Then, something unexpected happened. Instead of rising further and developing into an all-out war, the sectarian violence began to drop. At first, nobody knew quite what to make of it, and even suspected the Pentagon of statistical skulduggery. MoveOn.org, a prominent activist group, accused General Petraeus, commander of the coalition forces in Iraq, of "cooking the books" for the White House. When Petraeus presented the data to Congress in September, he faced similar scepticism.

Over the past two months, however, it has become increasingly apparent that the drop in violence is both real and significant. Official US military statistics have been corroborated by independent sources such as the Brookings Institute, Iraq Body Count and iCasualties.org. These groups measure different things -- some focus on Iraqi civilians killed, others on coalition military casualties, and others on particular types of violence, such as sectarian killing. But they all show the same broad trend: a sharp and sustained drop in violence during the second half of 2007. The statistical evidence is backed up anecdotal evidence from Baghdad, where reporters speak of city cautiously returning to normalcy. Shops and markets have reopened, and a tentative nightlife has returned to the city. The latest positive indicator is the return of refugees: according to the Iraqi government, 46,000 refugees came back to Iraq during October alone.

This does not mean that Iraq's problems are over. The political violence has been reduced from the staggering heights of 2006, but it remains at roughly the level of the 2004-2005 period, when Iraq was hardly a peaceful country. And any reduction in violence, no matter how steep, will mean little in the long run unless Iraq's factional leaders can agree on a durable political settlement. Even so, compared to the horrific sectarian reprisals that gripped Baghdad a year ago, this is a remarkable reversal, and it is worthwhile trying to understand how it happened and what it means.

The first question is easier to answer. The November elections were a turning point: defeat at the polls seemed to shock Bush out of his complacency, forcing him to face up to the fact that his strategy in Iraq was failing. A raft of personnel changes were announced, including Robert Gates as the new Secretary of Defence, Douglas Lute as the president's "war czar", William Fallon as chief of CentCom, and Ryan Crocker as the new US ambassador to Iraq. Many of these new figures had been critical of the Iraq War from the start, and were deeply sceptical of the Wilsonian idealism that underpinned it. The neoconservatives were out; the realists were in.

Most crucial of all was the appointment of Petraeus, a scholarly military officer who nevertheless inspires enthusiastic support from troops under his command. Though public opinion had turned against the war and Congress was agitating for withdrawal, Petraeus convinced Bush that a "surge" of new troops, accompanied a radical change of strategy, offered the best chance of success. Petraeus adopted the classic counter-insurgency approach - largely forgotten by the US Army in the post-Vietnam era - of concentrating on securing the civilian population rather than fighting the insurgents. He gained a foothold in violent areas by flooding them with troops, provoking sharp battles in Baquba and Baghdad. Once those initial battles were over, he did not simply withdraw to fortified bases, as previous American commanders had done. Instead he deployed small groups of US troops in neighbourhoods, allowing them to mingle with the civilian population and prevent the insurgents from returning.

The new strategy put American troops at greater risk, which lead to sharp upturn in combat casualties during the first six months of the surge. But there were also early successes. In the province of Anbar, once the focal point of the Sunni insurgency, local Sunni tribes dramatically severed their ties with Al-Qaeda in Iraq (AQI) and formed an alliance with local American forces. AQI had alienated the tribes with their attempts to marry local women, enforce Islamic law, and inflict brutal punishments for smoking in the territories it controlled. Whatever the reasons for it, the alliance between the US military and the Sunnis proved to be very strong, and rapidly lead to the expulsion of AQI from the region. Anbar went from being the most dangerous province in Iraq to one of the most peaceful in a matter of months.

Strictly speaking, the success in Anbar can not be directly attributed to the surge. It stemmed from a variety of factors, not least of which was luck. However, Petraeus soon began looking for ways to copy the "Anbar model" in other parts of Iraq. In Baghdad, the US military enlisted the help of civilian volunteers to fight common enemies such as Al-Qaeda and the Mahdi Army, the militant Shiite group commanded by Moqtada al-Sadr. There are currently almost 70,000 of these volunteers countrywide. They volunteers are officially known as "Concerned Local Citizens," and are given salaries and uniforms by the Americans. In reality, many of them are former insurgents, which makes the Shiite government nervous and detracts from the Manichean moral narrative preferred by Bush's speechwriters. Keeping former insurgents on the payroll is, however, an effective way of taking them off the street while simultaneously ensuring that new militants do not rise up to take their place. This in turn has created an impetus for further positive developments. Most notably, the collapse of the Sunni insurgency has robbed the Mahdi Army of its popular legitimacy, forcing Moqtada al-Sadr to stand down his forces while he tries to reassert control over the movement.

This is the situation as it stands now. Through a combination of the surge, war-weariness among Iraq's ethnic factions, and simple luck, something approximating a ceasefire has been achieved. The question is whether Iraqi leaders are capable of capitalising on the situation and arriving at a political settlement. To date, most of the "benchmarks" put in place by the Bush Administration to measure political progress are unfulfilled. The two most important political initiatives - an agreement for sharing oil revenues, and re-integrating former members of the Baath Party into Iraqi civil society - are stalled in the Iraqi legislature. And any security gain predicated solely upon the presence of US forces has a built-in expiry date. Sooner or later, political pressure for a US withdrawal will become inescapable, regardless of who succeeds Bush next November.

Despite these caveats, the fact remains that for the first time in years, there are genuine reasons to be positive about the future of Iraq. For the sake of Iraq's civilian population, who have spent the last five years being brutalised by war, and decades before that under the harsh rule of Saddam Hussein, this is something to be welcomed.

Laurence Caromba

This article was first published by the Centre for International Political Studies (CiPS), University of Pretoria. Reproduced by permission.