The underlying reasons for these two endstates have not materially changed. First, an army can't impose a new political order. Eastern Europe, when it was liberated from the Soviet Union, snapped to the shape it had had before the Second World War - namely, democratizing and mechanizing states. Central Asia became as nest of despots - again, snapping to a modern version of its traditional forms. There was no reason to believe that Iraq minus Saddam would be very much different from say, Yugoslavia minus Tito.

There is however, a conterveiling force - and that is that Iraq cannot develop until there is peace and security. However bad a deal the Sunni are destined to get under a unified, or federalized, or even partitioned, Iraq, the deal they are getting now is worse.

This is why the Mission Abolished thesis is as it is: that what the American Public should be demanding is not an exit strategy, but peace talks. As Burke said "Peace in its usual haunts, not peace by means of war".

America has created in Iraq a new Northern Ireland - where we back one sectarian faction in suppression of another politically and economically. The British have lost more than a thousand soldiers 40 years of military involvement in Northern Ireland, a constant drain on a nation that could ill afford it. America's involvement in Iraq is more treacherous and intense, and rivals the Soviet quagmire in Afghanistan.

The only solution is peace - brought on by means of general talks, and backed by an organization which is both credible and within reach. That organization must then be either the Arab League or the United Nations. Since the Arab League has, at best, a poor track record in enforcing talks and creating peace, it falls then to the United Nations.

The precondition for these talks would have to be a cease fire, and a cease fire backed by a recognition of territory. Iraq, like post-war Germany, would be divided into sectors, as would Baghdad. The insurgency would be recognized as being responsible for sectors, the United States would have sectors, and, this is crucial, other nations would enter into the process through the United Nations.

The current political process is the "bandwagon" process - where a bandwagon is started, and groups are urged to jump on board or be left out. Right now US friendly Sunni factions are begging Sunnis to get involved, because they are left out. The Sunni population was willing to vote last time, but many stayed away out of fear - fear of both the US and the insurgency itself.

The bandwagon process works with risk adverse participants and a powerbroker who controls the flow of money. This is why Bush is able to govern on such a thin plurality here: he starts bandwagons, and enough conservative Democrats jump on board, desperate to be with the President.

However, actors in both Iraq and Afghanistan are not risk adverse actors. It is interesting to see how the Republican reactionaries in the United States have utterly failed to deal with actors like themselves - gamblers going for control of everything, and willing to suffer short term losses in order to reach for that goal. The Taliban in Afghanistan and the drug warlords are not interested in bending a knee to the Republic of Kabul, simply because they can do much better than begging Karzai for a slice of American aid.

In Iraq the situation is more dire, because the stakes are much higher. Instead of a meager opium trade and human slave trade living, there is oil, light, sweet and cheap to produce. One could flatten all of Iraq, and it would still be valuable real estate. More over the insurgency understands that it is dealing with one of its own - the Iraqi defense minister is a former security general under Saddam, and is using the same tactics that Saddam used - torture, prison camps and creation of a large security force. They know how he will act, and they know that the object to target is the police force - which they bomb relentlessly.

It would not be too much of a stretch to believe that at some point Sa'dun al-Dulaimi, or someone very much like him, will reach for the brass ring himself. Particularly if staked to the only working security force south of the Kurdish area. At that point a simple de facto partition takes place. The strong man gets central Iraq and the nearby fields, control over the monetary apparatus and mechanisms of legitimacy, the Kurds get the north, its oil fields and pipeline to oil hungry Turkey, while the shia get the south, its oil, and the oil terminal to the oil hungry Pacific Rim. The insurgency gets sand, with whatever potential oil is under it, and will come to terms.

This the insurgency wishes to avoid, getting on board with the future strong man - now that the Defense portfolio is in the hands of someone they understand - would be advantageous for him and for them. There is pressure, not for them to negotiate peace, but to negotiate joining the Sunni wedge in the government, and eventually turning the sights on the top position. Iraq had coup or revolution in 1958, twice in 1963 and in 1968. It had a virtual palace coup when Saddam went from being second in command to being virtual ruler of Iraq. It is not a process which is difficult in a nation where control over a single brittle infrastructure is key to the whole country.

A de facto partitioned Iraq is the worst of all possible worlds. A look at Iraq's history under the Qasim government of 1958-1963 shows the proable course: at first the central government, backed by the urban poor, negotiates alliances with major groups, including the Kurds and southern Shia. However, these fall apart one by one, leaving a government with a popular base in the "Baghdad Street", but with no effective military or security power. Then there is a coup mounted from within by those with that security power, and a struggle for control. Qasim gave way to the first Baathist government, which stood only a few months. Then Arif came to power, and repeated the same cycle over again.

