Margaret Geller first met the stickman in the fall of 1986. While the exact date has faded from her recollection, she remembers the time as midafternoon and her reaction as a kind of euphoria. No one had ever seen the stickman before--at least, not really. Valérie de Lapparent, who was Geller’s graduate student, noticed it but says she was too inexperienced to understand its implication. John Huchra, who was Geller’s collaborator at the Harvard-Smithsonian Center for Astrophysics (cfa), says he took one look at the stickman and assumed he had botched his observations. It took Geller’s eye to recognize the stickman as something real and important.
Geller, Huchra, and De Lapparent had mapped the nearby universe, taking several months to carefully measure the distance to 1,000 galaxies, some as near as 30 million light-years away, others as far as 650 million. De Lapparent had fed the distance and positions of those galaxies into a computer program that printed out a two-dimensional representation of their three-dimensional distribution in the universe. On the printout was this slice of the northern sky, sprinkled with 1,000 distant galaxies, and smack in the middle, says Geller, was this remarkable stickman figure. The distribution of galaxies looked like a child’s drawing of a somewhat bowlegged person. It’s a whimsical name for a grand figure: the stickman extended 500 million light-years across the universe. Its torso was composed of hundreds of galaxies, a massive congregation known to astronomers as the Coma cluster. Its arms were two more sheets of galaxies streaming across the night sky.
The stickman was grand not just in dimension but in destiny. You might even say it changed our understanding of the universe. Until the stickman, the universe appeared to be a smooth and homogeneous place. Astronomers believed that galaxies were distributed at random, although they might occasionally form clusters like Coma containing as many as a thousand or so galaxies like the Milky Way. There was even some evidence that the universe contained at least one enormous void, in the constellation Boötes, which seemed to extend for some 200 million light- years--and other suggestions that galaxies could be found strung out on long filaments. But in 1985 most astronomers assumed these structures were products not of the universe itself but of the methods used to survey it.
Then Geller saw the stickman, which constituted compelling evidence that galaxies were congregating on two-dimensional structures, as though they had condensed out of the cosmic nothingness on the surfaces of invisible bubbles. Indeed, when Geller later wrote up the results of the cfa galaxy survey, she described the distribution of galaxies in the universe as looking like a slice through suds in the kitchen sink. Her metaphor implied that astronomers were mightily confused about how the universe had formed.
The very early universe, around the time of the Big Bang, was a smooth place. We know that because the Big Bang left an imprint: the cosmic background radiation, which is a radiation 3 degrees above absolute zero that pervades the entire universe. That background radiation is considerably smoother than a baby’s behind, and it means the universe, when it was a couple of hundred thousand years old (and maybe even younger), was equally smooth. Now it’s not. It’s full of these enormous two-dimensional structures. Perhaps the most awe-inspiring is one that Geller and Huchra discovered in 1989, known as the Great Wall: a sheet of galaxies extending for at least 500 million light-years, stretching across the entire northern sky. It may indeed be bigger than 500 million light-years, but no one can yet tell.
The confusion comes about because astronomers can see the huge structures at the very limits of their vision, which means when the universe was considerably younger than it is today. When we look out into space, we’re looking back in time; the light from a galaxy a billion light- years away, for instance, will take a billion years to reach us. It’s an amazing thing,’’ says Geller. The history is there for us to see. It’s not mushed up like the geologic record of Earth. You can just see it exactly as it was.’’
So what happened? The universe is full of these prodigious two- dimensional structures as far out as we can see, and thus was full of them as far back as we can see. In the 10 or 15 billion years the universe has taken to grow up, it has evolved from something unimaginably smooth into this sink suds of a structure, and no one yet knows how or why.
Geller is at work trying to answer this question, as are at least a hundred other astronomers around the world. Mapping the structure of the universe has become a cottage industry in astronomy; since the stickman made his appearance, astronomers have initiated more than a dozen surveys to chart the distribution of galaxies. Geller’s, with cfa astrophysicist Dan Fabricant, may be one of the deepest. The two are working on a survey that should begin probing the universe in late 1998 at the rate of thousands of galaxies a night. By the time they’re done, they will have surveyed more than 50,000 galaxies and mapped strips of the universe out to a distance of 5 billion light-years. This might just be far enough out-- that is, far enough back in time--to understand why the universe we see appears so profoundly different from the universe of the Big Bang. Now that we know something about what the nearby universe looks like,’’ says Geller, the issue everybody wants to understand is how it got that way. And the race is on to find that out.’’
