A couple of weeks ago in this space I complained that no one in authority was troubling to explain why this country needed more nuclear weapons when we were already capable of inflicting so much damage on anyone who started anything with us. Since then, in a somber and straightforward speech, Secretary of State Alexander Haig has addressed the question. We need to maintain a level of nuclear armaments and capacities, he said, that will disabuse Soviet leaders of any notion they might have that some advantage could accrue to them as a consequence of nuclear war. Deterrence, he said, "depends upon our capability, even after suffering a massive nuclear blow, to prevent an aggressor from securing a military advantage and prevailing in a conflict. Only if we maintain such a capability can we deter such a blow." So we must never let the Soviets get in a position relative to us that would permit them even remotely to entertain a notion that they might be able to "win."

OK. But where is this train going? Is it ever going to stop? And doesn't every extra mile and every new load of weapons we take on increase the chance that something will go wrong? It is the open-ended quality of the argument Haig espouses--an argument, incidentally, that pretty much represents the traditional American policy over the years--that is its main defect. What are we headed for? An MX missile in every garage? An SS18 in every dacha?

Haig talks about arms control agreements and the need to be strong enough to push the Russians into accepting sound deals. But people who have seen the vast increase in the number and lethality of nuclear weapons that seems to have accompanied every new U.S.-Soviet arms- control agreement will not be overwhelmed by the prospects for reversing the trend by means of such deals.

You would think this position would be highly vulnerable to attack on grounds of its extravagance, its evident dangers, its costs and its murky destination--and it is. The trouble is that the critique of the position has its own huge built-in flaw, thanks to the predicament we and the other nuclear powers have created for ourselves over the past three decades. That predicament may be summed up as follows: it is precisely the horror, the wanton destructiveness and the resistance to control of these weapons that gives them their "beneficent" deterring power. Attempts to "stabilize" their profoundly destabilizing character or to make "rational" our exploitation of them or otherwise to inject logic and sense into their handling, while perhaps reducing the danger in one way, increase it in another. They increase the danger that these weapons will be used.

When people say that deterrence has worked, what they mean is that we and the Soviets have not used these weapons against each other and have kept out of a lot of direct conflicts we might have engaged in if it weren't for the terror of the bomb. This result has its own indisputable, overwhelming moral justification, but no one should think it has been cost-free. Ask the Poles. Ask the folks in Afghanistan. The superpowers are scared of each other. So they let a certain number of things occur. And they fight their wars in a vicarious, indirect way, sometimes with actual third-party or covert proxies and sometimes with gigantic investments of money in ever more brutal and exotic weapons systems that they show off and test for each other's intimidation.

Deterrence in this sense has worked precisely because the weapons themselves are so terrifying, so indiscriminate in their instinct to kill, so unreliable in their patriotism and allegiance, which depend, quite literally in certain respects, on "which way the wind is blowing."

People who have not spent much time pondering the baleful history of human warfare sometimes misconstrue the past in relation to nuclear weapons, romanticizing it beyond all recognition. Make no mistake: siege warfare was a cruel disease-bearing scourge of civilian populations. Gore flowed, and fighters along with their prey were brutalized in those "glorious" battles we tend to envision as stately tapestry art. But surely in terms of mismatches and disproportion--the excess of punishment for provocation and the victimization of the innocent--none of this can approach the prospective monstrosity of 15-minutes-to-hell nuclear war. Intuitively we all know this. We are afraid. That fear, it is true, could conceivably get someone to shoot first at a time of great tension. But until now it has worked the other way.

Our problem is that when we try to tame the beast--with substantially less lethal, "tactical" nuclear weapons, civil defense precautions, efforts to redirect retaliatory strategy so it should not automatically involve urban massacre, first- strike/second-strike and limited/general war distinctions and the rest--we seem to make nuclear war more "thinkable," that is: less cataclysmic, less total, less final and so, less unacceptable. The defense intellectuals who sought to rescue us from a primitive approach to nuclear strategy some 20 years ago have only been answered with what must strike them as rank ingratitude: attacks on their cold- blooded Strangelovian discourse, on their apparent easiness with the idea of nuclear conflict. This is unfair. But it is also at some psychological level both fair and right. People know what kind of a beast we are riding, what our predicament is.

This, I think, not the hawk and dove business, represents the real locus of our argument over nuclear policy. Or at least it should. To make a threat (and therefore deterrence) "credible" you must make it a threat that doesn't vaporize you along with your foe. But when you have done that, you have made nuclear war just a little more practical, something a little easier to countenance. From day one, the tension has been between the horrendous consequences of taming, i.e., "controlling" these weapons and the horrendous consequences of not doing so. I've read the various proposals of the critics of the status quo and I've contemplated Haig's comprehensive statement of the rationale for current policy, and I don't think that any of them has found the right way yet. I am certain of only this concerning our nuclear condition: things are greatly worse and harder to fix than we think.