Abstract: The author’s book Incoherent Empire was written in 2002–3 in a burst of anger at the malevolence and stupidity of Western foreign policy. The book did not primarily take an ethical stance against American breaches of the Geneva Conventions and defiance of United Nations resolutions, but rather took a Machiavellian view, judging the morality of actions by whether they may lead to better or worse outcomes. Revisiting the book more than decade later, its predictions of disastrous outcomes for the invasions of both Afghanistan and Iraq proved to be amazingly accurate. America and its allies failed notably to understand that social order is a prior condition for good government. Rule of law and political accountability may follow, but order is the prerequisite. This failure of understanding has had catastrophic consequences not just in Iraq and Afghanistan, but in Libya and Syria too. The conclusion is that there should be no further interventionism, with two possible exceptions: when action has been approved by international authorities such as the UN or ICC; and, on Machiavellian grounds, when favourable outcomes can be realistically expected.

Keywords: US foreign policy; military invasions; Iraq; Afghanistan; Libya; Machiavelli

I wrote my book Incoherent Empire (Mann 2003) in haste in late 2002 and early 2003 in a burst of anger against the malevolence and stupidity of the foreign policies of my two governments, for I am a dual citizen of the United Kingdom and the United States. I essentially finished the book just before their invasion of Iraq actually happened, though I was able to hastily add an account of the invasion up to the looting spree in Baghdad in April 2003. I was confident that this invasion and the one in Afghanistan would not only be disastrous for these countries but would also create a massive blowback for America and Britain.

What amazes me today is how accurate my predictions were. Iraq is in sectarian chaos, Afghanistan is embroiled in tribal and ethnic struggles, as I predicted. US and UK troops have essentially left both countries, in an orderly but humiliating retreat. We in the West now live under massive and costly security arrangements which have restricted citizen freedoms and led to torture as a normal security practice approved in reality, though not in rhetoric, by our governments. Today we need more security than in the past (but not curtailments of civil liberties or torture), since there are terrorists out there trying to kill us. Yet our governments have helped cause all this to happen. What particularly distresses me is that people in the UK and especially in the US cannot perceive cause and effect, cannot see that it was our actions that helped to produce their deadly reactions.

There were other causes too. At the time that I wrote, the Islamist extremists were debating whether to attack the ‘near enemy’ or the ‘far enemy’ – that is their own governments or the West (this word being largely interchangeable with the United States). The root causes of the jihadi upsurge lie in the problems of the Middle East itself, more specifically in an understandable Islamist reaction against the region’s somewhat secular and highly corrupt authoritarian regimes – the near enemy. But our violence was the main cause of their attacks on the West. Of course these two enemies were entwined in terrorist eyes, since the attack on the near enemy was partly because they were seen as ‘apostates’ seduced by Western secularism. So they might have launched some attacks on Western targets anyway. But our violence greatly increased the scale of their attacks on us, and intensified the venom which led to alienated young Muslims in Western countries joining the international jihadi brigades. In the UK those captured before their plots came to fruition have virtually all said that British foreign policy was the cause of their action. The chickens have come home to roost, as I and many others predicted

My book did not dwell on the morally indefensible conduct of the US in routinely violating the Geneva Conventions on war, including bombing and invading foreign countries which did not threaten us, and in the case of Iraq contrary to UN resolutions. Instead I focused on the practical consequences of the invasions, adhering to Machiavelli’s notion that morality is to be judged in terms of whether actions lead to better or worse outcomes. Actions are immoral in this sense if they lead to worse consequences which were predictable in advance. I did predict disaster for several reasons but the main one was that it was a terrible mistake to invade foreign countries without having local allies on the ground. Local allies can provide some military assistance, but much more importantly after the invasion they can form a new government with some degree of legitimacy. I argued that imperialists in the past nearly always had this planned before they invaded and that was the key to their ability to rule over foreign countries. The US lacked such allies and such planning. It has shown itself to be a military giant but a political dwarf.

It is true that the US did have one group of local allies on the ground in Iraq, the Kurds. But the Kurdish forces would not assist in the reconstruction of a government in Baghdad, since they actually wanted to secede from it. The US had at first no Iraqi allied forces who would assist regime reconstruction. Eventually Washington realised its gross mistake and fell back on the traditional second-best imperial strategy: dividing and ruling by exploiting ethnic and regional divides. The US began to rely on the Shi’a majority to rule over the previously dominant Sunni minority. Of course, the US would prefer to see a government of national reconciliation willing to bridge this divide with power-sharing politics. It constantly pleads with the Shi’a-dominated government to stop discriminating against Sunnis, but Shi’a leaders largely ignore them. They have gleefully seized their chance to be the oppressors rather than the oppressed, with all the spoils that control of government (and of oil) brings. One result of that has been the support shown among Sunni areas of the country for the current ISIS militants. Again we see blowback, again we must see cause and effect.

