91av

When did you become aware of dark matter?

I took a class in the mid-1980s in which astrophysicist Scott Tremaine said it had to exist. The class was sceptical, me included.

How did he explain the need for dark matter?

Newton’s law of gravity says that the stars and gas in a galaxy should orbit the galactic centre ever more slowly with increasing distance. But in the 1970s, the astronomer Vera Rubin and others had found that orbital speeds first increase as you move away from the centre, and then stay more or less constant with increasing distance, giving what’s called a flat rotation curve.

Tremaine showed us a graph of such a curve. It meant that those outer stars were moving too fast, and Newtonian gravity just couldn’t account for it. He showed us how both normal and dark matter were needed to explain that flat curve: the mass of dark matter provided the additional gravitational pull.

Did the idea of dark matter appeal to you?

Not at first, but I became convinced by the data that it had to be so. By the mid-1990s, however, I ran into a problem. It had to do with low-surface-brightness (LSB) galaxies, of which even the brightest examples are dimmer in the sky than the darkest night. I was a postdoc at the University of Cambridge, and I had developed my own theory of how galaxies formed using models involving dark matter. Basically, everything I had learned about these dim galaxies suggested that they were simply stretched-out versions of brighter galaxies. One of my predictions was that groups of LSB galaxies would cluster differently compared with brighter galaxies. That prediction was confirmed. I was impressed with myself.

So where was the problem with dark matter?

My theory failed to correctly predict how fast these dim galaxies were rotating. That was a big blow to my ego. The only way to make them rotate correctly in my theory was to fine-tune the dark matter: the more I stretched out the stars in a given galaxy, the more dark matter I had to throw in to balance things out. To me, this fine-tuning was a problem. You feel like the Wizard of Oz: the man behind the curtain turning knobs and dials. If you don’t look behind the curtain, then yes, you can produce something that sort of looks like the data, but it’s completely unnatural. I pounded my head against that problem for months.

And that struggle soured you to dark matter?

Yes, but only thanks to a chance event. Towards the end of my stay in Cambridge, physicist Mordehai Milgrom came to give a talk on his theory of gravity, called modified Newtonian dynamics (MOND). His idea was that the gravitational force is as Newton prescribed for the most part, but acts slightly differently at very low accelerations due to gravity, so there was no need for dark matter. I almost skipped the presentation because I thought it was crazy talk. Still, I went. Then at one point, in just three lines on the board, Milgrom derived equations that predicted the behaviour of LSB galaxies. And that was exactly what I had been observing and had been unable to explain satisfactorily with dark matter. I was flabbergasted.

So, did you begin working on this theory of modified gravity?

Testing MOND was not my priority, but when I got around to it, my attitude was “My data for LSB galaxies will falsify this stupid theory”. The data told a different story. Then, shortly after I met Vera Rubin while we were both visiting the University of Groningen in the Netherlands, she offered me a fellowship at the Department of Terrestrial Magnetism in Washington DC, where she worked.

What was your reaction to the job offer?

Here was the queen of dark matter offering me a job, when I had suddenly become interested in this radical non-dark-matter theory! So I said, “Look, I’m interested in MOND, is that OK? Do you still want to hire me?” She had a brief look of shock on her face, but, to her credit, she said, “Of course. It’s a fellowship, you can work on whatever you want.”

So that’s when you started working on this radical theory in earnest?

Yes. I spent much of my time at the department going through different kinds of astronomical data to see whether MOND could explain them. People weren’t giving the theory a fair shake. There was a huge dark-matter bias in the field – and it’s still there. That brought out the social justice warrior in me. It turns out that MOND explains many scenarios well, though admittedly not all. It can’t account for the dynamics of clusters of galaxies, for example.

Your new work is raising hackles again. What did you find?

My colleagues and I observed a large sample of different types of galaxies. We found a direct correlation between the distribution of normal matter – the stars and gas – and the rotation curves, even though the curves don’t behave according to Newton’s laws. It’s like a there is a single, universal force law in galaxies that depends on normal matter alone. One that happens to look exactly like MOND.

Could dark matter explain your results?

It seems unnatural to invoke dark matter, because you have to arrange it to be just so, in just the right place, to get this one apparently universal force law to work. Also, the dark-matter theorists have to explain why it is that normal matter alone so gracefully maps on to the motions of galaxies (see “Strangely attractive“).

“It’s like God shouting: ‘There’s more to gravity, not more mass in the universe!’”

To me it’s a big stop sign in the sky. If it’s true that there is one force law in galaxies, it’s like God shouting, “There is something more to the theory of gravity, not something more to the mass of the universe!”.

Does that mean dark matter is dead?

No. While the dynamics of galaxies scream MOND, the dark-matter picture works well at larger scales in cosmology, and MOND doesn’t. It doesn’t seem that we can just pick one. Somehow we must reconcile the irreconcilable.

What would Rubin have said about your find?

She’d say it is early days and we still have a lot to learn. She was an open-minded scientist: always very positive about everything, incredibly curious and just wanting to know how things worked. From her I learned to appreciate the accomplishment of others, even if I didn’t agree with their world view.

When Rubin died in December, some lamented that she had missed out on the Nobel prize…

Vera really didn’t care about that, because she self-identified as an astronomer. She explicitly said to me, “I’m not a physicist”. So, if the prize is for physics, then I think in her own mind, it didn’t apply to her. I wouldn’t agree. Her work had a profound influence on physics. Some say she discovered dark matter, but I don’t think she’d agree with that herself. She discovered that the rotation curves of galaxies were flat. I think she found a law of nature, and that in itself is worthy of a Nobel prize.

This article appeared in print under the headline “From light into dark, and back again”