To avoid this eventuallity requires either the Shia government forming that can maintain security - which is not happening and is not going to happen - or the Sunni strongman will have to be quite strong. It is dangerous to under-estimate the power of the al-Dulaimi clan and its connections, both within the official Iraq, and within the insurgency. Once this is done there is little reason to believe that a reconstituted army and security force would not wish to turn its guns upward.

<HR>

If the US is fortunate, such a political arrangement would be pragmatic and want to pump oil to rebuild the nation. However, to do so, and maintain its own power base, would require that it take far more of the profits than the fraction of oil actual in Sunni dominated lands. Saddam's method was to build refineries in Sunni areas.

However, there is no reason to believe that, even should this arrangement be fortunate at first, that it would remain so. Power not only tends to corrupt, it also tends to seek its own level.

A large fraction of the US public would be willing to take this gamble. Perhaps with the veneer of peace talks to bring the insurgents into the government.

This is the force of gravity alternative, the gradual cutting out of the Shia from the real mechanisms of power - though giving them control over the courts and law formation, though not its actual implementation. The Defense post would eventually have control over a mechanism that would be able to overthrow the state, and would have no reason not to use it at the appropriate time. Before the war the British found having a new strong man an acceptable end-state, and this leads to that end-state.

<HR>

The only other end-state which is politically viable is some form of de jure partition of Iraq. There are three models: federalism, confederation, balkanization.

Of these the last can be dismissed almost at once as a desireable end-state. The problems with trade and populations would create an intractable mess. There is too much ethnic, tribal and sectarian mixing in Iraq to draw clean lines that would create functional countries without ethnic relocation, at best, and ethnic cleasning, at worst. The break up of the former Yugoslavia is not a model we wish to pursue in its broad outlines, even if it offers tactical lessons in peace making and reconstruction.

The second two, however, are both viable.

Federalism would be the creation of an Iraq with strong powers in decentralized regions. Since Iraq is already on its way to dividing, federalism would be a way of blessing this process. It would also mean that there would be a motivation to participate in a central government as a counterweight to local conditions. This is seen, for example, in Kirkuk, where local groups complain about the Kurdish take over of government. Iraq divided into "States" or "Provinces" that had their own mandate of power would create the ability of the Federal Government of Iraq to interpose itself on the states, but also the local units to press back on the federal level.

Federalism has the advantages of unity, keeping the fiction of a single nation, the easy of interface with a federal government capable of signing contracts and so on.

Confederation would be the creation of theoretically sovereign states which none the less have trade and monetary union. A level of federation stronger than the European Union, but allowing much wider latitude for local laws, and forbidding the interference of the central authority in local affairs beyond rather restrictive boundaries necessary to maintain a national government.

This arrangement, because it comes with a great deal more nominal independence, has the feature that individual groups would turn inward to their own squabbles and factions, and have much less incentive of trying to "overthrow" the weak central authority. It has the same as a downside - a weak "Confederation of Iraq" would be less of a counterweight to Iran, and it would be far easier to pry apart by outside influences.

But either would be preferable to a de facto partitioned Iraq under an undemocratic, or only nominally democratic, president. Even if there is constituted a nominal parliament. Turning Iraq into Pakistan achieves none of the important objectives that must be attained. It does not keep the peace, it does not make it so that individual groups have a stake in the success of any entity known as "Iraq", and if Pakistan is any lesson, it does not prevent the acquisition of a deterent. In fact, a strong man has every reason to acquire a deterent, given that there are 3, and possibly 4, states with deterent access on its borders or in its region: Turkey is part of NATO, a declared deterent power, Israel is known to have or have had deterent capabilities, Iran is nearing deterent status and Saudi Arabia has unknown capabilities but could easily acquire them at any time.

Thus the way to head off the "glide path to dictatorship or failed state" is for the UN to enter and offer peace talks. But for the same reason the UN cannot be the security apparatus. First, it does not have the experience, but second, it cannot be a neutral broker and the military power.

The first step to talks would be the negotiation of a cease fire and demarked territorial control. The sector map of Iraq could be drawn in numerous ways, by province, ethnic area, along road transit lines, any of which would be better than the present circumstances.

The second step would be the introduction of security forces from nations other than the United States. In this "peace enforcing" role there are two nations that present themselves as having both spare military capacity and the ability to assemble a security zone which could head an Islamic mission to Iraq: Indonesia and Malaysia. These nations have military surplus capacity, experience with occupation. While not trained to desert warfare, they both have experience with insurgencies and guerilla tactics. Both nations are making the transition to being members of the global community. Both nations are ethnically distinguisable from both natives, and from Americans of all kinds. It may not be politically correct to say this, but "race matters" in Iraq.