The discovery of the stickman may have changed Margaret Geller as much as it did our conception of the universe. It launched her into the stratosphere of science and linked her name with the idea of mapping the universe and with the structure of the universe itself. The morning after Geller first displayed the stickman to her fellow astronomers at a meeting in Houston, she made an appearance on the Today show, complete with the stickman and the news that the universe was a considerably more perplexing place than previously imagined. She went on to win a MacArthur Fellowship and to become a star of the Smithsonian Center for Astrophysics and the Harvard Observatory. All those achievements were not enough, however, to make Geller feel at home in astronomy.
There is a theory that creativity arises when individuals are out of sync with their environment. To put it simply, people who fit in with their communities have insufficient motivation to risk their psyches in creating something truly new, while those who are out of sync are driven by the constant need to prove their worth. They have less to lose and more to gain. The theory of asynchronicity might help explain Geller, who has been struggling to fit in since she began studying astronomy. Geller’s first love was acting, but her father, a physical chemist who worked on crystal structures, did what he could to encourage her in science. He would take her to his lab at Bell Laboratories in New Jersey, then in its heyday, where she would play with his state-of-the-art, hand-cranked calculator. The biggest challenge was to make it make a big racket for as long as possible,’’ says Geller. By the time she was ten, she was working out simple calculations. Because she didn’t like going to school, her parents let her teach herself. Her mother, from whom Geller inherited her fascination for language, would take her to the library and help her choose books, then supervise her study at home. Geller would show up at school to take tests and little more.
As an undergraduate at the University of California at Berkeley, Geller first took up math but moved on to physics. I didn’t know what kinds of questions to ask in mathematics,’’ she says. In physics, I could see there were things that were known and things that weren’t.’’ She went back east to Princeton for graduate school, where she studied astrophysics, learned about galaxy surveys, and pored over galaxy catalogs. She also lost her confidence, falling out of sync with her environment. Until then, she says, it never occurred to me that there might be something I wasn’t able to do.’’
She hated Princeton. Only one woman--Glennys Farrar, now at Rutgers University--had successfully obtained a doctorate from the Princeton physics department before Geller arrived in 1970. Princeton had just admitted its first women as undergraduates that year, and the atmosphere in the physics department, according to Farrar, was terrible.’’ Geller says she wasn’t mature enough to deal with it: Students would ask me what I was doing in physics at Princeton when men couldn’t get jobs in physics, or they would say, ‘Only one woman has passed her generals in the department and three have been admitted since, and they all failed. So there’s a 75 percent chance you’ll fail.’
Until she got to Princeton, Geller says, she simply never realized how few women there were in science. Somehow it hadn’t registered. But I never had a single female professor throughout my whole education, from the beginning of university to the end. Even all the books were about men; I never really liked reading books about the history of science, and I never really understood why.’’ At Berkeley, it hadn’t bothered her, perhaps because the physicists were used to dealing with women as undergraduates. I had a great deal of confidence when I graduated from Berkeley,’’ says Geller. I had almost none when I was at Princeton. After a while when people tell you you can’t do something because you’re a woman, you begin to believe maybe they’re right. It’s amazing, because even though you know these things are totally irrational, they stick with you for many, many years.’’
Geller often thought of quitting Princeton, but her parents talked her out of it. They said I shouldn’t quit, because I’d never failed at anything, and if I left I’d feel that I failed and that would haunt me my whole life. Get the degree and then quit.’’ She got the degree but didn’t quit. The experience toughened her. The Nobel laureate Steven Weinberg, who was at Harvard in the mid-1970s when Geller arrived from Princeton, says this is what stood out about her. I admired her,’’ he says. Not only her intelligence, but she had a certain toughness, a strength of purpose in pursuing her science, in making sure what she thought should be done would be done.’’
Nonetheless, as Geller admits, her first years at Harvard could be called her lost years. She had yet to recover from the Princeton experience, and she now had mixed feelings about astrophysics: I felt I could do it, but did I want to do it?’’ Finally, in 1979, she spent a year at Cambridge University in England and thought things over.
