If I told you that five-sixths of the universe’s matter is invisible and undetectable by any method known to science, you’d probably think I was messing with you—or just crazy. But that’s what most physicists and astronomers believe. They have good reasons, too. The stars and gas that astronomers can see with their telescopes don’t produce nearly enough gravity to hold galaxies together, according to well-established laws of physics. Some sort of “dark matter” seems to be serving as galactic glue. Dark matter also explains why galaxies clump together, and its fingerprints seem to show up in light that reaches us from the early universe.
Like every astronomer of his generation, Professor Stacy McGaugh came of age believing in dark matter. But the more he explored how galaxies behave, the more his data seemed to conflict with what he had learned. His quest has made him a reluctant heretic: He’s spent two decades trying to convince his colleagues to question one of their bedrock beliefs. McGaugh’s efforts have earned him attacks and enemies, but also respect and some measure of fame. And they may have put him in a position to help point an entire scientific field in a new direction.
If his doubts about dark matter were to prove correct, the consequences would be revolutionary, says Professor Chris Mihos, one of McGaugh’s colleagues in the Department of Astronomy. “I can’t even begin to say what kind of a shift that would be in our understanding of the universe.”
All the stars visible to the naked eye, even on a dark night in the middle of the desert, live in one galaxy—our own, the Milky Way. Through a telescope, things far more distant start to come into view. For centuries, astronomers debated what these objects were; some thought they might be massive clouds of gas. But in the 1920s, newly powerful telescopes revealed that, in fact, they too were galaxies, as massive and star-filled as our own—and that the universe teems with them. We now know that we share the universe with up to a trillion other galaxies, and that they began forming soon after the universe came into being. Still, there is much to learn about how our galactic neighbors are born, how they live and how they die.
McGaugh got hooked on galaxies almost by accident, in the mid-1980s. He was studying physics and astronomy as an undergraduate at MIT, and for his senior project, he joined a research team studying galaxies that had been discovered through their X-ray emissions. To try to see these objects in visible light, McGaugh traveled to Arizona’s Kitt Peak National Observatory. He found both the place and the science captivating. “You had to stay up all night to run the telescope, but you had plenty of opportunity to step outside to see the stars, which is an amazing thing in a dark place,” he says. “It’s like having special effects in your face—you can’t believe how much is out there. I found compelling the beauty of it and the way in which the science was done.”
McGaugh briefly veered toward lab physics during a year of graduate school at Princeton, but quickly realized he really belonged behind a telescope. So he transferred to the University of Michigan to gain access to the telescopes of MDM Observatory in Arizona. McGaugh wanted to study galaxies that had fewer stars and were much fainter than the Milky Way and other familiar, bright, star-filled spirals—fainter, even, than the night sky itself. At that time, most astronomers thought these “low-surface-brightness galaxies” were rare members of the galactic zoo, hardly worth worrying about.
McGaugh suspected they might actually be much more common and important. To test this hypothesis, he took advantage of a recent invention called the charge-coupled device, or CCD, which can detect tiny electric currents created when single particles of light hit a detector. The devices are now so ordinary that they’re packed into digital cameras by the millions. But during the 1980s, they were the hot new thing, and they made telescopes more than 10 times more efficient at gathering light. Suddenly, faint objects came into much sharper focus.
McGaugh‘s work showed that astronomers had vastly underestimated how abundant low-surface-brightness galaxies were relative to bright galaxies; the prevailing belief was that most galaxies were roughly equal in surface brightness. “That had misled the field for 20 years,” McGaugh says. He calls it his first eureka moment.
McGaugh’s discovery “pushed forward the correct idea that bright, beautiful galaxies were just one piece of the puzzle,” Mihos says.
After completing his studies at Michigan, McGaugh took a postdoctoral position at the University of Cambridge in England. There he did a study that confounded him. Two astronomers had previously demonstrated a simple mathematical relationship between a galaxy’s mass and its rotation speed. (Just as the planets in our solar system revolve around the sun, individual stars in a galaxy revolve around the galactic center.) The famous law of gravity worked out by Isaac Newton, supposedly after seeing an apple fall from a tree in his family’s orchard, would predict that for dim galaxies, whose mass is spread more thinly than in bright galaxies, the same relationship would not apply. But in an analysis of radio telescope data from dim galaxies, McGaugh found that it did.
“That sent me into a great depression for about half a year,” he says. Newton’s law—that the gravitational force between two objects depends on the product of their masses divided by the square of the distance between them—was simple and conceptually intuitive, and had held up for three centuries. It’s taught in first-year physics classes around the world. That it could be wrong was virtually inconceivable.
Then McGaugh had a second eureka moment. It occurred after he heard a talk by Mordehai Milgrom, an Israeli astrophysicist. Milgrom was dealing with a different problem about rotating galaxies—that they don’t have enough visible matter to keep their outermost stars rotating at the speeds astronomers had observed. To solve this problem, astronomers starting in the 1930s had proposed the idea of dark matter—some substance that exerts gravitational force but does not emit or absorb light. Dark matter, it turned out, also did a good job of explaining the makeup of the radiation that reaches us from the universe’s infancy—the so-called cosmic microwave background—as well as the arrangement of galaxies into clusters.
But Milgrom had a different idea. What if, instead of positing some strange, seemingly undetectable matter, you tweaked Newton’s law of gravity at galactic scales? That could solve the galaxy rotation problem. Moreover, Milgrom claimed, this approach would yield predictions that could be tested. And the best galaxies for testing his hypothesis, which he called modified Newtonian dynamics, or MOND, were the low-surface-brightness ones—McGaugh’s specialty.
