The Day We Found the Universe - Marcia Bartusiak (2009)

Exploration

Chapter 10. Go at Each Other “Hammer and Tongs”

The year 1920 was one of achievements—illustrious, infamous, resourceful, and humorous. American women got the vote, Joan of Arc was canonized by Pope Benedict XV, prohibition was initiated throughout the United States, an employee at the Johnson & Johnson company invented the Band-Aid, and the U.S. Post Office ruled that children may not be sent by parcel post. Moreover, astronomers didn't yet know exactly how the Sun generated its tremendous power or that the solar orb was largely composed of hydrogen, even though Einstein's newly introduced theory relating mass to energy, nicely summarized as E = mc2, was offering a fresh clue.

What 1920 is best remembered for in the annals of astronomy is Harlow Shapley and Heber Curtis meeting in Washington, D.C., before members of the National Academy of Sciences to argue the arrangement of the universe. The sides were now clearly drawn, and it was time for a showdown. Shapley had, of course, recently pronounced that the Milky Way was far bigger than previously assumed, easily imagining the spirals as minor players hovering on the edge of our vast system of stars. Curtis, on the other hand, thought otherwise. This epochal encounter is commonly known as the “Great Debate,” al though in truth that's hardly an apt description at all. More like two lectures back to back, the event wasn't covered by even the science-oriented press. In astronomy circles, the venerable legend that surrounds that April session—the memory of it as the mighty clash of cosmic titans, astronomy's version of High Noon —developed gradually over time, the embroidery added so profusely over the years that it was eventually described as a “homeric fight,” two opposing sides battling it out in the highest court of scientific opinion.

This odyssey, though, began quite simply. George Ellery Hale had suggested at a council meeting of the academy that the 1920 Hale lecture on a topic of interest to scientists, an annual event established in honor of his father in 1914, be held during the academy's upcoming spring conference. He himself was leaning toward a discussion of Einstein's general theory of relativity, the trendiest scientific topic of the era. But the academy's home secretary, solar physicist Charles Greeley Abbot, feared that the revolutionary new view of gravity would already be “done to death” by the time of the meeting. The triumphant British solar-eclipse expedition was the science story of the year, still garnering headlines around the world. More than that, Abbot was wary of relativity's radical and difficult-to-comprehend concepts: “I pray to God that the progress of science will send relativity to some region of space beyond the fourth dimension, from whence it may never return to plague us,” he declared. Abbot wondered whether the “cause of glacial periods, or some zoological or biological subject” might be more appealing. The Prince of Monaco was even suggested as a lecturer, to speak on oceanography. But eventually Hale's second choice came to the forefront—the unresolved issue of the island-universe theory.

There was no question that Shapley, the thirty-five-year-old rising star, would defend his Big Galaxy idea. But who to pick for the other side? Lick Observatory director W. W. Campbell was briefly considered to champion the island universes, but Curtis, who had been devoting his professional life at Lick to this issue, was ultimately chosen, as by then he had become the leading spokesman for the claim. In terms of personalities, it was an interesting matchup. Shapley was acknowledged to be the “daring innovator, pressing the last bit of information from his observations, unafraid to extrapolate from the known to the unknown…occasionally depending upon intuition to supply connecting links.” Curtis, on the other hand, was considered a “cautious, sometimes overcautious, conservative who weighed every observation and more often concluded ‘not proven’ than ‘not so.’” Though not as prominent a figure as Shapley, Curtis was a respected astronomer nonetheless; by then forty-seven years old, bespectacled, and far less brash than his younger contender, he struck one as being a distinguished banker. Despite this stolid appearance, however, he proved to be the more venturesome one in regard to the upcoming talk.

With more professional experience under his belt, Curtis was quite comfortable at a podium and eager for a good tussle. But for Shapley, then ill at ease as a speaker, public exposure at this time was problematic. British historian Michael Hoskin first pointed out that Shapley had come to believe he was the front-runner for the directorship of the Harvard College Observatory, one of astronomy's most prestigious positions. Edward Pickering had recently died, having established a monumental legacy, and the search for his replacement was actively under way. Though young and completely untested in managing a world-class research institution, Shapley submitted his name for consideration, wanting to advance his career and strike while the iron was hot. Though he would be leaving the world's biggest telescopes, Shapley was enticed by Harvard's extensive collection of photographic plates, which offered a lush resource for the problems in which he was most interested. “Perhaps Harvard is amateurish, compared with Mount Wilson,” he told Russell, “but you and I…realize the enormous possibilities of the place.” More than that, it was an opportunity for Shapley to get away from his troubled relationship with Mount Wilson's deputy director, Walter Adams. Given this ambition, he worried how he would come across to certain members of the National Academy audience, who might have influence in the final decision. Curtis was known to be a dynamic lecturer; Shapley feared he would look bad by comparison. A letter from Curtis before the debate didn't calm his fears: “I am sure that we could be just as good friends if we did go at each other ‘hammer and tongs.’ … A good friendly ‘scrap’ is an excellent thing once in a while; sort of clears up the atmosphere.”

