THERE ARE two means by which the United States might achieve some of its objectives in the Persian Gulf: It might impose a result, the way the Allies conquered Germany in World War II, or it might influence a decision, the way the United States influenced the Soviet Union to withdraw its nuclear missiles from Cuba.

Our preferred method has been to avoid war and to seek to influence Saddam Hussein to withdraw from Kuwait, restore its government and, as it now promises, release the hostages.

But our ability to influence Saddam will depend upon what is going on in his head -- how sharply he sees the contrast between the two sides of his choice. We can influence him to withdraw from Kuwait by convincing him that the consequences, should Iraqi forces stay in Kuwait, are far worse than the consequences he faces if they withdraw.

Until now, we have been focusing on only one side of this influence equation: What happens to Iraq if it stays in Kuwait. By blockading the export of Kuwaiti oil we have prevented Iraq from benefiting from its invasion (beyond its looting). Economic sanctions have prevented Iraq from selling its own oil, costing Iraq about $100 million a day. And restrictions on imports threaten Iraq with real economic hardship. Of course, there remains the risk of war.

But almost no attention has been devoted to the other half of the influence equation: making withdrawal more attractive for Iraq. In fact, much has been said that could convince Iraq that withdrawal from Kuwait would only make it worse off: The identification of Saddam Hussein with Hitler, whether accurate or not, undoubtedly suggested that Iraq would face a high risk of war even if it were to withdraw. Public discussion by officials of the U.S., Saudi, Kuwaiti and other governments of additional objectives may have communicated that even if Iraq withdrew from Kuwait, sanctions and the threat of war will probably continue. Such added claims include compensation to Kuwait and the hostages, war-crimes trials, elimination of Iraq's chemical and biological weapons, overthrow of the Saddam Hussein regime and the reduction of the Iraqi army. After a poll suggesting that the elimination of Iraq's nuclear threat was the only objective for which a majority of Americans thought a war would be justified, that objective was added to the list of potential demands.

Given the array of signals we have been sending, it is likely that Saddam Hussein currently perceives his choice and its probable consequences as suggested in Chart A. Whatever errors the chart may have, its message is essentially correct: Saddam Hussein is remaining in Kuwait because for him, getting out looks worse. How can the United States and its allies alter that perception -- without compromising their core demands? The easiest way is to make clear, formal commitments in the United Nations or elsewhere that sanctions and the risk of war will stop if Saddam withdraws from Kuwait and releases all foreign nationals. It is not enough for U.S. government officials to offer hints that we may consider giving assurances. What matters is Saddam Hussein's perception, and he is likely making worst-case assumptions about our intentions, just as we are about his.

It may not be necessary to stop all sanctions in order to make withdrawal look more attractive than remaining in Kuwait. If the threat of war ends, and economic sanctions are lifted sufficiently to permit Iraq to import civilian supplies and earn immediate revenue on the sale of oil, that might be enough. Frozen assets might remain frozen and restrictions on the import of military supplies might remain in effect.

We can also increase the chance that sanctions will work by making clear to Saddam Hussein the ways in which Iraq will be better off by withdrawing from Kuwait. We should not squander our power to persuade by demeaning such assurances as mere "face-saving" gestures and by refusing to consider them. What in others we call "saving face," in ourselves we call "adhering to principle." If we have the ability to influence Iraq by giving it no more than that to which it is entitled, it is a mistake not to do so.