The third step would be the negotiation of peace itself, which could take a very long time. However, talks are superior to what is being done now, with an "inside/outside" structure. The present structure encourages groups to continue fighting and building their own capacity, separate from Iraq, in case Iraq should go wrong. The Kurds and the Shia are building their own military capacity.

The sample partition map above should show the thinking. The Sunni insurgency would be offered the chance to have governmental control of a large section of Iraq, if, and only if, it agreed to "no more attacks", including suicide bombs and ambushes. The US would withdraw to a small sector, and would not engage in offensive strikes of any kind. The British would be asked to fill a larger sector. The Islamic force would take the Euphrates routes, and, this is particularly important, take over the contentious road that runs west from Baghdad. There would be major cities that would be put under UN peacekeeping missions, including Baghdad itself, Ar-Ramadi and Al-Fallujah.

The cost of this would be as much as the current occupation, there would be no savings. Peace talks could be held in Al-Najaf. Part of the confidence building phase would be an agreement would be that the insurgency adopt "an insignia visible at a distance" and give, as well as expect, POW treatment. Importantly, opening the Sunni sector to humanitarian aid, distributed by UN NGOs, would be dependent on this step. Aid workers would go under non-US, non-NATO UN escort.

<HR>

This peace oriented solution is not dependent on a withdrawal time table. There is no such time table. It would instead require presence as long as is necessary, though as the talks continue the US and UK could gradually hand over more and more duties to UN sponsorted peacekeeping missions. The implicit threat of wakening the US dragon would be maintained, but US casualties would drop dramatically as the US would neither be engaged in offensive operations, nor would it be patrolling many of the most dangerous roadways.

The offer for peace, itself, along with an offer to suspend offensive operations in Iraq, would be the opening step. It could be conveyed by the UN Secretary General.

The word "immediately" is now being floated by many members of the left with regard to Iraq - and immediate is what the offer of cessation of offensive actions would be. The US would agree to suspend them for one month, including supsending of arrests of insurgents and other "hunt them out" operations. If there were no more hostile actions by the insurgency, this could be extended for another month, and so on, on a month by month basis.

The composition of the insurgent negotitating would have to be determined by their own process, under the umbrella of knowing that there would be no US offensive actions during that time. This would allow them to gather in one place, and chose some effective leadership. The impulse to "box out" the leaders of the insurgency, to give them nothing, is one of the worst impulses in current US policy.

The process could be politically kicked off in the US by resolutions in both House and Senate to "negotiate peace with all armed groups in Iraq, beginning with an offer of an immediate cessation of offensive operations by the US." The theory that the US must "keep up the military pressure" in order to push the insurgency over the brink is clearly incorrect. What is more important is to gain time - time for the Iraqi security forces to be trained - and I reiterate my reccomendation that they be trained outside of Iraq and then returned to Iraq once they have reached military competence, thus putting them out of reach of the insurgency.

It could then be put before the UN General Assembly, and the UN Security Council - which could move resolutions stating that the UN, "concerned about the loss of life and civilian deprivation in Iraq, urges all armed parties to engage in a unilateral cease fire, and enter into negotiations for a peaceful transition in Iraq."

This would also shift the war aim in Iraq, from "pacification to a compliant US state" to "a stable and peaceful Iraq with all parties involved in the government". It is the belief - strongly reinforced by American actions - that it is American intent to pick the new leadership of Iraq, and force out those that they do not like, which is the most important stumbling block to an end of Iraq as a conflict zone.

Negotiations will take time, at least two years at the very minium, and more likely three to five. This will not save money, though it will save lives and equipment. If the Sunni insurgency will not negotiate in a "no preconditions" format, then it will be clear to the wider world that they are intent on a hostile overthrow of the government in Iraq, and that the consequences for allowing them in power would be dire. At that point realpolitik solutions could be come to which would involve training a much larger Iraqi army and accepting a federalism with a strong man solution in the iterim, accepting that it will almost certainly come to a simple strong man at some point.

In short, negotiate now. There is a clear, and implmentable road map. One that has the UN as the broker, but individual forces as the security apparatus. One that will not save money, but will save lives. It is time to realize that the mission of creating a "United States of Halliburton" in the Middle East is the stumbling block, that the neocon attempts to salvage something of their grand designs the expensive error, and that the road to peace lies in accepting that "regime change" has happened, and that the imperial mission should be immediately abolished.