There was a flurry of correspondence between the participants and the National Academy in the months before the event, aimed at establishing the rules of engagement. Curtis was eager to air the controversy in a no-holds-barred debate. He told Shapley he wanted to “‘take the lid off’ and definitely attack each other's view-point.” But Shapley had a different agenda altogether. He wanted to discuss solely his new super-sized model of the Milky Way and even informed Russell a few weeks before the debate that he didn't intend to say much about the spiral nebulae at all. “I have neither time nor data nor very good arguments,” he lamented. In fact, Shapley was relieved that the chosen title for the lecture, “The Scale of the Universe,” was ambiguous enough to allow him to carry out his plan. Shapley was quite reluctant to dwell on the spiral nebulae, a subject with such uncertain evidence. He hated airing science's dirty laundry in public.

Ardently voicing these concerns, Shapley convinced Hale that the so-called debate should be more of a discussion, “two talks on the same subject.” And instead of forty-five minutes for each speaker, as originally posed, Shapley asked for thirty-five. “My sympathies are with the audience, always,” he argued. “Could it listen to or endure nearly two hours of nebulosity?” Curtis was dismayed by this suggestion; he firmly believed he needed more time to lay out his scientific arguments. “We could scarcely get warmed up in 35 minutes,” he pleaded with Hale. After a while, they all compromised at forty minutes. And there would be no rebuttals. “If you or he wish to answer points made by the other, you can do so in the general discussion,” Hale told Curtis.

Shapley and Curtis were each paid an honorarium of $150, out of which they paid their travel expenses to journey from California to the East Coast. For Curtis it was $2 for the stagecoach to San Jose, then another $100 for the round-trip railroad ticket. By chance both Shapley and Curtis took the same train out to Washington via the southern route, but they agreed not to hash out their ideas ahead of time in order to keep their arguments fresh. When the train broke down at one point in Alabama, they got out and walked around for a while, keeping their conversation focused on flowers and the classics. Shapley didn't forget to collect a few native ants. In all likelihood, they were also, silently and unobtrusively, sizing up their competition.

The annual meeting of the National Academy of Sciences that year extended over three days. During the daytime sessions, a number of outstanding scientists presented talks. Franz Boas, the father of American anthropology, spoke on “growth and development as determined by environmental issues,” and rocket pioneer Robert Goddard advocated the use of rockets in weather forecasting. The “debate,” however, took place on the cool and showery evening of April 26, 1920, at the end of the conclave's first day. The audience of around two hundred to three hundred gathered in the Baird Auditorium of what is now the Smithsonian Institution's Museum of Natural History, prominently positioned along Washington's national mall directly across from the Smithsonian “castle.” In a news report the day before, the Washington Post announced that “Dr. Harlow Shapley, of the Mount Wilson solar observatory, will discuss evidence which seems to indicate the scale of the [Milky Way] to be many times greater than is held… Dr. Heber D. Curtis, of the Lick Observatory, will defend the old theory that there are possibly numerous universes similar to our own, each of which may have as many as three billion stars.”

The proceeding started at 8:15 p.m., and Shapley was the first to speak. He had been right to be nervous; two friends of Harvard president A. Lawrence Lowell—George Agassiz, a member of the Harvard astronomy department's visiting committee, and Theodore Lyman, chairman of its physics department—were in the audience to size him up. But Shapley came prepared. He made sure that Russell, still a valuable supporter of his cosmic model, was in the audience to back him up during the discussion period.

What exactly happened that night—the tenor of the speakers, the reception of the audience—is largely guesswork, based on the limited evidence left behind. Recollections of the event are riddled with false memories. Shapley, for instance, recalled an interminable banquet beforehand with honored guest Albert Einstein whispering to his table-mate that he “just got a new theory of Eternity.” But the conference dinner was the following night, and the noted theorist of relativity didn't make his first visit to America until the following year. However, Shapley did save the typescript of his talk, complete with last-minute scribbles (some in shorthand, a talent honed in his reporting days), which revealed his style and manner. Given the diversity of his audience, many not schooled in astronomy, Shapley chose to avoid technicalities and spent a good portion of his time just presenting basic astronomical facts: He carefully described the size of the Milky Way, its structure, and its constituent parts—the stars, gaseous nebulae, and clusters. He accompanied it with slides of the 100-inch telescope, the Moon, the Sun, the Pleiades cluster of stars, globular clusters. It was a visual tour of the known universe, with special attention paid to making the audience understand the meaning of a light-year. “You do not see the sun where it is, but where it was eight minutes ago,” he instructed. “You do not see these stars as they are now, but more probably as they were when King Cheops was a little boy.”

