Physics & Math / Dark Matter

Two spiral galaxies twist and pull at each other. Astronomers are still debating how gravity mediates such encounters.

Mordehai Milgrom never wanted to be a heretic. Twenty-five years ago, while poking around for a meaty research problem, he found one that changed the course of his career—and that might yet transform our most fundamental understanding of the universe. His ideas, long relegated to the fringes of physics, where all but cranks fear to tread, have finally become too intriguing for his mainstream colleagues to ignore. Milgrom's heresy? He denies the existence of dark matter, the shadowy and thoroughly hypothetical stuff generally held to make up 80 percent or more of all matter in the universe. Even though dark matter has eluded all attempts at detection, most cosmologists are convinced it must be out there. Without it, there's no explanation for much of what they see in the cosmos.

Or at least there hadn't been until Milgrom's break with orthodoxy. His alternative not only eliminates dark matter, it strikes at the heart of modern physics. In short, Milgrom thinks that Isaac Newton's laws of gravity are incomplete. Unlike many radical alternatives to conventional physics, Milgrom's brainchild—known as modified Newtonian dynamics, or MOND—has not withered under scrutiny. Attacked? Yes. Ridiculed? Certainly. Refuted? No.

A little more than a year ago, Milgrom, a professor of physics at the Weizmann Institute in Rehovot, Israel, gained new support for his ideas when his longtime collaborator, Jacob Bekenstein, published a new, more powerful version of the theory, one fully consistent with Einstein's general theory of relativity. With this advance, MOND is poised to go head-to-head with dark-matter theories in describing how galaxies form and evolve. Should MOND prove successful, thousands of papers in mainstream cosmology will become obsolete overnight. And that is just half the story. If Milgrom can claim victory, he will have wrought the most dramatic revision of our understanding of gravity since Einstein's work of almost a century ago.

Mordehai Milgrom is a physicist who has boldly updated Newtonian motion and, in doing so, has reconceived Einsteinian gravity.

Mordehai Milgrom began his career studying objects called ultracompact neutron stars in binary star systems. In 1979, sensing the need for a new challenge, he bundled his family and headed on sabbatical to Princeton University, one of the world's leading centers for the study of galaxies. Even after half a century of unraveling the structure and evolution of galaxies, astronomers still had much to learn about them. Milgrom became particularly intrigued by one intractable problem. "I had heard there was this trouble understanding the so-called galactic rotation curves, which describe the way stars rotate around the centers of galaxies," he says. "I thought I would apply myself and try to think about this problem."

No one knew it at the time, but the problem posed a challenge to a cornerstone of physics, the fundamental relationship between mass and gravity. Isaac Newton figured out the rules for weighing orbiting bodies close to four centuries ago. The simplicity and accuracy of Newton's laws let any undergraduate transform observations of a satellite's orbit into a direct measurement of the parent body's mass. It is Newton's laws that tell us the sun weighs a thousand trillion trillion tons.

Throughout the 1970s, astronomers applied these laws to galaxies, hoping to extract a measure of their total mass. The orbital speed of stars circling a galaxy can be teased from an analysis of their combined light. Repeating this process for a sequence of positions from the center of the galaxy out to its visible edge allowed astronomers to determine rotation speeds at various distances. Placing these points on a graph produced the galactic rotation curves that Milgrom had heard about. The data were new and messy, but by the end of the disco decade it was clear that something was terribly wrong.

The skinny black line on a plot of stellar rotation speed versus distance was expected to go down—stars close to the galactic center should orbit faster than stars at the edge because all the mass concentrated at the center of the galaxy pulls most powerfully on the closest stars. The same thing happens in the solar system: Mars moves faster than Jupiter because the sun's gravity pulls harder on it, Jupiter orbits faster than Saturn, and so on, out to Pluto and beyond. A plot of orbital speeds and distance—a rotation curve of the solar system—does decrease with distance. The skinny black line falls, just as Newton's laws say it should.

The rotation curves for spiral galaxies do not. At a certain distance from the galactic center, the rotation curves for stars in most every spiral galaxy simply do not fall; instead, at some point they flatten. All the stars in the middle and outer parts of these galaxies orbit with the same speed, in seeming defiance of Newton's laws. Why don't the outer stars move more slowly than the inner ones?