neutron stars

An artist’s impression shows two tiny but very dense neutron stars at the point at which they merge and explode as a kilonova. (ESO/L. Calçada/M. Kornmesser)

The latest LIGO observations rekindle a fiery debate over how gravity works: Does the universe include dark matter, or doesn’t it?

In October, LIGO and its European counterpart, VIRGO, witnessed gravitational waves rippling out from a breathtaking collision between two neutron stars. This unprecedented event looked like yet another triumph for a new kind of astronomy, one that could use gravitational waves to probe some of the universe’s deepest mysteries. But in all the excitement, most people didn’t notice that something had died: a whole group of theories that posit a universe with no dark matter.

That’s right: a lesser known consequence of the LIGO news is that we may be about to see a significant reshaping of the debate over dark matter—the stuff that seems to be gravitationally attracting the visible material and bending space, but can’t be seen. According to a paper posted to the ArXiv preprint server by Richard Woodard, a professor of physics at the University of Florida, the recent observation nullifies a class of theories that sought to explain gravity’s behavior on galaxy-sized scales without dark matter.

Woodard notes that some of those theories, known as modified gravity (MOG) or modified Newtonian dynamics (MOND), predict that gravitational waves and light waves would arrive at different times. Yet LIGO picked up the gravitational waves and light from two colliding neutron stars within about 2 seconds of each other. Since the source of both was 130 million light years away, that’s a difference of only 1 part in about 1.5 quadrillion. Essentially, they arrived at the same time.

The kinds of models Woodard is talking about—which he calls “dark matter emulators”—attempt to duplicate the effects of dark matter, by assuming that gravity behaves differently than most scientists think. “Our paper definitely does not rule out all modified gravity models that dispense with dark matter,” Woodard clarified. “It applies just to the large class of them.”

Yet while they may have faced a blow, anti-dark matter theorists aren’t going down without a fight.

In Albert Einstein’s theory of general relativity, space is curved by massive objects. A beam of photons—otherwise known as light—travels along the shortest distance between two points (which isn’t always a straight line). General relativity says gravitational waves and light move on the same lines, or metrics.

But while general relativity has been vindicated as of late, it isn’t the last word. Some alternative theories of gravity had gravitational waves moving on a different path, or metric, from light. To get this effect, a modified gravity theory would have to posit that gravitational waves’ paths are affected only by the visible matter we see, whereas light (photons) would be affected by the visible matter and whatever duplicates effects that look like dark matter.

In that scenario, gravitational waves and light would arrive at widely different times. But since LIGO saw both arrive so close to each other, it looks like a powerful piece of evidence that gravity works the way Einstein’s theory says it does—which in turn would bolster the case for dark matter.

However, long before LIGO, some physicists were unsatisfied with dark matter and devised other theories that sought to explain what astronomers see. One set of theories is known as Tensor-vector-scalar gravity (TeVeS), which adds an extra field to gravity. Developed by Jacob Bekenstein in 2004, was already under some fire because it seemed to require neutrinos more massive than what physicists have estimated so far, and it did not always produce stable stars. Scalar-Tensor-Vector-Gravity (STVG) also adds another field, though in a different way from TeVeS. The theory says gravity gets stronger as you scale up from the solar system to galaxies and then to galaxy clusters. It’s those two classes of theories that Woodard says are ruled out by the latest data.

You’d think that physicists would finally accept that dark matter is out there, in whatever form it may be. Right? Well, the proponents of modified gravity say they aren’t done yet.

John Moffat, a researcher at the Perimeter Institute in Waterloo, Canada, says Woodard simply mischaracterized his theory. “They provide no explanation as to why my MOG is falsified,” he said in an email. “It is true that these MOND theories are excluded by the neutron star merger data. Therefore, it appears that my MOG is the only surviving gravity theory that can explain the galaxy, galaxy cluster data and cosmology data without detectable dark matter in the present universe.” Moffat says that his theory in fact does predict that light and gravitational waves would arrive at the same time.

“The best way to interpret this result is not as proving that dark matter is correct, but rather as constraining how modified gravity theories must be constructed if they seek to dispense with it,” Woodard said.

Different Paths

In the 1970s, the late astronomer Vera Rubin, then at the Carnegie Institution, found that that visible matter didn’t move slower as one goes out from the galactic center (the way that planets move slower as one moves away from the sun). At a certain point it was all moving at the same speed. Either there was a lot of diffused mass around the galaxies we couldn’t see, or gravity behaved in ways that weren’t apparent before.

Early explanations for the unseen matter included: gas, rogue planets, neutrinos, and even black holes. Eventually all were discarded in favor of the current conception of dark matter as made of something that only interacted via gravity.

Yet a few physicists felt that the idea of dark matter was too convenient, something invented just to make the mathematics work. Maybe gravity worked differently at different scales, and general relativity simply didn’t account for it, they theorized.