Instead of talking about the nature of the spiral nebulae, the very reason for the encounter, Shapley focused on his Big Galaxy model. He figured that if he proved the Milky Way was immense, the spiral nebulae would automatically be relegated to minor status in the cosmic scheme of things, mere hangers-on. Anticipating that Curtis would challenge his use of the Cepheids as standard candles in determining the globular cluster distances, Shapley simply ignored the technique in his remarks. “[Curtis] may question the sufficiency of the data or the accuracy of the methods of using it,” he said. “But this fact remains: we could discard the Cepheids altogether, use instead the thousands of B-type stars upon which the most capable stellar astronomers have worked for years, and derive just the same distance [to the globular clusters]…and obtain consequently the same dimensions for the galactic system.” But Shapley was being disingenuous. Two years earlier he had reported to the Astronomical Society of the Pacific that the Cepheids carried “so much greater weight” for his distance measurements and that the magnitudes of red giants and blue stars “can best be used as checks or as secondary standards.”

Shapley went on to stress his finding that the Sun is not at the center of the Milky Way: “We have been victimized by the chance position of the sun near the center of a subordinate system [of stars], and misled by the consequent phenomena, to think that we are God's own appointed, right in the thick of things.” As for the spirals? “I shall leave the description and discussion of this debatable question to Professor Curtis,” he said. Shapley conceded that the possibility remained that they were comparable galactic systems, but only if the Milky Way were cut down to a tenth of his newly defined dimensions. He believed that unlikely and preferred to think of the spirals as nebulous objects. He maintained “that it is professionally and scientifically unwise to take any very positive view in the matter just now.”

One can imagine Curtis's mounting dismay as his opponent was progressing through his talk. Shapley had spent most of his time on just the basics of astronomy, while Curtis had prepared a full-fledged analysis, laden with scientific detail. The Lick astronomer was about to address the audience on issues that Shapley had never brought up. While he anxiously awaited his turn at the podium, his mind raced, wondering whether he should change his approach on the fly, making his presentation more relaxed and general. But in the end he decided to stick to his original plan.

Unlike with Shapley, a copy of Curtis's script no longer exists, but some of his slides, displaying his essential points, do survive, and they provide a glimpse of the flow of his arguments that evening. Contrasting sharply with Shapley's popular approach to the topic, Curtis's talk was more technical, although by all accounts he spoke more spontaneously. At first he focused on one of his major disagreements with Shapley: the size of the Milky Way. He carefully outlined his reasons for believing that the Milky Way was a tenth the size that Shapley was hawking. Mainly, he had no confidence in Shapley's use of the Cepheids. Shapley himself knew that his momentous refashioning of the Milky Way's size stood on the foundation of eleven “miserable” Cepheids, as he had earlier described them in a letter.

From there Curtis went on to focus on the spiral nebulae, the subject that Shapley conveniently avoided. Curtis showcased his best evidence, echoing many of the points he had made to the Washington Academy of Sciences just the year before: He stressed that the spiral nebulae displayed the spectra typical for collections of stars—not gas; that not one spiral had ever been found within the Milky Way itself; that the spirals are primarily seen away from the Milky Way, because obscuring matter blocks the view through the plane of our galaxy. He paid special attention to the many novae being sighted within some spirals. He showed that if the flare-ups in Andromeda were half a million light-years distant, their luminosity would roughly match those seen in our own galaxy. Any closer and they would be far too bright. And then there was the movement of the spirals detected by Slipher; the spirals traveled speedily through space unlike any other celestial object in the Milky Way, which suggested that they had to be located outside our galaxy's borders.

All in all, the two men were simply talking at cross purposes. Shapley primarily defended his new vision of the Milky Way—its unexpected bigness—while Curtis hammered away on his contention that the spiral nebulae were far-off galaxies. In hindsight, each turned out to be partly right and partly wrong. Shapley argued for his larger Milky Way (true) but insisted that the spirals were local (wrong). Curtis still believed in a smaller home galaxy (wrong) but persevered in his belief that the spiral nebulae were situated far outside the Milky Way and rivaling it in size (true). At the end of the day, it was a wash.

