How old the cosmos?

ASTRONOMERS who study the age of the universe have an embarrassment. They cannot decide whether the cosmos is 10 billion or 20 billion years old. Moreover , the lower of these two ages, which a number of recent analyses favor, would imply that the universe is younger than some of the stars it contains.

Gerand de Vaucouleurs of the University of Texas, who leads the young-universe group, calls the age uncertainty ''intolerable.'' Yet it is the kind of intolerable situation that astronomers have found themselves in before.

Back in the early 1950s, the best estimates of the age of the universe and the age of Earth were both just under 2 billion years. Then geophysicists found they had erred in dating rocks and meteorites. As they improved their methods of estimating age from the decay of radioactive materials, they found Earth and the solar system to be several times older than they had thought - 4.5 billion years old for Earth, by the current estimate.

This left astronomers contemplating a universe which their estimates of size and age suggested was younger than the planet on which they resided.

But this quandary was only a prelude to new discoveries which showed the astronomers had also erred in their estimates. In essence, they had not adequately understood the nature of certain stars used to estimate distance. These stars, called cepheid variables, have a known relationship between their intrinsic brightness and variability which enables astronomers to know how bright they really are. Then, by noting how dim they appear when seen from Earth , astronomers estimate the stars' distances. The new distance estimates, based on a better understanding of these stars, indicated a much larger, and hence older, universe.

By the early 1960s, Allan Sandage of the Mt. Wilson and Palomar Observatories had showed that the universe was at least 10 billion years old. Astronomers felt comfortable again.

Their comfort was short-lived. As understanding of the life cycles of stars progressed, it became evident there should be many stars that are very ancient. Estimates gave ages of 15 to 20 billion years.

Meanwhile, Sandage, working with Swiss astronomer Gustav Tammann of the University of Basel, was refining the cosmic-distance scale. By a decade ago, they were confidently giving the universe itself an age of around 20 billion years. Again the universe was at least as old as the stars within it.

This is the estimate that de Vaucouleurs and some other astronomers have come to consider twice as large as reality. Sandage and Tammann stick by it. But the controversy has highlighted serious problems with the way astronomers estimate cosmic distances and the age of the universe.

What is at stake, basically, is astronomers' grasp of how the universe is expanding. They characterize this expansion by a number they call the Hubble constant, named for its originator, the late Edwin P. Hubble. This relates the speed with which a distant object recedes as part of the universal expansion to its distance from us. Also, by inverting this constant, astronomers can estimate the age of the universe.

Thus Hubble's original number of 560 implied a universe less than 2 billion years old. The current controversy, in effect, is between a Hubble constant of about 50, with an age of 20 billion years, or a constant of around 100, with an age of 10 billion years.

The Hubble number is first estimated by studying the rate of recession and the distances of galaxies relatively near. Then the distance-estimating techniques are extrapolated further out to give an estimate of the number that presumably characterizes the entire universe.

This procedure is beset with uncertainty, for no one can be sure such extrapolation is really valid.

If the younger age were found to be correct, astrophysicists would have to rethink their theories of how stars evolve, for these would make stars older than the universe. If the 20 billion-year age is correct, there will be no challenge to stellar theory. But until the dispute is cleared up, no one can be sure it does not hint at serious flaws in our understanding of stars and of the cosmos.