THE TWO ALTERNATIVE U.S. campaign plans entail horrendous risks. One calls for a direct south-north offensive to dislodge Iraqi forces from Kuwait, and the other for an "indirect approach" from the southwest to cut right across Iraqi territory so as to isolate Saddam Hussein's forces in Kuwait. Even if accompanied by an amphibious landing of the Marine brigade kept afloat for that purpose, a straight dislodgement offensive would amount to a frontal attack against the minefields, earthworks, dug-in troops and artillery fires of the Iraqi deployment in Kuwait.

Very weak in the air, and incapable of fast armored maneuver in the Israeli style, the Iraqis are by contrast very experienced in static defensive warfare. Their skilled and well-equipped combat engineers have turned much of Kuwait's territory into a huge fortified zone, guarded by vast minefields, massive sand walls, anti-tank ditches and strong points that contain hundreds of dug-in tanks. With such defenses, even mediocre infantry can resist for several days of bloody combat.

Above all, the Iraqis have gathered much artillery in Kuwait. There was little of that in Vietnam and none at all in Grenada or Panama. Yet it is rapid-firing artillery that characterizes modern ground combat and inflicts most of the casualties. Hence few of the U.S. Army and Marine officers now in Saudi Arabia have a real idea of what they are facing.

Even if preliminary air attacks greatly reduce their number, Iraqi guns and howitzers are still likely to kill and wound many American troops -- while quickly scattering their local Arab allies (the Saudis have sketchy training and no combat experience at all). While the widely-circulated estimates of 30,000 U.S. casualties including 10,000 killed were too high given the original size of the U.S. ground force, they are becoming credible now that the president has more than doubled the size of the combat echelon that would be in contact with the enemy.

Actually the straight dislodgement option looks so bad that one must wonder why it is being considered at all. One possible reason: some Saudi princes are lobbying for it -- and so are the Marines. For the former, the great advantage is that it would entail no intrusion into a "sisterly Arab state," in the words of Sultan ibn Abdul Aziz, the Saudi defense minister. Some princes, including Bandar, the Saudi ambassador to Washington, agree with King Fahd that Saddam Hussein must be finished off by any possible means.

Others are still opposed to the idea of cooperating with infidels and non-Arabs against Moslem and Arab Iraq. Hence a counter-invasion of Kuwait but not Iraq is a compromise position that makes sense in Saudi political terms. That Americans should die to preserve the harmony of the Saudi ruling family may seem absurd -- but then of course one must recall that President Bush's entire commitment to the interests of both the Kuwaiti and Saudi absolute rulers seems absurd to many around the world.

As for the Marines, their desire to fight is commendable, but it is simply an unfortunate geographic fact that they can only do so if the attack moves straight up the coast from where they are deployed, incidentally offering the opportunity for their much-desired amphibious landing.

The seemingly dominant plan -- the "indirect approach" offensive -- would be indirect precisely to avoid the Kuwait fortified zone. A swing through Iraqi territory from points beyond the former neutral zone might well make swift progress without encountering much effective resistance. Not even the huge Iraqi army can strongly hold the entirety of the very long border, while U.S. air strikes should be able to break up significant Iraqi counter-attacks. The trouble with this option, however, begins precisely when the operation comes to an end.

At that point, the U.S. deep-penetration columns would have hundreds of thousands of Iraqi troops on each flank. Moreover, to isolate Iraqi forces in Kuwait, the U.S. columns would have to reach all the way to the border with Iran. Again, air strikes should prevent any Iraqi attempt to cut off the U.S. forces in turn by a concerted pincer movement.

But the necessarily small U.S. flank-guard units might simply be engulfed in some cases, even by Iraqi forces only trying to retreat from Kuwait. Certainly, casualties will continue to accumulate, until a final surrender that may be long delayed. Even the necessity to advance north to occupy Baghdad and the rest of the country could become unavoidable merely to end the fighting.

Iranian "revolutionary guards" have already clamored for a chance to fight Americans. They too may become involved in combat once the U.S. spearheads approach Iran's border -- as they must to cut off Iraqi routes to Kuwait. The government of Iran is most unlikely to oppose the United States by overt warfare (it naturally supports the anti-Iraqi coalition). But tens of thousands of "volunteers" may well be allowed to cross the border -- and they too could inflict casualties by ambushes and lesser attacks, day after day without an end in sight.