It was my journalistic good fortune in the late summer of 1961 (my first week in Washington) to be ushered into the Oval Office to take part in a colleague's interview of President Kennedy. In addition to my paralyzing fright, I remember best his interrupting the proceedings briefly to take a call and them sharing with us the news that the Russians were summarily breaking the U.S.-Soviet moratorium on nuclear-weapons tests. The call had been confirmation of reports he had earlier received. Kennedy then spoke of the response he would probably have to order -- a resumption of American tests -- which he subsequently did.

All this came wafting back to intrude on my nostalgia as I was thinking about my two decades in Washington the other day. Since that August afternoon we have seen 20 frenzied years of international effort to (1) test, develop and deploy ever more sophisticated nuclear weapons and (2) negotiate agreements limiting their numbers and more menacing characteristics and the likelihood of their being used.

How are we doing? Better, it seems, at the first than at the second. For 20 years and many arms-control agreements later, the United States and the Soviet Union have put in place thousands of new nuclear explosives that can reach each other's heartland; additional nations have acquired nuclear weapons, and the world is sitting around debating whether Iraq was three months away from a weapon, or five years, when Israel destroyed part of its "peaceful" nuclear installation.

To many people the obvous conclusion is that the treaty-seeking part of the process has not been pursued with equal vigor. that may be true, but I think the relationship between our bomb-producing and bomb-controlling impulses is more complicated that that. Everyone recognizes the difficulty of adjusting the pace of political and diplomatic activity to that of technological change, and this is part of the problem. When the Russians, for example, moved on Prague in 1968 it became politically and diplomatically impossible for him to hold back progress on the so-called MIRV technology, which took nuclear arms to a feasible to the Nixon administration to engage in arms talks, the MIRV-limiting opportunity was pretty much gone.

There is another dilemma: although formal legalized commitments to forswear certain weapons and certain conduct may be required to achieve arms control, the accords themselves and the process whereby they are reached have, in some cases, the effect of increasing levels of armament, not decreasing them.

The thought may strike you as radical and heretical. It is not put forward by way of concluding that all arms control (for example, the ABM treaty) has the perverse effect of building bombs or that treaty making should be abandoned. My idea is merely that at a moment when there is much agitation about re-entering arms negotiations with the Russians and in Europe and about the efficacy of the nuclear nonproliferation treaty, we could do worse than to reflect on the relationship, with a view to entering the next few round with a keener understanding of the risks.

These are not, in my view, the risks you hear so much of from the friends and enemies of arms control. In the friendly estimate it is the "process" of working out agreements (the "SALT process" has almost been canonized) that brings great benefit. Yet I think it is demonstrable that the "SALT process" has contributed to an increase in some of the most destablizing and even slightly nutty weapons systems and proposals.

You begin (SALT I) with a reasonable effort to limit those things that can be limited, which is to say seen and counted -- i.e., launches. You then have a race to see how much (before the next accord) you can pile onto each launcher in the way of multiple warheads, etc. You end up with weapons so loaded for bear that you need to protect them at all costs from attack -- ergo, MX, the monster, movable (fugitive?) missile that no one wants hidden in his basement.

This impetus that the arms talks may give to arms development is what is often wrong with the complaint of the other side, too, those critics who believe that it is the likelihood of the other fellow's cheating that constitutes the danger of arms accords and arrangements to reach them. Yes, there is cheating. But I believe it is far easier to argue that the Russians, for instance, merely lived up to the maximum reading of the accords we have reached with them than that they went off and systematically and grossly violated them.

Not always, but often, so-called "ceilings" in these accords become "floors." Politically; at home, it may be necessary to go for everything allowed, whether you need it or not, under the treaty's terms (in SALT II, it was even lamented that we were prohibited from acquiring certain types of missiles we didn't want). Sometimes it is the simple binge-before-the-diet instinct that comes into play, causing each side to stock up in a big way in the year or two that may remain before a scheduled limitation goes into effect. You don't have to be for or against any particular level of armed preparedness on our part to observe that this is a hell of a way to run a railroad -- to make our weapons choices.

And internationally the same is true continuance of the allegedly harmless nuclear technology and material between nuclear haves and have-nots was a condition of getting the nonproliferation treaty accepted in the Third World. That misguided traffic should have been much more severely limited and regulated years ago. Maybe Colonel Kaddafi will make the point -- with a bang -- one of these days.

People will say we would have been much worse off without these agreements. I say we aren't so well off with them, and that in some measures, anyway, they have rationalized and incited increased arms production. My tentatively offered point is that we need to look at the agreements we have made and are going to embark on in a much more intelligent, critical and unsentimental way.