Is Iraq entitled to anything? A number of Iraq's concerns may be legitimate. These include: The Ramaila oil field. Iraq contends that Kuwait inched forward its oil wells onto the Iraqi side of the unmarked border in this rich field. It also contends that Kuwait exceeded its OPEC quota, stole oil from Iraq and drove down the price of oil, costing Iraq billions of dollars. Iraq's war debt to Kuwait. During its eight-year war with Iran, Iraq borrowed about $14 billion from Kuwait. Iraq would like Kuwait, in addition to contributions previously made to the "Arab war against Iran," to cancel this debt or at least renegotiate it. Access to the gulf for a deep-water port. Perhaps the major interest of Iraq is in acquiring sufficient access to the Persian Gulf to build a deep-water port for supertankers to be able to reach the open sea without having to go through either Iranian or Kuwaiti territory. Withdrawal of U.S. and other Western forces from the gulf. Iraq has an interest in having the United States honor its commitment to withdraw its forces once stability has been achieved. Limits on the claims for financial reparations. Iraq does not want to resolve the present crisis only to find itself facing enormous financial claims that would for years prevent it from earning any revenue from its oil. A fairer distribution of Arab wealth. Iraq, whatever its own record has been, now wants to be seen as a government concerned with the millions of poor Arabs who today get no share of the enormous revenues derived from "Arab" oil. The Palestinian cause. Saddam Hussein obviously has an interest in being seen as an Arab leader who champions the cause of the Palestinians, and who is able to do something for them. Not being denigrated or discriminated against. Iraq does not want to be subjected to standards that the world is unwilling to apply to others, whether these involve chemical weapons, nuclear reactors, military capability or forced compliance with Security Council resolutions. A fair process for dealing with its concerns. Apparently Iraq believes that Kuwait has refused to negotiate directly in good faith on issues that divide the two governments. (Kuwait has been reluctant to engage in direct private negotiations for fear of being coerced.) Some process could readily be devised that takes both sides' legitimate concerns into account.

Recognizing the possibility that such legitimate interests exist does not prejudice our legitimate interests or the means by which we pursue them. But convincing Iraq that following withdrawal from Kuwait, these issues will be dealt with on their respective merits by fair procedures could greatly increase the effectiveness of the sanctions and other U.N. resolutions.

Such assurances are not at all inconsistent with the Security Council's demand for unconditional compliance. This demand means that Iraq may not set the conditions for its compliance, but it does not prevent us from indicating what will happen after Iraq withdraws and releases all hostages.

For example, demands that Japan surrender unconditionally toward the end of World War II did not stop Truman, Churchill and Chiang Kai-shek at Potsdam from assuring Japan that it would be permitted to rebuild its industry, have access to raw materials and return to world trade -- but not to re-arm. Occupation forces would be withdrawn, Japan was assured, as soon as the declaration's objectives were accomplished. Secretary of State James A. Baker III and Gen. Colin Powell, chairman of the Joint Chiefs of Staff, have said we are using a carrot-and-stick approach in which the carrot is "that we don't use the stick." Yet, as the foregoing suggests, we can offer a more enticing carrot at no cost to our vital interests. In congressional testimony this past week, Baker reiterated that the United States would not negotiate because we could make no concessions on the demands of the Security Council. What Baker apparently had in mind, and rightly objected to, was the "standard bargaining" material of negotiations where each country starts off by adopting an extreme set of demands. The two parties look forward to approaching possible agreement through a "haggling" process in which each demands major concessions from the other, while itself making minor concessions or none. That is a game of positional bargaining familiar to all of us, but as a way of settling difficult international problems, it has many undesirable features.

It encourages the adoption of extreme partisan positions, the making of public commitments to those positions and a competition as to which side can be more stubborn. Standard moves are those which we have seen in recent weeks with respect to the gulf: Each side tries to demonstrate that it is prepared to take enormous risks, that it will not yield, that it may appear to be acting irrationally and that the choice of avoiding war depends upon the other side's backing down.

The real world is never a zero-sum game in which one side's loss equals the other side's gain. In the real world, it is always possible for both sides to lose. In fact, the present situation in the gulf is a "lose-lose game" in which no matter what happens, every country will be worse off because of Iraq's invasion of Kuwait. We cannot expect that anyone will be made "whole." So the purpose of negotiation is not to "win" in the sense of coming out with a positive gain but rather to minimize total losses, and to see that Iraq -- the perpetrator of the situation -- does not benefit from it.