Everyone in essence went home maintaining the beliefs they held at the start of the lecture. The data were so muddled that Curtis and Shapley could take the same facts and arrive at completely contradictory conclusions. At the time of the debate, there was no overwhelming evidence to settle the inconsistencies either way. Both men were traveling along a precarious road, each viewing his destination through an obscuring fog and interpreting the hazy view in different ways.

There was a winner, however, for best presentation. Curtis headed off that night feeling pretty good about his performance. He received assurances afterward that he “came out considerably in front.” Shapley, on the other hand, was judged more poorly. Russell wrote Hale afterward that his former student sorely needed to enhance the “gift of the gab.” Agassiz, the Harvard evaluator, was not impressed by Shapley's performance at all. “He has … a some what peculiar and nervous personality…lacks maturity and force, and does not give the impression of being a big enough personality for the position,” he reported back to Harvard's president two days after the event. More attractive to Agassiz was Russell, who spoke quite eloquently that night in support of Shapley's arguments during the audience response period. He said that Russell had “more balance more force and a broader mental range.”

The two opponents came to acknowledge what others sensed all along over the course of that April evening. “Yes, I guess mine was too technical,” admitted Curtis to Shapley a couple of months after the debate. “I thought yours would be along the same line, but you surprised me by making it far more general in character than I had expected.” As captivating scientific theater, the so-called Great Debate was ultimately a letdown.

A full year after the debate, however, the two astronomers battled it out once again within the Bulletin of the National Research Council. The original intent was to simply print the lectures they had given before the National Academy. But as the articles were being prepared, each man deepened and extended his arguments. It was not during that misty spring night in Washington that Shapley and Curtis had their great debate but rather within the pages of the Bulletin. It was the written version, vastly altered and amended, that ultimately established the legend handed down by succeeding generations of astronomers, many coming to believe it was the bona fide transcript of the April scrap.

At first Curtis wasn't keen on publishing his comments, but he indicated he would be willing if both he and Shapley delved more deeply on the technical issues. Shapley agreed. They were originally asked to keep to ten pages, which Curtis joked would force him to follow the laws of writing “generally observed in composing telegrams.” Perhaps, he wrote Shapley, they could “shoot our arrows into the air, to let them fall we know not where.” With his customary down-home wit, Shapley suggested that he would provide “ten pages of buncombe and flapdoodle,” while Curtis could supply “ten more pages of wisdom.”

Shapley also wondered if they should exchange their papers, providing the opportunity to rebut each other's arguments. “Should I go ahead, shoot my shot (or wad), then you use your shillelah (or hammer), then I sneak up behind you and apply my ole horn-handle.” Curtis was game, and over the ensuing months a lively train of drafts and comments went back and forth between them. In the process Curtis pushed Shapley to devote more space to the spiral nebulae, “at least a brief statement of how you explain them if not island universes.” Upon completion, their published remarks each expanded from ten pages to twenty-four, and though it wasn't mentioned until his penultimate paragraphs, Shapley's strongest and freshest ammunition against Curtis involved the spirals. It was there at the end that he played his definitive trump card: The spiral nebulae could not possibly be island universes, because the rotations measured by Adriaan van Maanen at Mount Wilson “appear fatal to such an interpretation.” Shapley now seemed more at ease in dismissing the spirals as simply minor objects. “I see no reason for thinking them stellar oruniverses,” he told Russell during the course of his writing the Bulletin article. “What monstrous assumptions that requires before you get done with it.” From that point on, Shapley's strongest weapon against supporters of distant galaxies was van Maanen's twirling spirals.

Although in his heart of hearts he never believed it would happen, even Curtis had to grudgingly concede in his published response that if van Maanen's findings held up “the island universe theory must be definitely abandoned.” Over the succeeding years, van Maanen and his observations stood like a giant wall before island-universe advocates. If the spiral nebulae were truly remote and massive galaxies, how could you possibly explain seeing them rotate over just a few years from so far away? The island-universe theory would not gain general acceptance until its supporters figured out how to breach this formidable rampart.

Van Maanen had begun his measurements of the spiral nebulae in 1915 and continued into the early 1920s. Astronomers took his results seriously because his reputation was exemplary. He was known to be a careful observer who followed intricate astronomical procedures to the letter. And it was easy to accept his conclusions, as they supported an idea of the universe that many readily believed at the time: The Milky Way defined the universe, and the spirals were mere appendages that from their swirling appearance had to be turning. Stars, planets, and moons rotated; planets revolved around the Sun; rotation was a natural feature of the universe. Given that, it was not surprising to hear that the spirals were rotating. In 1914 Slipher had already reported on a spiral rotation from his spectroscopic data, but simply viewing the curving lines of a spiral's misty arms, captured so vividly in photographs, made it impossible to think otherwise.

