When the Republican-controlled Senate Budget Committee voted to cut in half President Reagan's proposed 10 percent real increase in defense spending, it knew it was taking issue with his strong personal commitment.

Just hours before the committee voted, House Minority Leader Bob Michel, a skilled and respected politician of the old school, told a group of Washington Post reporters and editors that the president rigidly insisted on the full percentage increase.

"He's got a very fixed mind on this," Michel said. "His attitude is that there must be no territory lost to communism 'on his watch.'"

Yet loyal Republicans like Michel, and Senate Budget Committee Chairman Pete Domenici bit the bullet and followed their intuitive political instincts: they oppose Reagan because there is a bipartisan belief in Congress that a huge outpouring of defense dollars will merely perpetuate unmanageable budget deficts. In turn, these deficits will turn the nation into an economic basket case, making it militarily insecure.

Michel wants the debate moved away from the projected percentage increase in the defense budget to the more revealing question of what the $1.8 trillion Reagan proposes over the next five years is supposed to buy, program by program.

A bipartisan group of former high government officials gathered by New York banker Peter G. Peterson--all very much "establishment" figures--puts the same thought in very cogent fashion. In a letter dated March 25 to presidential assistant William P. Clark, defending their call for a slower but substantial defense buildup, these elder statesmen say:

"We would . . . rather (ask) what defense capability we need, what missions our military units should be able to perform, what situations and specific threat projections we should be ready to meet, and what technologies and hardware we need to develop and procure to meet them."

It's when you look beyond the dollars in the administration's program, and instead try to figure out what Caspar Weinberger wants to buy, that a chill comes over you. The proposal to build the highly publicized MX missile--whether deployed in Wyoming and Nebraska silos or anywhere else--is only one small piece of an arms race designed to give the United States first-strike nuclear capability. (We're still the only nation ever to have used a nuclear weapon.)

MIT's William W. Kaufmann, in a new Brookings Institution analysis of the budget, reminds us that the central objective of the Reagan-Weinberger defense budget is to gain the ability to fight a "protracted nuclear war" and at the same time have the capability, with conventional forces, to expand non-nuclear conflict by attacking the enemy's vulnerable points for an indefinite period.

At the heart of this sweeping proposal is the frequently cited warning that the Soviet Union has moved out ahead, and thus the United States must play catch- up. But most of the comparisons of the way the United States and the U.S.S.R. have invested arms money are crude and almost meaningless. Kaufmann cites one: the public is often reminded that the last American B52 bomber came off the assembly line in 1962. But in the past 20 years, the money spent on rebuilding the B52 exceeded the original cost.

"The issue, therefore, is not its calendar age, but whether (the B52) can continue to have a high probability of getting through Soviet anti-bomber defenses," Kaufmann writes.

But the frightening element is the Weinberger conceptualization that there can be a "protracted" nuclear war--and that somebody (preferably our side, of course) can win it. On this dubious proposition, Kaufmann is eloquent:

"A number of demanding conditions would have to be met before it could be said that either the U.S.S.R. or the United States had obtained the capability to conduct and successfully conclude a protracted nuclear war. Leaders, weapons, and communications would have to be made survivable for more than a few hours or days.

"Post-attack damage assessment and retargeting facilities would have to be available. Major economic and population targets would have to be saved from destruction. Not only would the belligerents have to be able to communicate with one another, somehow, and with timeliness; they would also have to arrive at an agreement on the course and probable outcome of the conflict. . . ."