Let's try again. Some weeks ago, I wrote a column touting an idea of Lt. Gen. Dan Graham (ret.), former head of the Defense Intelligence Agency. Basically, it was that the United States should "put into space a non-nuclear system that could shoot down Soviet nuclear weapons over Soviet territory"--an approach that, according to Graham, doesn't even require new technology. Graham's system (he calls it "High Frontier") would be useless as an offense against unlaunched missiles, because its high-velocity projectiles would be burned up in the friction of the Earth's atmosphere. But since it would be deadly against missiles in space, it would be a perfect defense against a Soviet nuclear attack and (I thought) a major step toward military stability.

A few readers, among them Jeremy Stone of the Federation of American Scientists, saw the flaw immediately. If we could neutralize Soviet-launched missiles, the Soviets would be left with no defense against our own. Just think how we would react on learning that the Soviets could shoot down our missiles over the U.S. mainland while retaining the deadliness of their own missiles. It would be, said Stone, "the most destablizing development imaginable."

I wrote a second column apologizing for the oversight. Here we go again. At least three readers responded with the same general suggestion. Here is how Jack C. Saichek of Rockford, Ill., put it: "You endorsed the (Graham) plan until Jeremy Stone pointed out that the Russians would be intimidated by the knowledge that we were developing such a system which would incapacitate them but not us. They might panic and launch a pre-emptory attack.

"The ideal solution to the dilemma would be to get the Russians to work jointly with us on a system that would destroy any and all missiles launched, including our own. Such a plan would have at least three salutary effects: 1) It should allay Soviet fears of being at the above-mentioned disadvantage. 2) It should allay Soviet and American fears that some other power might launch a nuclear attack. 3) It should, therefore, end the missile-building race. . . . Recent history predicts that it will be hard to get the Russians to agree with us on terms for such a joint project. In this case, I suggest that we announce that we are going ahead with the project on our own, but that Russia is welcome to send observers or participants. I'm sure they will decide that this is an offer they can't refuse, since it will allow them to assure themselves that the system will really shoot down our missiles as well as theirs."

No doubt there is some fatal flaw in Saichek's modification of the Graham scheme, but I haven't spotted it yet. I find it hard to imagine that we and the Soviets could so easily end our arms race. Still, the High Frontier concept (as modified here) might give us and the Russians something other than throw-weights, megatonage, warheads and MIRVs to talk about while we grow up enough to talk about genuine arms reduction. And it just might leave both sides with enough leftover resources to feed their hungry, educate their ignorant and get their economies going again.