Her self-appointed task at Cambridge was to assess the state of knowledge of the universe’s structure. I realized almost nothing was known about anything outside our galaxy,’’ she says. Galactic surveys had been done, but they were small and useless for drawing firm conclusions about how galaxies were distributed in the universe. To gauge galactic distances, the surveys studied redshifts, which are measures of how much the wavelengths of light from distant objects stretch toward the red end of the spectrum the farther away those objects happen to be. Huchra and Harvard astronomer Marc Davis, now at Berkeley, had measured redshifts of 2,400 galaxies going out barely 300 million light-years, our own backyard by the standards of the universe. That survey was too shallow to see the remarkable patterns that were uncovered later.
Bob Kirshner, who was then at Michigan and is now at Harvard, had led a deeper survey of a few hundred galaxies but had measured them all in a few tiny regions of the sky. Geller compares that technique to sticking a few needles in a haystack’’ to map the internal structure of the stack. The survey had detected the great void in the constellation Boötes, but few astronomers believed the void was real. Everybody was skeptical,’’ says Geller. I thought there was something wrong with the survey, because the void was much larger than any structure anybody thought existed.’’
Geller wrote no papers during her year in England. What I did was much more valuable, she says. I figured out the kinds of problems I thought were interesting.’’ Back in Cambridge, Massachusetts, she began studying galaxy clusters with John Huchra, which led her to think about the distribution of galaxies on a larger scale. Somebody, she decided, should do a survey reaching deep into the universe, one that would be able to see very large patterns like the Boötes void, if it was real. She had no trouble convincing Huchra, who had worked on the redshift survey with Davis and knew that the field needed a deeper one. Geller would design the project and crunch the data, and Huchra and De Lapparent would spend the nights at the telescope making the observations. They just had to decide what part of the sky to map.
That’s where they got lucky, as they all admit today. Geller wanted to measure the redshifts of galaxies that could be found in a continuous strip across the sky. She argued that by examining long continuous strips rather than isolated patches, they could find large structures and understand their geometry. Huchra agreed because strips are easy to measure; you just let the sky roll over your telescope, and it does the hard work for you. The place where luck was involved was the width of the strip, says Geller. We were lucky it was wide enough to see the stickman.
Geller thinks visually, seeing problems as patterns and geometries, so the appearance of the stickman represented a convergence of some dominant themes in her life. The choice of strips was a geometric choice, and the stickman itself was a pattern that emerged from the background clutter of the universe. It had been waiting, in a sense, for Geller to interpret its meaning. In the months after the discovery, she spent considerable time creating a video of the stickman, one that would convey to her colleagues the beauty of the pattern, and with it her sense of wonder and revelation. Using what she calls incredibly primitive graphics,’’ she and her cfa colleague Michael Kurtz made a display of their slice of the universe. We would sit there absolutely mesmerized by this one slice moving around,’’ says Geller. We would stare at this thing over and over and over again. It was as if we were high on something.’’
With the discovery of the stickman, Geller had arrived. She had established herself, in Weinberg’s words, as an adornment to the astronomy establishment at Harvard.’’ And when she and Huchra completed four more slices of the redshift survey, she wrote a paper for Science on the Great Wall and the art of universe mapping. She then won the MacArthur Fellowship, which gave her the cachet to raise $200,000 to make a documentary film about her work with Huchra and other colleagues. The film rejuvenated her, taught her new skills, and gave her new ideas.
With all the accolades, however, Geller was still only a senior researcher at the Smithsonian Astrophysical Observatory, and although she has a professorial appointment at Harvard, she has never been given tenure. I drive myself crazy with this periodically,’’ she says. I keep wondering, ‘Why me, why don’t I have this?’ I’ve had periods where I find it very difficult to work because of this. Then I think, ‘The hell with it, I’ve shown I can do it. Why don’t I go do something else?’
That question is borderline rhetorical. If it needs an answer (beyond her passion for her pursuit), one might be the new survey and the instrument Dan Fabricant is constructing. The program Geller set in motion at the cfa is about to pay off, and she wants to be there to see it. While Fabricant’s reputation puts him in the top handful of astronomical- instrument builders, Geller gets some small credit for his accomplishments. She met Fabricant, who’s now 45, while he was still in graduate school. Later, when the Harvard-Smithsonian optical astronomy community was ignoring his ambitions, Geller was cultivating them.
