FROM SECRETARY of Defense Caspar Weinberger comes a major contribution on the subject of a perennial dilemma of democratic policy: deciding when to use military power. Mr. Weinberger heads straight for the hard part -- the "gray-area" Third World conflicts where the source and nature of the challenge are uncertain. His is no routine call to arms in the name of American global interests. Rather, his commendable purpose is to ensure that American power, when it is used, will be used to good effect and that American lives will not be put at risk casually. To this end, he lists six cautions or tests for U.S. forces to be sent into combat abroad:

1)The commitment must be deemed vital to our national interest or that of our allies. 2)It should be made "wholeheartedly, and with the clear intention of winning." 3)Political and military objectives and the ways to meet them must be clearly defined. 4)As conditions change, whether the commitment remains in the national interest must be reassessed. 5)Before a commitment is made, there must be "some reasonable assurance" of popular and congressional support. 6)A commitment to arms must be a last resort.

Secretary Weinberger, who came of age in the 1930s, is still stirred by the failure of the democracies to respond to Hitler in a timely and forceful way. What is more on his mind now, however, is a Vietnam-type situation in which the United States may succumb to the "danger of (a) gradualist incremental approach which almost always means the use of insufficient force." His implicit message is that if a commitment meets his six tests, it should be embraced. His explicit message -- the "Weinberger doctrine" -- is: "These tests can help us to avoid being drawn inexorably into an endless morass, whee it is not vital to our national interest to fight."

He identifies Central America, where he has stoutly resisted the dispatch of American combat forces, as one place where "the president will not allow our military forces to creep -- or be drawn gradually -- into a combat role." He might also have spoken of Lebanon, where he insisted first that troops, if they were to be sent, be sent for peace- keeping and not combat, and later, when their mission became untenable, that they be withdrawn.

In a sense Mr. Weinberger is simply distilling the post-Vietnam consensus -- in a way that, strangely, relates to Gary Hart's minority plank on "the selective, judicious use of American military power" in last summer's Democratic Party platform, even though there are large differences of stress. He has absorbed, too, the military's well-known and understandable reluctance to be assigned again, as in Vietnam, a mission which it successfully accomplished -- handing over the war to the South Vietnamese -- but which stopped well short of victory. Certainly his demand for rigorous precommitment review by Congress and the executive branch alike makes sense.

Notwithstanding Vietnam, however, it still needs to be asked whether a combat commitment short of winning should not continue to be an American option. Our Korean commitment became an "endless morass," but by hanging on, the United States ensured South Korea's independence and established that North Korean prisoners would not be returned to their homeland involuntarily. In other situations, might a president not wish to consider, among other choices, a policy of flexible response in order to set the stage for talks before a larger war broke out? An insistence on "a strong consensus of support and agreement" before a commitment -- a commitment to win -- would no doubt facilitate the distributing of responsibility for a foreign policy success or failure. But would it not also deny the president the compromise option of muddling through, which sometimes can be the best and only way?

Secretary Weinberger has not ended the debate on these essential questions, but he has reopened it in a serious and stylish way. His speech now becomes the central text to which others must respond.