McGaugh found the idea of modifying Newton’s law and throwing out dark matter laughable. Expecting to prove Milgrom wrong and move on, he did the tests Milgrom suggested. “My reaction was, ‘Great, now I have the data to disprove this stupid theory,’” McGaugh says. “That is literally what I thought.” But when he looked at his results, he found that all of Milgrom’s predictions had come true. “This provoked a crisis of faith for me,” he recalls.
Imagine trying to unlearn that the Earth revolves around the sun, or that dinosaurs once existed. That’s the level of cognitive dissonance McGaugh faced. “I believed in dark matter very much. It was really hard to wrap my head around the possibility that it could be wrong,” he says. “It kind of drives you crazy.”
He was still processing it all when he was offered a second postdoc, this time at the Carnegie Institute in Washington, D.C., with famed astronomer Vera Rubin. Rubin and a colleague had in the late 1960s and early 1970s observed the inexplicably large speeds of stars at the outer edges of rotating galaxies. The study was considered the first indisputable evidence for dark matter, and many observers think it should have earned Rubin the Nobel Prize. (Rubin died in 2016, and the prize is never awarded posthumously.)
Naturally, it was a bit awkward for McGaugh to tell the “queen of dark matter” that he thought her discovery pointed to something else instead. Come work with me anyway, she said. The two followed up McGaugh’s earlier radio-wave study with a more detailed analysis of visible light from dim galaxies. It confirmed the original findings.
McGaugh’s career later took him to the University of Maryland, College Park, and then to Case Western Reserve, where he has been chair of the astronomy department since 2015. His skill at bringing high-quality data to bear on his discipline’s biggest problems has earned him admirers. “Stacy may be one of the last ‘core’ old-fashioned astronomers,” says James Schombert of the University of Oregon, who has collaborated with McGaugh since the 1990s. “He has a keen sense of where you can make progress given the technology and data available to us, and where you are wasting your time trying to solve a problem that simply is not doable at present.”
Whether the observations of rotating galaxies point to dark matter or MOND is “a really emotional issue,” McGaugh says. “I’ve seen conferences dissolve into shouting about it. I’ve never observed wisdom to emerge from shouting.” Though his support for MOND has drawn the ire of many advocates of dark matter, he has also pointed out some shortcomings of MOND, particularly when applied to clusters of galaxies, thus provoking ire from the opposite side.
The theory of dark matter has other challenges besides MOND. Because dark matter interacts so little with light, it cannot be made of electrons, protons or any other known elementary particles. Physicists have for several decades sought new particles that could provide dark matter with an identity. In the mid-2000s, many physicists felt all but certain that relatively massive dark matter particles would appear at the Large Hadron Collider in Geneva, Switzerland. But since the machine was turned on in 2010, it has delivered only one new particle—the Higgs boson. Other experiments designed specifically to detect dark matter have also come up empty.
The null results have disappointed physicists, but for the most part have not shaken their faith. “There is no push within the physics community to say dark matter needs an overhaul,” says Leslie Rosenberg, a physicist at the University of Washington who leads an experiment to detect tiny hypothesized dark matter particles called axions, though he adds that “it’s a good idea to consider alternatives.” And few astronomers are ready to abandon a hypothesis that has answered multiple independent questions in what most experts consider a satisfactory way.
A small but growing contingent, however, is becoming more open to alternative ideas, and McGaugh has become one of its public leaders. He appears regularly in the media, and he “rants” (his word) on his blog about scientific intransigence. At the same time, he has continued to gather evidence that something is amiss in our understanding of galaxies.
Several years ago, McGaugh and Schombert decided to track the motions of stars in 153 galaxies using infrared radiation—light with wavelengths slightly longer than those of visible light. Their study included bright galaxies like the Milky Way, low-surface-brightness galaxies and everything in between. The data, when analyzed by Federico Lelli, who worked with McGaugh at CWRU as a postdoctoral fellow and is now at the European Southern Observatory, showed that any star’s radial acceleration—how fast its angular speed changes—can be predicted by an equation involving its galaxy’s visible matter distribution alone. “This is exactly what is expected in MOND,” McGaugh says.
Published in 2016 in Physics Review Letters, one of the world’s top physics journals, the paper by McGaugh’s team drew wide attention. Two groups of scientists have already attempted to show how the observational data could be consistent with the standard dark matter paradigm. But McGaugh says their efforts require unconvincing conceptual contortions. In most theoretical models, he points out, dark matter is distributed in a spherical “halo” throughout and around a galaxy, so it should not give rise to the same relation as MOND.
Others are intrigued by McGaugh’s results but aren’t sure how they fit into the bigger picture. “It’s an impressive demonstration of something, but I don’t know what that something is,” Oxford University astrophysicist James Binney told a reporter from Physics World.
Even though a clear picture has yet to emerge, McGaugh “is doing important work that will help the field to make progress,” says Sabine Hossenfelder, a physicist at the Frankfurt Institute for Advanced Studies in Germany. “If you ask a theorist like me which theory is the best, everyone will tell you, ‘Well, my theory is the best.’ That’s why it’s important to have people like Stacy who don’t have a favorite theory, but who can just look at the data and say, ‘This one fits, this one doesn’t fit.’ And not everyone will like what you tell them.”
McGaugh confesses that he himself remains uncomfortable with MOND. Modifying Newton’s law of gravity destroys some of the beautifully simple qualities that have made it a favorite of physicists for centuries. Moreover, he agrees with most astronomers that dark matter best explains the cosmic microwave background. Yet his findings have forced McGaugh to conclude that MOND makes accurate predictions about how galaxies actually behave. Perhaps, he says, MOND and dark matter will someday prove to be partial glimpses of a complete theory of gravity that no one has yet had the vision to see: “It seems as if we’re missing something deeper.”
Gabriel Popkin is a freelance science and environmental writer in Mount Rainier, Md.