"Astronomers. . . are used to the impossible," says Guy Pooley of Britain's Mullard Radio Astronomy Observatory at the University of Cambridge. That, he explains, is why they didn't dismiss out of hand the intriguing reports that some cosmic radio sources seem to be expanding at speeds faster than that of light.
Since such reports began coming in a decade ago, they have seemed to contradict a pillar of modern physics -- the principle that neither matter nor any signal can move faster than light. Either this principle is under challenge , the observations are faulty, or something interesting is going on far out in the cosmos which creates the illusion of what the physicist calls superluminal speed.
As Dr. Pooley notes in reviewing the subject recently, astronomers, being accustomed to the seemingly impossible, are neither eager to abandon their fundamental physical principles nor quick to discount careful observations of trusted colleagues. Their inclination has been to withhold judgment and investigate the maverick radio sources as thoroughly as possible.
This is not easy. The sources are billions of light years away. No optical telescope has the resolving power to see the detail needed to explore their mystery. No single radio observatory can do it either. However, by cooperating with one or more other observatories thousands of kilometers away, observations can be made of extremely fine detail.
Now, at last, one set of such observations has produced the detail needed to provide what the research team calls "the first direct and unambiguous evidence of superluminal expansion in any radio source." It was the publication of some of this evidence in Nature recently that was the occasion for Pooley's review.
This sharpens the kinds of questions that astronomers now want to pursue with these sources. It better defines possible explanations of their illusory superlight speeds. And it raises the hope that the mystery may itself be resolved within this decade.
So far, four of the puzzling sources have been identified. One is a galaxy known only as 3C120. Three are what are called quasars, known respectively as 3 C279, 3C345, and 3C273. Quasars are powerful, often violently eruptive, objects billions of light years distant. They are so compact they appear no larger than faint star images in sky surveys. Yet, intrinsically, they outshine a whole galaxy of stars, such as the Milky Way galaxy in which we live.
To resolve the detail needed for these studies, radio astronomers need very large instruments indeed. As a rough rule of thumb, resolving power increases with the size of the antenna dish, if one thinks in terms of the traditional radio telescope. Astronomers literally synthesize antennas with the effective size of a continent or larger using a technique that is simple in principle but that has been given a long and formidable name -- Very Long Baseline Interferometry (VLBI).
To make VLBI work, two or more radio telescopes thousands of kilometers apart track a radio source simultaneously. The radio emissions they pick up are recorded on magnetic tape along with precise time "hacks" from atomic clocks to provide synchronization. Then the tapes are fed into a computer, again with precise time synchronization, and their recorded "signals" interact with each other. Technically, they are said to interfere with each other, hence the name of the technique.
With this kind of data processing, radio astronomers gain the effective resolving power of a single radio telescope with an antenna dish as wide as the distance between the two observing stations. In the superluminal studies, they have been resolving details a few light years across at distances of several billion light years. That is roughly comparable to distinguishing an object a third of a meter across at the distance of the moon.
Because, radio astronomers do not, in fact, have a dish antenna that large, their data are incomplete. The radio source, maps they produce are ill-defined. This has muddied the study of the faster-than-light sources in the past. But over the last few years California Institute of Technology radio astronomers have cooperated with several other observatories in the United States and West Germany to use several different baselines at once. This has given the detail needed to yield the "unambiguous" evidence for faster-than-light expansion that now has been reported by T. J. Pearson, S. C. Unmin, M. H. Cohen, R. P. Linfield , A. C. S. Readhead, G. A. Seielstad, R. S. Simon, and R. C. Walker.
Working with the quasar 3C273, they have produced maps of its structure for a period from mid-1977 to mid-1980. These show an expansion with an apparent velocity of 9.6 times the speed of light, give or take 5 percent.
Seen in optical photos, 3C273 is an object some 2.5 billion light years away with a prominent jet, a million light years long, protuding to one side. It is also a powerful object at radio wavelengths with most of the emission coming from two locations. One is a compact object, known as 3C273B, coincident with the visible quasar. The other is more diffuse and near the end of the jet.
The newly published maps include only the 3C273B object and a small region more or less in the direction of the jet where a smaller radio source is moving rapidly away from 3C273B. This marks the seemingly faster-than-light expansion.
As the Caltech team notes, the cosmic speed limit is probably not being violated. Earthbound instruments may be "looking" more or less straight down a jet along which objects are emitted with speeds approaching that of light. Standard calculations with the theory of relativity, which must be used when objects move this fast, show that such an arrangement could give the appearance of superluminal expansion.
Although this is a possible explanation, it is not yet established. Other explanations for faster than light speeds include the flashlight effect. If you shine a light on a wall and flick your wrist, the light spot on the wall can move at speeds faster than light. But nothing physical -- no object or signal -- has moved that fast. There is also the Christmas tree effect in which lights flashing randomly can create the illusion of something moving faster than light.
As Pooley points out, these explanations don't seem to be holding up as more details of the puzzling sources are discovered.
Among other things, an adequate explanation must account for the fact that the superluminal sources seem always to expand. None have been found so far that are contracting. Indeed, Pooley says, "The evidence is that rapid expansion is a very common feature among compact radio sources." This argues in favor of some kind of expulsion of material from a central source as seems to be happening with 3C273B.
What is needed now, he says, is a program of systematic mapping of these sources to build an adequate data base for theorizing. In fact, the Caltech team says that the kind of detailed mapping they are doing probably is the best way to study the expansion that seems to be going on.
This will put a premium on long-distance cooperation among radio astronomers. And budget cutbacks or not, it increases astronomers' desire to build a giant VLBI network in the United States. In putting out its own announcement of the Caltech mapping, which it supports, the National Science Foundation cited a Caltech study that "recommended that the US build a nationwide array of radio telescopes, stretching from Massachusetts to Hawaii and from Texas to Alaska."
It remains to be seen whether or not the "impossibles" raised by national budget priorities are more intractable than those presented by the universe.