In Afghanistan the situation was a little different, since the US did have a substantial ally, the Northern Alliance, which was already fighting against the Taliban-dominated government. But the roots of that struggle were tribal and ethnic, Taliban Pashtuns against a Northern Alliance dominated by Tajiks. So the US, in supporting one side against the other, defectively perpetuated an Afghan civil war which has intermittently flared up for over half a century. It has done so again. That struggle might remain stalemated, as it is right now, but the other most likely outcome would be victory for the Taliban since the Pashtun are the larger population, and since the US and its allies have tried to crush the one industry which brings serious revenue to the country and its warlords – cultivation of the opium poppy. My reasoning is again a practical one: this US policy diminishes further Afghan support for the kind of government the West prefers. Afghanistan has seen some improvements in life expectancy and education. But these are the same improvements as the Soviet-backed regime brought earlier – and that (with the help of the US through the CIA) resulted in the Taliban.

Americans and to a greater extent Britons have come to see the futility of these two invasions, and this has weakened the will of most American politicians to intervene. But they cannot quite rid themselves of military desires since their military is so powerful. The Obama administration tried to pretend that it had not intervened in Libya by hiding behind the British and French, but these countries could not have succeeded in overthrowing Gadaffi without American logistical support. The success, however, was limited to the killing of Gadaffi. Afterwards the country rapidly descended into chaos with militia fighting against militia and no effective government – a worse outcome than even in Afghanistan and Iraq. Despite political pressures coming from Republicans, the Obama administration dithered and then gave only a little aid to the Syrian rebels. Only the horrors inflicted by ISIS on Kurds, Christians and Westerners got the US into bombing mode again.

The US has been intervening for several reasons, some good, some bad. The most virtuous one, trumpeted by Washington over and over again, is to lead the region towards democracy. I accept that for many American politicians, even some of the neo-cons of the Bush period, this was sincere belief; other more realistic interveners sought a new military strongman. But this leads me to question the relative virtues of democracies and dictatorships. All these cases would seem to reveal that dictators have their merits. Saddam, Gadaffi and Mubarak provided more order than their successors, whatever their vices. Dictators must accomplish two things. They must maintain order and they must be willing to do that if necessary by repression. If they cannot preserve order they will be overthrown. Second, a more positive aspect of their rule is that they must establish a balance between relying, on the one hand, on their personal base of support, be it through ties of kinship, tribe, region, ethnicity or sect, and, on the other hand, on an ability to divide-and-rule between other social groups. Not only the core but also some more peripheral groups must share in the benefits of rule. Thus dictators (or oligarchies) can become quite adept at dealing with tensions and conflicts between different ethnicities and sects, playing them off against each other while retaining the ability to repress if necessary. Where moderately successful, these strategies bring order.

Some dictators are more successful in this goal than others. President Tito was a master at this. Even though he was widely perceived afterwards as having given privileges to Serbs within Yugoslavia, he made sure that all ethnicities were well-represented in the Communist Party, and in fact he needed very little repression. Gadaffi was moderately successful at playing off tribes, but he was undone by his regional bias against the Eastern provinces. The biases of Saddam Hussein for Sunnis and the Assads for Alawites did not prevent those regimes from also bringing others into power, although they also needed more repression. Of course, they will all tend to repress some more than others, and they all deal savagely with rebellion – Saddam Hussein’ s gassing of the Kurds is a particularly vicious example.

As Fukuyama (2011) has argued, political order is the first requirement of good government. If it exists then the rule of law and the accountability of leaders to the public may come too. Without order they certainly will not come. Saddam Hussein, Mubarak, Gadaffi and Assad to varying extents brought order. I am not suggesting that any of these dictators would actually institute the rule of law or representative accountability. But it might be possible for their successors to inch forward liberalisation measures, as the kings of Jordan and Morocco have done. In the one success story of the Arab Spring, Tunisia, the overthrow of Ben Ali and replacement by a somewhat democratic regime, the reforms had begun under his rule. Little good did it do him; much good did it do the country. In contrast, all the US interventions have brought more disorder, as in Afghanistan, Iraq and Libya.

What about the virtues of democracy, or rather of the democracy brought by American intervention? For the sake of argument I will assume that American intentions are largely benign, and indeed in all these American interventions there were efforts by the Americans to introduce some rule of law and some representative institutions. In all these cases the dominant local group either ignored them or steered them toward its own interests. This might still be considered an improvement on the Taliban or Saddam Hussein, but the downside is less order, in fact civil war.