Fabricant started his career building instruments to do X-ray astronomy from rockets and later from satellites. But when the space shuttle Challenger exploded in 1986, the launch of any scientific satellite, X-ray or not, began to look like a debatable proposition. Fabricant switched his focus to optical astronomy, which was about to undergo its third major revolution in 300 years. The technology that would make it happen is known as multiplexing--not to be confused with having a choice of 17 movies every time you go to the theater, although the concepts are similar.
The first revolution began when Galileo invented the telescope. For the next 300 years, the science of astronomy advanced because astronomers made bigger and bigger telescopes, which allowed them to collect more and more light and see ever fainter objects. By the 1950s, with the advent of the 200-inch (5.1-meter) mirror at Mount Palomar near San Diego, telescopes had gotten about as big as they were going to get. In the following 30 years, the advances came from new technology to collect the light hitting the equipment. This was revolution number two. Astronomers stopped using photographic plates, which might capture one-half of 1 percent of the incoming light, and turned to electronic detectors, which can capture more than 90 percent.
In the mid-1980s, astronomy was embarking on revolution number three. New technologies were emerging that allowed astronomers once again to build bigger telescopes--ones like the 10-meter Keck in Hawaii, which dwarfed Mount Palomar’s. Meanwhile, researchers had been developing instruments that could look at multiple objects simultaneously (hence the term multiplexing). A telescope that could measure the redshift from 100 galaxies simultaneously, for instance, would be a 100-fold improvement over a telescope that could look at just one. We don’t get factors of 100 very often for astronomy,’’ says Fabricant.
When the Challenger disaster made Fabricant turn to optical astronomy, the Smithsonian and the cfa were trying to decide what to do with an optical telescope at Mount Hopkins near Tucson, Arizona, known as the Multiple Mirror Telescope, or mmt. The telescope, designed in the 1970s, was composed of six identical mirrors that worked together as though they were one telescope six times as large. It was a remarkable idea because building six modest mirrors was considerably cheaper than building one huge one. But the mmt focused on one object at a time, which pretty much condemned it to obsolescence from the day it opened its shutters to what astronomers call first light.
At the suggestion of colleagues at the University of Arizona, the cfa astronomers thought about rebuilding the mmt as one mirror that could focus clearly on a patch of sky four times the size of the full moon, instead of on a single star or galaxy. No one was sure just what they would do with this new mmt (which they planned to rename the Magnum Mirror Telescope so the initials could remain the same). Fabricant, who was only a young X-ray astronomer, attended meetings, but no one took his ideas very seriously. They were kind of discouraging,’’ says Fabricant. Their attitude was, ‘We don’t really need anyone for this, and if we did, it wouldn’t be you.’
Geller was the exception. The stickman had convinced her that a massive sky survey was needed, as deep into the universe as technology would allow. By 1990 astronomers had mapped a paltry 10,000 galaxies; Geller liked to say that interpreting the structure of the universe from the positions of 10,000 galaxies was like trying to understand the surface of Earth from a map of Rhode Island. The deep redshift survey Geller wanted would be possible only with some new multiplexing instrument, because the deeper into the universe you look, the more galaxies there are in each patch of sky. When Geller and Huchra mapped the nearby universe a few hundred million light-years out and uncovered the stickman, they were lucky to see a single galaxy in every square degree of sky. Look 5 billion light- years out into the universe, however, and you’ll see more than a thousand galaxies in the same amount of sky. That just gives you an idea how big the universe is,’’ says Geller. With the right instrument, a sky survey could map a good number of those thousand galaxies simultaneously, and Geller believed Fabricant, X-ray astronomer or not, had the talent to help build it.
Geller also felt that she now had enough prestige to throw her weight around. Over the next six or seven years, she managed to get Fabricant the funding and permission to build two new instruments for redshift surveys. The first--called the Decaspec because it could look at ten galaxies simultaneously--went on a 2.4-meter telescope on Kitt Peak near Tucson. It worked the very first night, unlike most instruments in the history of astronomy. The second, which went on a smaller telescope at Mount Hopkins, not only worked perfectly the first night but also was the most efficient instrument of its type ever built. Fabricant and Geller used that instrument to perform a galaxy survey twice as deep as the one that revealed the stickman and the Great Wall. Their new survey showed the same inexplicable two-dimensional structures, with still no sign of how and when they formed.
