How many guns, and which ones?

Gen. Maxwell Taylor, former chairman of the United States Joint Chiefs of Staff and one of the world's most respected military experts and thinkers, proposes that the US adopt what he calls a ''task-oriented'' national strategy.

And high time.

How does the Pentagon justify its present requests for military appropriations from the Congress and on what basis does the Congress decide to appropriate the funds?

You and I both know the answer.

The Pentagon asks for all it thinks it can get and then justifies it by charts showing what the Soviets have. The ''threat'' is always the ''Soviet threat.'' If the Soviets have something, or if we think they may have something, then the Pentagon wants the same thing - if possible, better and more of them.

I am exaggerating and oversimplifying, of course, but by and large it is true that Washington allows the Soviets to determine the decisions in Washington on what guns to buy and how many of them.

Is this the best way to protect the security of the US, and also protect the Treasury from unnecessary spending?

General Taylor proposes that the US first identify its national interests. Then single out those national interests which require protection. Then identify the possible threats to the interests. Then work out the amount and type of military power necessary to protect the specific interest against the specific threat.

In this process you start with a definition of national interest, not from what the Soviets have. Instead of saying we must have something because the Soviets have it, you start by deciding what is vital. If you start that way, you may end up with quite different force requirements and weapons systems. And you will find that in some cases you will be concerned about threats other than Soviet ones.

In some ways, of course, this is done now on an emergency basis. When the revolution in Iran threw out a friendly Shah and replaced him with a definitely unfriendly Ayatollah, Washington immediately decided to build up a ready deployment force for the Middle East.

But those entrusted with developing such a force faced a number of difficulties, including the fact that the new battle tanks being built for conceivable war on the north German plain are unsuited to desert conditions prevailing in Arabia and much too heavy for quick air deployment.

The British experience in the Falkland Islands is a case in point. The British did a remarkable job of improvising a task force which indeed proved to be an answer to the need. But as the story comes out in its full proportions we learn that it was a close-run thing which could easily have failed.

The British had not adequately estimated the importance to themselves of the Falkland Islands. They failed totally to estimate the threat aimed at that interest. And when the threat was unleashed they did not have a task force in hand designed specifically to meet that threat. It is truly remarkable that they were able to put together an effective answer as quickly as they did.

No country could anticipate every possible threat to its national interests. And the mere definition of national interests would be a formidable task. It would bring out wide differences of opinion. Another obstacle, as General Taylor points out, is that there is a tendency for each new president to want to discard the work of his predecessor, particularly if the predecessor is from the other party.

The definition of interests can change suddenly from one administration to the next. And, of course, each new president tends to think that the measures taken by his predecessor were the wrong ones. Continuity in foreign and defense policy thus becomes a particularly difficult thing for the US with its constitutional imperative for a presidential election every four years.

General Taylor proposes an ideal condition under which it would be possible to obtain a bipartisan agreement on interests and then on measures to sustain those interests.

But even a start in the direction of ''task-oriented'' strategy could, the general thinks, reduce the justification for an arms race with the Soviets and make it harder for defense contractors and the Pentagon services to justify expensive weapons systems. They would first have to prove a rational ''task-oriented'' reason.