Faster Than What?
Say you're the editor of a science journal deciding whether studies arriving over the transom deserve to be published -- or forwarded to "The X-Files." And say that this morning's FedEx delivers a paper reporting that a beam of light traveled faster than--how to put this?--faster than the speed of light. Then your e-mail brings a paper describing how a particle of light -- navigating an obstacle course of slits and detectors -- "knows" what lies ahead of it. At quitting time, your fax shrieks with an arriving P.S.: in that last experiment, the authors add, you can change the past.
Fodder for "The X-Files"? Not in today's physics. Physics parted company with common sense in the 1920s, when such giants of modern science as Niels Bohr, Max Born, Erwin Schrodinger and Werner Heisenberg discovered quantum mechanics. This science of subatomic particles took a wrecking ball to all sorts of ideas we take for granted, such as the belief that particles have both a definite location and a definite momentum. This is the famous Heisenberg Uncertainty Principle. Worse, quantum mechanics allows what Einstein called "spooky action at a distance": by the mere act of observing something here you affect something there, Several papers at a conference last week at the University of Rochester focused on this spookiness, and a new book by English astrophysicist John Gribbin -- "Schrodinger's Kittens and the Search for Reality" (261 pages. Little, Brown. $23.95)--explores it with both whimsy and scientific rigor. In a paper to be published later this year in the journal Quantum Optics, physicists led by Raymond Y. Chiao of the University of California, Berkeley, analyze experiments that show what a long, strange trip quantum mechanics is still on. "It's completely counterintuitive and outside our [everyday] experience," says Chiao. "But we [physicists] have kind of gotten used to it."
Though not easily. In one experiment, Chiao, Aephraim Steinberg and Paul Kwiat seemingly got light to travel faster than the speed of light (186,000 miles per second, give or take). The Berkeley team sent particles of light--photons--toward a detector. Half the photons sped along un-impeded. The others headed toward a special multilayer mirror; of these, 99 percent bounced off and out of the experiment. The surviving 1 percent tunneled through the mirror, heading straight for the finish line where photon detectors waited. They didn't have to wait long. "The tunneling photons arrive, on average, before those that traveled through the air," says Chiao. So much before, in fact, that the light penetrating the mirror had an apparent velocity 70 percent faster than the speed of light.
Faster than the speed of light is not possible, says every reputable physicist from Einstein to your high-school science teacher. So Chiao's team calls its re-suit an illusion. How was the illusion created? Something odd, they say, must happen to the photon inside the mirror. According to the Uncertainty Principle, nothing is exactly here. Everything, including a photon, has a chance of being slightly ahead of, or behind, "here." Somehow, going through the mirror shifts forward where the photon is likely to be found. Unfortunately, "no one yet has any physical explanation" for this shift, they admit. Steinberg, now at the National Institute of Standards and Technology, admits that the explanation is convoluted. Critics, he says, call the refusal to acknowledge faster-than-light travel the mark of "imbeciles refusing to believe common sense."
'Mind-boggling effects': Until the 1990s most studies of quantum weirdness were "thought experiments," in which scientists used pure reason to deduce what would happen if they ran an experiment. That is never as satisfying as the real thing. But ingenious advances in light detectors and emitters, including an almost magical crystal that absorbs one photon and emits two (low-er-energy) photons, are turning thought experiments into real ones. "The field had basically been dormant since the [1940s]," says physicist Daniel Greenberger of the City College of New York. "Now we have these new devices that bring out all sorts of mind-boggling effects."
In one mind-boggler, a particle of light seems to "know" what experimenters have in store during a "double slit" experiment. A little background: double-slit experiments have a long pedigree. In the 19th century, Thomas Young fired light beams at a two-slit screen, and produced on a second screen an "interference pattern" of alternating dark and light bands (diagram). The bands are the result of light rays going through both slits and interfering with each other, like the waves of water from pebbles dropped into a pond. The zebra pattern disappears, though, if the experimenter sends light through the slits one photon at a time, and then does something to identify which slit any particular light particle goes through. Having this knowledge means that the light particle cannot have gone through both slits, a prerequisite for the stripes. Nobel-winning physicist Richard Feynman called this the "central mystery" of quantum mechanics, that something as intangible as knowledge changes something as concrete as a pattern on a screen.
Now researchers have pushed this craziness to the limit. The eminent physicist John Archibald Wheeler, Feynman's thesis supervisor, suggested one wily version of the double-slit experiment. Researchers place detectors between the screen and the two slits. The detectors monitor the photons, revealing whether each photon passed through one slit or another. Just as in the experiment that confounded Feynman, knowing about a photon's behavior at the two slits makes the interference pattern vanish. But by switching off the detector, Wheeler points out, the zebra stripes return.
Changed behavior: But wait. Whether or not photons create the zebra pattern is determined by their behavior at the double-slit screen. How can what photons do at the slits be affected by a detector that they encounter after the slits? How, when a photon reaches the slits, does it "know" how to behave in order to match the presence or absence of the detector behind the slits? Even if experimenters slip in the detector after the photon has passed the slits, the photon still "knows" what to do. Its past behavior--at the slits--is made to conform to whether or not it "should" make the zebra pattern. In versions of this experiment carried out recently at the University of Munich and at the University of Maryland, physicists sent single photons toward a screen with two slits. The behavior of the photons, writes Gribbin, "is changed by how we are going to look at them." One implication of all this, admits Chiao, "is that you can affect the past."
The resurgence in quantum experiments is fueled by the hope that, if physicists push experiments to the limit, they may find a deeper, better theory than one that says scientists can affect reality by making an observation. This was Einstein's hope until his dying day. Chiao, whose students describe him as "a concert-quality pianist to within experimental error," points out that "all the experiments we do show that quantum mechanics is correct. But who knows? As a scientist, you must hold all theories provisionally and be open to new data showing that a theory is wrong." For now, though, quantum mechanics is the only game in town. Nonsensical, counterintuitive, crazy--sure. But as Henry Kissinger has said about less abstruse matters, "It has the added virtue of being true."
We have known that . . .
Particles of light, called photons, leave a source . . . and pass through two slits in a screen interfere with each other . . . and form bands of light and dark on screen.
Then we found that . . .
When photons are emitted one at a time . . . and a detector monitors which slit each photon passes through . . . the photons don't form the stripped pattern, instead making two bright spots.
Somehow knowledge of the paths of the photons through the slits affects their behavior. It's as if they know they're being watched.
Now we discover . . .
When photons are emitted one at a time . . . and a detector monitors which slit each photon passes through . . . and another device erases this knowledge . . . the striped pattern returns.
Losing knowledge of the photon's path, even after they have passes through the slits, bring back the stripes. It's as if the past has changed. No wonder Einstein was confused.