Now Fabricant is building the ultimate redshift collector, the one they will use to map 50,000 galaxies. It will be mounted on the tail end of the converted mmt, which is expected to observe first light in late 1998 with a single, immense, 6.5-meter mirror. Fabricant’s instrument, called the Hectospec, will collect light hitting the mirror from 300 galaxies at a time. (Hecto means 100 in Greek, which is the right order of magnitude and less of a mouthful than Trihectospec.) It will then robotically redistribute its 300 light-catching fibers, one at a time, so that five minutes later it can begin to observe redshifts from another 300 galaxies. And so on through the night, every night Fabricant and Geller can get time on the telescope.
Fabricant and his colleagues are finishing up the Hectospec design in a laboratory in Cambridge across the street from the Harvard- Smithsonian. To understand what Fabricant is building and how it will work, first imagine the telescope itself. The light from the heavens comes down and bounces off the huge mirror and then bounces back off a secondary mirror 6 meters above the main one. The doubly reflected light then comes back down and lands on a surface called the focal plane, which is sometimes covered with photographic film or electronic detecting sensors. In this case, it will be covered with the Hectospec.
The Hectospec’s light collectors are 300 tiny glass fibers, each of which ends in an equally tiny prism that sits in a metal button sticking magnetically to the focal plane. If the button is placed correctly, light from a galaxy will fall on the prism, which will direct it down the fiber, which will run, along with the 299 other such fibers, into a spectrograph-- the instrument that breaks the light into its component colors and makes it possible to measure the redshift.
That’s the relatively simple part. The hard part is positioning those 300 buttons: figuring out how to pick them up, one at a time, and put them down again exactly where the next galaxy’s light will fall, without tangling the glass fibers. To make matters even tougher, all this repositioning has to happen on the telescope, which might be pointed off toward the heavens at who knows what angle and at temperatures ranging from a comfortable 70 degrees Fahrenheit to a very chilly 20, depending on the time of year. Speed counts, too. The faster the fibers are repositioned, the more galaxies can be surveyed, which is crucial because, as already noted, it’s a big universe out there.
The job of moving the buttons goes to a pair of robots, which Geller describes as pretty massive pieces of machinery.’’ If you met them on the street, she adds, you wouldn’t recognize them as robots. Like refugees from some mechanized, futuristic drafting table, they move on a pair of perpendicular rails that allow them to cover the entire focal plane. On the bottom of each robot is a clamp that can close around the button on the end of the fiber. Once that clamp locks on, the robot will reposition the fiber one foot per second, which is about the speed of a violinist’s bow playing andante.
The guiding philosophy behind the Hectospec, says Fabricant, was to build the most ambitious project they could pull off, measuring redshifts from the greatest number of galaxies while still functioning the first day on the telescope. You can run the risk of being not ambitious enough and having something no one finds competitive by the time it’s finished,’’ he says. But if the instrument is temperamental during your assigned nights, you’re out of luck. You have to be awfully sure the thing is going to work.’’ Because of the tricky nature of the job and the Hectospec’s idiosyncratic requirements, Fabricant and his friends couldn’t entice any industrial builder to take on the challenge, so they’re building the instrument themselves. They expect to have it done by late 1997, which gives them plenty of time to make sure it works before the mmt opens its single enormous eye one year later.
At that point, Geller and Fabricant will start measuring their galaxies--300 at a time, as many as 3,000 a night, and tens of thousands a year. While they’re doing it, they’ll be competing with a half- dozen other redshift surveys. Some will map fewer galaxies but get on-line sooner. Some will map more galaxies but won’t go as deep. They’ll all be looking for signs of the beginning of structure, for the time when the universe started to form what Geller calls these beautiful patterns. And this may be why Geller is still in the business. It’s the grandeur and the aesthetics of the problem,’’ she says. I’ve seen these beautiful patterns that the universe makes, and I’d like to know how it does it.’’