Democratisation has its own problems. As I wrote in my book The Dark Side of Democracy: Explaining Ethnic Cleansing (2005) the democratisation of a country amid ethnic or religious or regional divides can be dangerous. The problem is that when the two Greek roots of the word democracy, the demos and the ethnos, become fused – as when any ethnic group believes it is the true nation, whereas other groups are not – discrimination almost always ensues from this, and sometimes ethnic cleansing too. The successful overthrow of a despotic regime by American military pressure might lead to some order, but order imposed for very undesirable purposes. Bad effects are particularly likely if (a) the transition to such a regime was sudden and violent, and (b) if a minority was beforehand the repressor (think Iraq for both). There is some truth in the old adage that democracy does not flow from the barrel of a gun. There is even more truth when the country being defeated is tribally, ethnically, regionally or religiously divided. In the Middle East religious divides exist between religious sects (e.g. Sunni against Shi’a) or between Islamists and relatively secular Muslims. The US would therefore have great difficulty in enforcing democratic institutions on any country in the region. Therefore it should keep out of the politics of the region.

There is an equally unacceptable alternative. It would be for the US to become a colonial power, not just invading and then withdrawing (as has been its practice), but staying and ruling through reliable but closely supervised local clients. These would be dictatorships or oligarchies, hopefully leading to more order but sustained by repression, perhaps with a limited rule of law, but almost no democracy. Such a policy would have two main problems, as I noted in Incoherent Empire: (a) in a post-imperial and nationalist age very few locals would regard this as legitimate, reducing the possible number of reliable clients and increasing the chances of rebellions; and (b) Americans would not accept it, since they do not see their country as imperialist and they cannot stomach sustained American casualties. Neither side favours colonialism, and so this is not going to happen.

What instead should be American policy?

First there should be no more US military interventions, neither bombings nor invasion, since on balance they do harm. There are two types of exception to this rule. First, when intervention has received international ratification under international law and norms. The Rome Statute of 1998 gave the International Criminal Court (ICC) jurisdiction over three international crimes: crimes against humanity, war crimes and genocide. The law is fairly clear in these three respects, although the statute added a fourth crime, waging aggressive war, over which there is not international consensus concerning what this includes. The court has the power to indict individuals on these grounds but has no power of enforcement itself. Yet foreign states can legitimately use force to pursue indicted individuals. For example, soldiers from several African countries plus the United States are currently pursuing indicted members of the atrocious Lord’s Resistance Army in northern Uganda. But for full-scale war the only legitimate agency is the United Nations, more precisely the Security Council. This is a highly imperfect institution, often immobilised by disputes between the Great Powers and usually rather slow-moving. Nonetheless, when it does call for intervention this reflects a broad international consensus. Waging war to stop perceived crimes should not be at the whim of a single state plus its allies. In other words there should be an end to ‘wars of choice’ by individual states (self-defence is of course a legitimate reason for war). This would lessen the number of interventions, but that would be no bad thing.

Second, intervention should be also subject to what I have called Machiavellian morality. Is military intervention likely to be successful in stopping, or at least containing, the above crimes? If this is unlikely, then intervention is not desirable. So military intervention in Rwanda in 1994 and intervention in Cambodia under UN auspices to stop genocides would have been desirable. The UN commander on the ground in Rwanda asked for a further 5,000 troops with which he said he could stop the genocide, but the US and France shamefully vetoed this.

Of course the two criteria of intervention need not be both fully present. One might compensate somewhat for the other, or there might be a measure of self-defence. The invasion of Cambodia by the army of neighbouring communist Vietnam did rapidly secure the fall of the Khmer Rouge, the installation of a more legitimate (though not democratic) regime, and the end of the genocide. This was without any international approval, yet it worked. In any case, Vietnam acted to some extent in self-defence, since the Khmer Rouge were killing ethnic Vietnamese living in Cambodia and had also launched punitive raids into Vietnamese territory. It was also obvious that the battle-hardened Vietnamese troops would easily rout the Khmer Rouge militias. This was a legitimate case.

The UN did give its approval in 1991 to intervention in Kuwait, which Saddam Hussein had flagrantly invaded, and the US and its allies had no difficulty in routing the Iraqi troops and pushing them out of Kuwait. This was legitimate on both criteria. On this occasion, wise Machiavellian heads then prevailed in Washington for the US stopped its advance and did not try to overthrow Saddam. It could have achieved this, but what then? –chaos and civil war a decade earlier than actually happened?