Adriaan van Maanen (left) with Bertil Lindblad (Photograph by Dorothy
Davis Locanthi, courtesy of AIP Emilio Segrè Visual Archives)

Van Maanen was the descendant of an aristocratic family in the Netherlands, whose ancestors were ministers, teachers, and noted jurists. Those who knew him attested to his meticulous integrity and high sense of personal honor, instilled by his family's esteemed heritage. After earning his doctorate in 1911, van Maanen had traveled to the United States to work as a volunteer assistant at the Yerkes Observatory. But after his mentor, the noted Dutch astronomer Jacobus Kapteyn, brought him to the attention of Hale, he was soon offered a permanent position at Mount Wilson. The observatory wanted him for his superb and proven skills in gauging the motions and distances of stars. In 1917, in carrying out such observations, he discovered the second known white-dwarf star, a rare find at the time.

Van Maanen was drawn as a student toward this line of work, an endeavor that other astronomers tried to avoid because of the tedium and difficulty in discerning the change in a star's position over time. The procedure involved comparing, with intense concentration, photographic plates taken over intervals of months or years. But to van Maanen the routine was heaven; he even went back to the pursuit two years before his death. “One always returns to one's first love,” he scribbled on a copy of his 1944 paper on stellar parallaxes. To carry out the task, he superimposed the pictures of the stars taken at different times in his special stereocomparator (more often called the “Blink”). This machine allowed the viewer to quickly alternate between two photographic plates taken of the same field at different times. The blinking proceeded so rapidly that an object that had moved between pictures would immediately stand out, while those that remained fixed appeared still. Van Maanen could then slowly turn a micrometer screw to measure the star's exact advancement across the sky. The number of turns measured off the change in the star's position—the amount the star had moved over the years. It was his most cherished instrument at the observatory's Pasadena headquarters, and everyone knew it: The warning “Do not use this stereocomparator without consulting A. van Maanen” was blatantly posted on its front. The sign remained there for decades, long after he had left.

Sociable and well-liked, “Van,” as everyone called him, played a good game of tennis and made sure newcomers to the mountain felt right at home. He was a lively storyteller and also a bit of a playboy. Once Shapley arrived, he and van Maanen became fast friends, as they were nearly the same age. “He could go to a dinner and soon have the whole table laughing,” recalled Shapley. An accomplished chef, van Maanen relished throwing parties where he could put his culinary skills into practice and prepare fine dishes for his cohorts. Shapley and van Maanen further bonded when they discovered they were both disliked by Adams, who was suspicious of them for their liberal outlooks and ambivalence toward the ongoing war in Europe, as well as their ambitions. “Van Maanen and I are in ill-favor because we do or try to do too much,” confided Shapley to a friend.

One of van Maanen's first jobs at Mount Wilson was to measure photographs of spectra taken over the face of the Sun, an endeavor that helped Hale map the Sun's magnetic field. Early reports suggested that the strength of the magnetic field varied with solar latitude, and van Maanen always seemed to see this effect, even though it was later found to be a mistake. Van Maanen's persistence in finding the change was a harbinger of trials to come in succeeding years—not concerning the Sun but rather the spiral nebulae.

Van Maanen first got involved with the spirals in late 1915, when George Ritchey asked him a favor. Ritchey was then using Mount Wilson's 60-inch telescope to produce superb photographs of spiral nebulae. Everyone agreed the images were breathtakingly beautiful. Part of this success was due to Ritchey's inventiveness. He had developed a fast camera shutter, which allowed him to build up an image from a series of short exposures, each taken when the atmosphere was calm. The total exposure time could last anywhere from two hours to more than eight hours, sometimes stretched over two or three nights. This resulted in rich nebular details never before captured.

When he approached van Maanen, Ritchey had recently taken a picture of the spiral nebula M101, the Pinwheel, which he had also photographed in 1910. With both images in hand, he asked van Maanen to put the plates into his trusty stereocomparator and see if any changes could be detected in the nebula over those intervening years. Van Maanen at first measured no variation but got permission from Ritchey to keep the plates to study them further. Adapting methods he had used previously in other work, he chose thirty-two stars, equally bright and positioned uniformly around the nebula on each plate, and measured how dozens of points within the spiral nebula may have shifted in comparison to those stars. Extending his study, he borrowed additional plates of M101 from the Lick Observatory, photographs taken in 1899, 1908, and 1914. He didn't rush. Van Maanen was so meticulous that he even made sure the temperature in the room where he was making his measurements was tightly controlled. He wanted to take care that any thermal expansion, in either the glass photographic plate or the measuring machine, would be negligible.

