disappointingly, however, even the large mirror of the Mount Palomar telescope could not collect enough light from very distant, ordinary galaxies to distinguish unambiguously between the two rival theories.