Obviously intervention based only on these two criteria would err on the side of allowing repressive regimes to remain in power. But we are over-estimating American power if we plump automatically for the opposite policy, of military intervention. The US has been badly burned by its ventures in Afghanistan and Iraq, as it was by defeat in Vietnam. But as happened after the Vietnam War ended, restraint might not last very long. Indeed, even now most of the Republican Party is urging more intervention in Syria and Iraq. However foolish such a policy is, it is more likely to happen than the two-principle policy I advocated above.

But what about ISIS? This is the current litmus test, since its atrocities do infringe international law. Indeed, on the first criterion there is in principle no problem, for the Security Council has called for joint action to stop the massacres being perpetrated by ISIS against Shi’a, Kurds, Christians and others. But it has not specified what actions should be taken, and the Great Powers are opposed to sending in ground troops. Some bombing has been inflicted on ISIS-held territory in Syria and Northern Iraq, led by the US but also including Britain, France, the Netherlands, Denmark and Belgium, plus Syria, Iraq and Iran, with Syrian and Iraqi forces also engaged on the ground. That is quite a broad coalition. But the participation of the Muslim countries brings its own problems. The West does not want to offer support to the Assad regime in Syria since the destruction of ISIS would benefit Assad and his repression. The other two Muslim countries are both Shi’a regimes while ISIS is Sunni. ISIS is mainly self-financing from oil revenue, but it also receives financial aid from Gulf Sunni countries, especially Qatar, while in Iraq, Sunni communities are often supportive of ISIS. Sectarian war has arrived.

This raises problems for my second criterion, the practical consequences of intervention. This is admittedly difficult to gauge, but that itself indicates a strong element of risk in interventions. Obviously a full-scale invasion by all these powers would have a good chance of military success over ISIS but there would probably be further blow-back, especially among Sunnis, the majority Muslim faith. Even if ISIS fell, anti-Western and anti-Shi’a terrorism would certainly reappear elsewhere, convincing more young Muslims that the West is evil. Blowback, though of lesser dimensions, would likely ensue if bombing were escalated. Every bombing incident which results in the death of civilians – and this is an inevitable consequence of bombing – is trumpeted across the Middle East. In contrast our media publicise the Pentagon’s dubious assurances that civilian casualties are very rare. By regarding ISIS militants as simply evil, the West avoids asking what their motives are and particularly what has been our role in their creation. ISIS has good reason to hate the West, though not of course to massacre other religions and ethnicities of the region.

In the end I retain the bias against interventionism I revealed in Incoherent Empire. Western intervention should be confined to economic aid, of which there should be much more. Our military power is great, but it cannot achieve goals that are essentially political. Even our ideology no longer goes down well in the region, for it is seen as hypocritical. We preach democracy but practice militarism. In any case who is concerned about democracy if there is no order? If we do wish to exert political power in the region (indeed, in most of the world) it should be limited aid to elites who are likely to bring more order to their country or to those who already do provide order in return for small liberalising steps. That general principle applies to ISIS also: give aid to the Iraqi government and the Kurds, and protect more vulnerable minorities, and that is all. Of course, this is one major thrust of the somewhat wavering policies of the Obama administration, alongside the opposite tendency elsewhere toward an ever-increasing number of drone attacks.

Middle Easterners must be largely left to solve the problems of the Middle East themselves. And they will. ISIS will not last long if Iraq, Iran, Syria and the Kurds increase their military presence on the ground, no matter how much oil money they have to buy weapons. They are too few in number and they cannot concentrate their forces without being vulnerable to air strikes. Extreme Islamism is having its day, and broadly speaking we must accept that. But when it fails to address the real problems of the people it will be either overthrown or learn under pressure to moderate, as has happened in Iran. We will only extend the life of extremist groups if we intervene militarily. This general principle also applies to the other long-running sore of the Middle East, the struggle in Israel–Palestine. End the enormous military and economic support to Israel and let the locals sort out their own problems – Israel is now sufficiently strong to stand on its own two feet.

My argument will probably not be much appreciated in the United States, because it comes close (outside the economic realm) to ‘isolationism’ which almost everyone in Washington except Rand Paul on the right and Bernie Sanders on the left rejects. In the past the world often benefitted from American military intervention. But those days are over. Give economic aid and focus the military on self-defence and its benign and much appreciated role of combating disasters like famines, tsunamis, shipwrecks and aircraft downings.


  • Fukuyama, Francis (2011)The Origins of Political Order. London: Profile.
  • Mann, Michael (2003) Incoherent Empire. London: Verso.
  • Mann, Michael (2005)The Dark Side of Democracy: Explaining Ethnic Cleansing. Cambridge: Cambridge University Press.

Biographical Note

Michael Mann is Distinguished Research Professor of Sociology, University of California Los Angeles, and Honorary Professor, University of Cambridge. His many publications include the acclaimed four-volume study of The Sources of Social Power.