In the end, he decided that the nebular material within M101 was moving after all, although exactly how was not immediately obvious. “If the results…could be taken at their face value, they would certainly seem to indicate a motion of rotation or possibly motion along the arms of the spiral,” he reported. If turning at his measured rate, M101 was completing one full rotation every eighty-five thousand years. As noted earlier, that meant if the Pinwheel were truly the size of the Milky Way and located way off in distant space, the nebula's edge had to be traveling faster than the speed of light, an impossibility given Einstein's special theory of relativity, which said that no bit of matter can move fast enough to overtake a beam of light.

Given what was at stake, van Maanen followed all the precautions: He switched the plates in the holders to eliminate machine error, and he got a colleague to redo the measurements with a different machine, to make sure there wasn't an instrumental error or personal bias. He came to believe that the matter within the spiral was drifting away from the center—outward along the arms—and noted in his report that this agreed with the Chamberlin-Moulton model on the origin of spiral nebulae, which involved a collision between a star and a nebula. Thomas Chamberlin was elated to hear the news from Hale. “While the recent revival of the notion that spiral nebulae are mere distant constellations has not seemed to me to have any substantial basis, it is a satisfaction to feel that definite evidence is about to give it a quietus,” he responded.

Van Maanen was aware that his work “might indicate that these bodies are not as distant as is usually supposed to be the case,” but he kept that speculation out of his early reports. That's partly because in 1917 he measured a rotation for the Andromeda nebula with error bars larger than his result. “So that we do not know yet if this is an island universe!” he told Hale.

But that was the exception. Van Maanen primarily got the answer that many expected: Spiral nebulae exhibited internal motions and so must be relatively nearby. Moreover, the announcement was being made by a widely respected astronomer working at the world's premier observatory, whose expertise in stellar measurements was lauded. “His wide experience in astrometric work,” Walter Adams later recalled, “gave his conclusions a high standing among astronomers.” Other observers even confirmed that the spirals were changing; concurring reports came out of Mount Wilson, the Lowell Observatory, and observatories in both Russia and the Netherlands. It became the conventional wisdom among astronomers. And why not? It fit the general opinion of the time.

Only a few, such as Heber Curtis, openly disagreed. Curtis, with his wealth of spiral nebulae photographs at Lick, had earlier attempted to measure a change in the spirals over the years but could only conclude that “a much greater time interval will probably be necessary before nebular rotations can be definitely established.” He knew that many of the older plates that van Maanen was perusing were very poor and useless for measuring anything. To Curtis comparing a photograph made at Lick in 1900 with a more modern picture taken with a completely different telescope at Mount Wilson was a fool's errand, which is why Curtis found it easier than Shapley and others to dismiss van Maanen's data right away. “The mean of five measures each of which is not worth a damn, has a maximum value of only five damns,” he liked to say sarcastically.

But Curtis's warning was not heeded. With hindsight, it now seems easy to dismiss van Maanen's measurements. But at the time it was extremely difficult to assess. Van Maanen was no slouch at telescopic measurements and his finding a spiral rotating appeared quite reasonable. One of the era's leading theorists, the Britisher James Jeans, was especially eager to jump on van Maanen's bandwagon. Upon hearing the Dutch astronomer's results, Jeans speedily sent off a letter to the journal Observatory, saying they were “entirely in agreement with some speculations in which I have recently been indulging.” In calculating the behavior of a blob of gas, rotating and condensing, he had determined that tidal forces would lead to the formation of spiraling arms. And now Van Maanen was providing the observational evidence to back him up. Jeans eventually wrote up his ideas in the book Problems of Cosmogony and Stellar Dynamics, which exerted a tremendous influence on astronomers at the time. Moreover, both van Maanen and Jeans began to calculate higher masses for the spirals. So, instead of a single solar system in the making, they began to think of a spiral nebula as the start of a dense (but still small) cluster of stars.

As more plates became available, van Maanen expanded his study to include other spirals. He measured a rotational period of 160,000 years for M33 (Triangulum), 45,000 years for M51 (Whirlpool), and 58,000 years for M81, a handsome spiral in the Ursa Major constellation. Other nebulae followed. All were rotating in such a way that the spirals appeared to be unwinding their arms, spreading them farther outward. He figured the spirals were no more than several hundred light-years wide and ranged in distance from one hundred to a few thousand light-years away.