Mordehai Milgrom, an emeritus professor at the Weizmann Institute of Science in Israel, was one of the early MOND theorists, having proposed his version in the 1980s. At its heart, his theory proposes that gravitational dynamics change when accelerations due to gravitational force get below a certain limit. He also posits that gravity and light travel on different metrics.

Taken together, these theories presented, if not a serious threat, at least the intimations of problems with dark matter -- until now.”

Dark Matter FTW

Dark matter didn’t just explain rotation curves. It also accounted for observations of gravitational lensing—the bending of light by massive objects. When we look at some distant galaxies, we see objects behind them as though through a lens, per general relativity. The light is bent by an amount that can’t be explained by the visible mass. This was another piece of evidence for dark matter (or something like it).

Dark matter can also explain why the cosmic microwave background looks the way it does: it’s uniform on average, but at smaller scales it’s clumpy, as one would expect in a dark-matter universe. “One of the things that alternative to dark matter theorists never talk about, is that if you don’t have dark matter you don’t get bumps in the [cosmic microwave background],” says Will Kinney, a professor of physics at the University at Buffalo. “To my knowledge none of the alternative dark matter theories ever had any explanation at all for bumps in (cosmic microwave background) spectrum. That in itself tells me those theories aren’t going to work.”

One good example is the Bullet cluster, a region of space in which two galaxy clusters are colliding. Observations of the cluster show lensing effects that don’t line up with the visible matter in it. Yet if one assumes dark matter is present but hasn’t settled yet around the cluster, then the lensing fits dark matter theory, Kinney said.

The Case For MOND

Even so, the architects of modified gravity counter with the problems that dark matter has. One is an anomaly around the Bullet Cluster—the same one that most would say supports dark matter theory. According to some observations the Bullet Cluster is accelerating too fast; even assuming dark matter the velocities are “wrong.” Also, dark matter predicts the rotation speeds of some galaxies less well than modified gravity.

In addition, some galaxies that appear to have less visible matter still appear more massive. That could be due to a lot of dark matter, but there’s no particular reason that should be the case. MOND theories do better on that score. “MOND has more predictive power. One can use it to predict the kinematics of apparently dark matter dominated galaxies. You cannot make the same prediction with dark matter. All you can say is ‘I bet that low surface brightness galaxy has a lot of dark matter!’” said Stacy McGaugh, an astrophysicist at Case Western Reserve University who has worked on modified gravity theories. “This based on previous experience, not theory, for which there is no agreed prediction.”

Another issue is the distribution of said matter. Milgrom notes that in almost all the galaxies that have been observed so far, the rotation curves are the same shape out to the point where acceleration due to gravity towards the center is about one ten-billionth of a meter per second squared (about the same gravitational force felt by someone two meters away from a 10-kilogram weight).

If dark matter exists, one wouldn’t expect it to always be distributed just so. It would be like going to all the countries on Earth and finding that the income distribution was exactly the same, despite the very different histories that each country has.

“In the [dark matter] paradigm, present-day dynamics are a result of the complicated, cataclysmic, and unknowable history of the individual galaxy under study: on how many mergers it underwent and how violent they were, on the ejection of baryons from the galaxy due to various poorly understood processes, etc.,” he says. MOND theories, he added, do a better job at predicting galaxy motion in that regard.

Even Milgrom, though, acknowledges there are some areas that MOND theories don’t predict as well, even in their relativistic MOG versions – not reproducing the observed cosmic microwave background, for example. “We need an extension of MOND that will account for cosmology. This is something we are working on.”

Sabine Hossenfelder, a research fellow at the Frankfurt Institute for Advanced Studies in Germany, agrees that Woodard’s observation would render some kinds of MOND or MOG obsolete, but also isn’t convinced that dark matter is the answer. “It is almost certainly correct that the observation rules out theories with the assumptions that they list in the paper. But it is unclear which, if any, modified gravity theories actually fulfill the assumptions,” she said. On her blog she noted dark matter works on all scales, while modified gravity doesn’t work as well for cosmology.

Ethan Siegel, an astrophysicst and author, said the odds are that a lot of modified gravity fields are nullified by the LIGO observations. Like Hossenfelder, he believes the problem for MOND is the scales it describes. “Moffat is right: MOND does better than dark matter on galactic scales. If you look at individual galaxies and their dynamical properties, MOND has the advantage. MOND fails on all scales other than that, however.” Some of Milgrom’s theories, he said, might survive – if Milgrom’s contention that gravity obeys different rules than the matter in the universe does is true, for example. “This is a theory that may still survive these gravitational wave results.”

And despite his work on alternatives to gravity, McGaugh said there are things that only dark matter can make sense of. ” I don’t see how to explain the cosmic microwave background or clusters of galaxies (all rich clusters, not just the bullet cluster) without it,” he says. “That doesn’t mean it can’t happen, but at present I see no other viable explanation.” At the same time, he isn’t yet committed to either side. “Neither are convincing,” he says.

In other words, expect the debate to keep raging for the foreseeable future—with the force of two neutron stars colliding.