Of late, the actions of the House in dealing with the defense appropriation bill have been perplexing, to say the least. On one day it inflicts a deep, though not lethal, cut in the MX budget. On the next day, it proceeds to vote the funds for MX research and development, as well as for two other major programs every bit as controversial as the MX -- the B1 bomber and the two supercarriers for the Navy. What goes on?

Various sources -- the government, the media, politicians, Pravda and company -- are advancing quite different reasons for this extraordinary behavior. Only one explanation seems widely accepted. Congress, pressured by the recession, constituent unhappiness and special interest causes, apparently intends to scrutinize the military budget far more closely than in the past, while continuing to support the essential elements of the president's rearmament program. If such is actually the case, congressional behavior and the fate of the president's military policy will hencefore depend largely on the soundness of legislators' views of what elements are truly essential.

In his speech of Nov. 22, President Reagan touched on the central theme of my argument here when he defended the MX program on the ground that it was an essential weapon. But at the same time, he made little effort to explain what he meant by essential and why it applied to the MX. My argument is that Congress should make it a practice to evaluate the essentiality of the MX and all other controversial programs of high dollar value prior to voting on them. Then, regardless of other possible motivations, members of Congress could tell their colleagues, the media and their consciences: "I voted for (or against) weapon A or program B because of solid evidence that it is (or is not) essential to our national security."

To illustrate a procedure for testing a program's essentiality, let us examine the claim to essentiality of the MX program. But first I must define what is meant by essential in this context. An essential weapon is one that, at a tolerable cost, can make an important contribution to the deterrence or defeat of an urgent danger which has 1) a high probability of occurrence and 2) an assured capability for inflicting serious damage. Justifying the need for the MX, the usual danger postulated is a Soviet surprise attack on our relatively unprotected land-based ICBMs, our so- called "window of vulnerability."

To what extent may we expect the MX to contribute importantly to deterring or defeating such a surprise attack? With its 100 new missiles and 1,000 modern warheads, the MX system might be expected to constitute a major reinforcement to the deterrent, retaliatory and first-strike effectiveness of our forces. But there are a number of factors tending to diminish this apparent asset.

The first is the uncertainty of the vaunted invulnerability of the Dense Pack silos. The confidence expressed in them by the administration assumes the working of the "fratricide" principle, which asserts that the first wave of Soviet missiles arriving at target will destroy all or most of the succeeding missiles. Administration scientists and nuclear experts say it will work. Many outsiders of equal qualification and reputation say it won't -- at least not for sure. We shall not know who is right until Armageddon Day.

It may also be argued that the danger constituted by a Soviet first strike is not a real threat in that it has a very low probability of occurrence even with our present strategic forces. The Soviets are deterred not only by the numbers of our weapons but also by concern about the reliability of their own missiles, which like ours have never been fully tested. They also are well aware of the delicate exactness required in the time of arrival of their missiles on target and the high probability of a disastrous U.S. retaliation. Lastly, we should not forget that all good Communists have an ideological right to believe in the eventual collapse of the capitalist system of its own accord. So why expect them to risk a first strike?

Our definition requires that an essential weapon should perform its function at a tolerable cost. At the moment, one cannot be sure of the total cost of

the MX system. Estimates vary from the

administration figure of $26 billion to as

much as $50 billion. Given the Pentagon's

record for overruns, the latter figure is the

more probable.

A final consideration before "buying" the

MX is whether or not there are other and better ways to achieve the contribution that the MX purports to make. In view of the uncertainty of the survival of the MX in its Dense Pack silos, a safer, surer and probably cheaper way to increase our deterrent power would be to rely more on submarine-launched missiles and air-and sea-launched cruise missiles while progressively eliminating our land-based ICBMs. Submarine-launched missiles are becoming much more powerful and more accurate; the cruise missiles are a cheap and accurate way to attack fixed Soviet targets without risking a bomber or a pilot. Without dependence on the land-based missiles, the "window of vulnerability," if it exists, would disappear along with tempting land-based targets on American soil.

Summing up the results of this probe of the essentiality of the MX system, I find that it fails to qualify as essential because of the uncertainty regarding its survivability, the unlikelihood of the first-strike threat it is designed to counter, the unpredictable dollar cost and the existence of apparently better and surer ways of accomplishing the purposes of the MX.

In reaching these conclusions, I hope to have demonstrated a procedure that both the Pentagon and Congress mighttwell find useful in many similar situations affecting the military budget. The procedure should help in measuring the true worth of weapons and programs, as well as facilitating the determination of the relative merits between competing programs. Above all, it offers a simple behavioral principle to guide the conscientious legislator, bemused by the complexities of the military budget: "If proved essential vote yes, otherwise, no."