Soon van Maanen was running out of spiral nebulae to measure, as few had been regularly photographed for comparison over the years. As a double check on his dexterity with the Blink, he measured a simple globular star cluster, M13, which was known not to rotate. If there were any instrumental error, he should have mistakenly measured a motion, but he didn't, which seemed to imply his methods were valid. A British astronomer independently checked his methods as well and concluded that no one “would be so bold as to question the authenticity of the internal motions…. In fact, the more one studies [van Maanen's] measures, the greater is the admiration which they evoke.”

Adriaan van Maanen's markings on a photo of M33 indicating the
rotation he measured (From Astrophysical Journal 57 [1923]:
264-78, Plate XIX, courtesy of the American Astronomical Society)

“I finished…my measures of M51,” van Maanen wrote Shapley in the spring of 1921. “The results look more convincing than M101… Motion outwards along the spirals + some motion away from the center…. By this time Curtis and [Swedish astronomer Knut] Lundmark must be the only strong? defenders of the island-universe theory.”

“Congratulations on the nebulous results!” responded Shapley. “Between us we have put a crimp in the island universes, it seems,—you by bringing the spirals in and I by pushing the Galaxy out. We are indeed clever, we are.” Shapley reported on his friend's latest results at that summer's American Astronomical Society meeting in Connecticut. “I think that your nebular motions are taken seriously now,” he told van Maanen afterward, “and nobody…dared raise his head after I explained how dead the island universes are if your measures are accepted.”

The two were feeling quite cocky. At this stage, van Maanen at last made it publicly known, in the Proceedings of the National Academy of Sciences, that his observations “raise a strong objection to the ‘island-universe’ hypothesis.” If M33, the prominent spiral in the Triangulum constellation, for example, were several million light-years distant, he pointed out, the motions he detected would represent velocities near the speed of light, “which, obviously, are extremely improbable…[and] afford a most important argument against the view that these nebulae are systems comparable with our galaxy.”

But this declaration was hardly a resolution to the Great Debate. While Shapley and van Maanen were smugly celebrating, Knut Lundmark was visiting the Lick Observatory, using the Crossley reflector to gather the extremely faint light of M33. It was a difficult task, requiring extremely long exposures, one totaling thirty hours collected over four nights. Lundmark eventually saw that the light from the nebula's spiral arms resembled nothing less than the light of ordinary stars. Where other astronomers had seen a fuzzy patch in a spiral arm and called it a “nebulous star,” Lundmark pondered whether each mistlike spot was instead “a great number of very distant stars…crowded together [to] give the impression of nebulous objects.” That led to his cogent conclusion: that his observations of the spiral arms “speak for a large distance.” The respected Swedish astronomer soon became one of the loudest voices championing the existence of other galaxies, and Shapley began to feel sizable pressure on his beloved model of the universe under Lundmark's onslaught.

Meanwhile, Slipher, in Arizona, had dispatched a story to the New York Times revealing that he had found a new “celestial speed champion,” a faint spiral nebula that he judged had to be “enormously large” and “many millions of light years” away. And in the following year, 1922, Ernst Öpik, at the Dorpat Observatory, in Estonia, carried out an elegant calculation demonstrating how the Andromeda nebula must be some 1.5 million light-years distant. He did this by assuming that its mass and luminosity were comparable to those of the Milky Way. This “increases the probability,” reported Öpik in the Astrophysical Journal, “that [Andromeda] is a stellar universe, comparable with our Galaxy.” Thrust and parry. Thrust and parry. The duel over the island universes continued. Nothing would be settled until astronomers obtained a clear and unequivocal distance measurement to a spiral nebula—an observation so clear, so decisive, so comprehensive, that it immediately quelled all doubts.

Poor Shapley, it turns out, did put himself in jeopardy with his performance before the National Academy of Sciences gathering. Still in his thirties, Shapley was judged as too impetuous and immature to be the head of the Harvard College Observatory. Instead, his Princeton mentor, Russell, was offered the position. “Shapley couldn't swing the thing alone,” Russell confided to Hale two months after the conference. “I am convinced of this after…observing Shapley at Washington. But he would make a bully second … if he grew intellectually he would be a prodigy!”

Russell gave the Harvard directorship intense consideration, with the understanding that Shapley would be his assistant. “At this point,” continued Russell to Hale, “I would like to see your expression! I know I have my nerve with me: but,—and here I am very serious indeed,—consider what Shapley and I could do at Harvard! Between us, we cover the field of sidereal astrophysics pretty fully…and I might keep Shapley from too riotous an imagination,—in print.”