There is, however, a more productive method of negotiation that would not require any concessions on our vital interests. That is for each side to try to persuade the other of what ought to be done in the light of precedent, international law, or some other objective criterion. If agreement is reached, neither has given in to the arbitrary position of the other. Each can explain to constituents why the result is fair. {See box.}

How would I suggest we proceed in talks with Iraq? The United States should keep the following points in mind:

1. Neither seek nor make commitments at the outset. It is far better to agree that the first several sessions will be exploratory. Discussions will be without prejudice. No firm commitments will be made or sought. In this context ideas can be exchanged, possibilities considered and the rigidity of a confrontation loosened.

2. Reach an understanding on the purpose of each session as measured by the product. But our purpose ought not be narrow, one-way communication. There has been much talk by the administration that the purpose of the meeting is simply to communicate clearly to Saddam Hussein that if he wants to avoid war, he has no recourse but to withdraw -- to make sure that he "gets the message." But President Bush, in dealings with other leaders, has demonstrated his recognition that the more serious the substantive difference between two nations, the more important it is that its leaders develop the kind of effective working relationship that enables them to deal well with those differences -- an approach that he should not abandon now.

A useful early product is a list of issues that might be included in an agreement. The United States will, of course, wish to consult Kuwait, Saudi Arabia, members of the Security Council and others before making any substantive commitment. A useful concept may be to think in terms of the possibility of developing a draft Security Council resolution that Iraq and the United States might jointly recommend. Or they might develop such a draft framework, report it to the secretary general, and ask the secretary general to submit a formal report to the Security Council.

However the United States, Iraq and others decide to proceed, it will be helpful to think in terms of developing a framework for a possible agreement. Chart B illustrates a possible outcome of an early stage of the upcoming talks

Rather than work from two documents, each of which reflects the position of one party, it is usually better to work side by side on the same rough draft without committing anybody to anything.

3. Consider different process options for working toward agreement. Critical at each stage of a negotiation is to think about the next stage and to see if agreement can be reached on the process for going forward. Such an agreement should not just identify who is to meet whom, where and when, but also indicate the work to be done and the kind of product that the will seek to produce.

4. Keep the sanctions options available.

The president's decision to increase troop levels in Saudi Arabia and obtain a U.N. authorization for the use of force was intended to make credible the option of going to war. That decision should not be allowed to preclude the option of maintaining sanctions if negotiations do not produce a quick result.

As President Kennedy discovered at the time of the U.S.-sponsored Bay of Pigs invasion -- when he found that invading Cuba was his only well-prepared option -- it is not enough to prepare only one option but tell the president he need not choose it. It would seem imperative now that the Pentagon be instructed to prepare plans for the option of not going to war, including presumably provision for rotation of troops, air-conditioning of some facilities in Saudi Arabia, the stationing of much heavy equipment in Saudi Arabia combined with a reduction in the number of personnel there at any one time.

Maintaining the sanctions indefinitely is a far less expensive way of winning than to wage a war. And whatever the difficulties, maintaining international support for sanctions is likely to be far easier than to maintain it for a war against Iraq.

The continued application of political and economic pressure and a military deterrent in concert with a broad Arab coalition is most likely to keep Iraq in line -- as such measures finally prevailed against the Soviet bloc. Few would wish now that we had waged a preventive war against the Soviet Union, as some once urged. A war could neither guarantee the now-promised safe return of the hostages nor an intact Kuwait. It is likely not only to be bloody and costly but to unleash Arab nationalist and anti-American forces that will create peril and instability for years to come. And the more that the United States talks about the need for a quick solution, the more we encourage Iraq -- if there is a war -- not to surrender but to outlast us as North Vietnam did.

Roger Fisher is director of the Harvard Negotiation Project and professor of law at Harvard Law School. He is the author, with William Ury, of "Getting to YES: Negotiating Agreement Without Giving In."