But Russell, after nerve-racking deliberation and an attractive counteroffer from Princeton, ultimately declined the job (“I would rather do astronomy,” he confided to Shapley). Harvard came back to Shapley, but not for the top position. Harvard officials brought up the title of “Chief Observer or something of the sort.” He, a bit miffed, curtly turned it down. A month later, however, Shapley reversed his decision when Harvard (spurred by a suggestion from George Hale) agreed to try him out for a year as chief of staff, starting in the spring of 1921. He obviously passed muster, for he was soon named full director and served at the post for thirty more years, working at his unique desk that turned like a wheel—“a kind of rotating galaxy for ideas,” noted a friend.

Shapley breathed new life into the sclerotic institution, bounding up the stairs two steps at a time and greeting everyone with a sporty cheerfulness. “He cast spells over people,” said one staff member. Pickering had run the observatory like an absolute monarch. Under the youthful and energetic Shapley, it became a band of enthusiastic workers. Leo Goldberg, a student at Harvard in the 1930s, compared him to a benevolent Mafia Godfather. On the one hand, “he inspired us all,” said Goldberg. “He pepped us up, he raised us out of the depths of discouragement many times.” But a darker side lurked within Shapley as well. Adopting a “divide and rule” principle, he could be a father figure to some, while a tyrant to others. He also stubbornly ignored new scientific data at times, if it conflicted with his personal vision of how the universe should work.

Even as Shapley settled into Harvard, his former employer requested one more task from him. He was asked to contribute to Mount Wilson's annual report, to recount the final work he carried out there in 1920. “I thought I told you that I left Mount Wilson just to avoid this ordeal,” he replied playfully. “Suppose I had lived wickedly and unrepenting died—would you even then haggle with His Majestic Nibs for your annual tithe of Blood-and-Brain?” Shapley was again being Shapley. It was his last hurrah for the California observatory. Mount Wilson got its notes.

Meanwhile, Curtis, who could have done much toward solving the mystery of the spiral nebulae, stepped out of the race entirely. Just a few months after the debate, he left the Lick Observatory to become director of the Allegheny Observatory, the same post that James Keeler once held. He had actually tendered his resignation ten days before the Washington debate took place. Being in charge of an observatory, a more highly paid position with increased prestige, was an opportunity hard to pass up, especially for a family man. But, as with Keeler, the urban setting, cloudy weather, and poorly equipped telescope at the Pennsylvania observatory ultimately prevented Curtis from making any further cutting-edge discoveries. Some considered it “the biggest mistake he ever made.” Even Curtis later confessed to his former boss Campbell that “the California combination of instruments PLUS climate is a hard one to beat… There is no place like the hill [Mount Hamilton] for astronomical work and…any man who leaves these opportunities is bound to be sorry for it.” A visiting colleague found him at Allegheny one day puttering with an instrument and chided him for turning into a toolmaker. “You play golf don't you? Well, this is my golf,” he responded.

Harlow Shapley at his wheel-like desk at the Harvard Observatory
(Harvard College Observatory, courtesy of AIP Emilio Segrè
Visual Archives)

Despite their differences in cosmic outlooks, Curtis and Shapley remained cordial over the years and kept in touch through correspondence. More than two years after the debate, Curtis looked back on the event—what he called their “memorable set-to”—with good humor. “I have always thought that the clubs we wielded at each other were all the more effective because politely padded,” he told Shapley, “and regard with approbation the view-point of the old lady who warmed the water in which she drowned the kittens…. I fancy we both are as stiff-necked as ever; am sure that I am; cant [sic] see that my views have changed in the slightest.” With his new responsibilities at Allegheny, though, Curtis had to remain on the sidelines, resigned to simply “watching the strife with interest,” as he put it.

A few years later, a friend from Lick asked Curtis what he would have done with the Crossley if he had stayed on Mount Hamilton in 1920. Curtis replied that he would have just kept “photographing, photographing, and yet more photographing.” He had in mind a “program of about 30 min. exposures of all the larger spirals at frequent intervals, to hunt for novae and variables.” In a nutshell, he would have done everything that Edwin Hubble later carried out at Mount Wilson using its 60-and 100-inch telescopes, but with a few years' head start. Was the Crossley up to the task? Curtis had total faith in his beloved telescope: “I am copying that instrument in my design far more than any other,” he said. “Could a ‘race’ be run between the 60″ and the Crossley, would bet on the Crossley every time.” Others, too, later judged the Crossley as having had a fighting chance at clinching the distance to Andromeda. But once Curtis left for Pennsylvania, no other Lick astronomer was interested in photographing the spiral nebulae. In effect, once Curtis left the Lick Observatory, it handed the baton over to Mount Wilson.