The war in Afghanistan is intensifying, especially in the southern province of Helmand where western coalition forces are attempting to take the fight to the Taliban. The inevitable result is an increase in deaths and injuries (often disabling ones) among British, American and other national contingents.

Martin Shaw is professor of international relations and politics at the University of Sussex. A historical sociologist of war and global politics, his books include War and Genocide (Polity, 2003), The New Western Way of War (Polity, 2005), and What is Genocide? (Polity, 2007). He is editor of the global site

Also by Martin Shaw in openDemocracy:

"The myth of progressive war" (11 October 2006)

"Genocide: rethinking the concept" (1 February 2007)

"The International Court of Justice: Serbia, Bosnia, and genocide" (28 February 2007)

"The genocide file: reply to Anthony Dworkin"  (6 March 2007)

"My Lai to Haditha: war, massacre and justice" (16 March 2008)

"Israel's politics of war" (19 January 2009)

"Uses of genocide: Kenya, Georgia, Israel, Sri Lanka" (9 February 2009)

"Sudan, the ICC and genocide: a fateful decision" (11 March 2009)

"The Kosovo war: between two eras" (31 March 2009)

"A century of genocide, 1915-2009" (23 April 2009)

"The trouble with guns: Sri Lanka, South Africa, Ireland" (10 June 2009)

"Sri Lanka - camps, media...genocide?" (30 June 2009)

The British death-toll since the Afghanistan war began in October 2001 is (as of 23 July 2009) 188, now more than the 179 killed in Iraq. The mid-2009 period has been especially painful; nineteen died in the first three weeks of July, and in the third of these weeks there were a shocking 150 serous injuries. The United States military forces too have endured great losses: 742 US soldiers have died in Afghanistan, while in Iraq - where around 133,000 US troops are still stationed pending a phased withdrawal by the end of 2011 - there have been to date 4,327 fatalities.

The routine media focus on individual deaths and on these overall figures, however, tends to eclipse the fact that by far the largest number of casualties in both war-zones have been civilians; and that - in contrast to the existence of reliable statistics on the western coalition's death-tolls - there is no official count of civilian deaths and injuries in Iraq and Afghanistan (or indeed in Pakistan, where US air-strikes - often by pilotless "drones" - have also caused many civilian deaths) (see Paul Rogers, "Drone wars", 16 April 2009).

The true numbers, then, can be only roughly estimated - from media reports, hospital and morgue data, and household surveys. This also means that they become (over Iraq in particular) matters of dispute, even among fellow critics of the United States invasion and occupation who share the same humanitarian concern to record the fate of its victims.

The "numbers-game" and its associated tensions are a perennial feature of the reporting and analysis of wars, genocides, massacres, and natural disasters (see Jean Seaton, "The numbers game: death, media, and the public", 6 October 2005). But behind the arguments are vital questions of responsibility: to be accurate and to establish the truth of what happened, and to create a foundation of evidence for political and moral judgment.

It clearly matters whether around 100,000 civilians died in Iraq as a result of the war in 2003-09 (as Iraq Body Count [IBC] records), or whether the real figure from 2003-07 was already about 600,000 [as indicated by the study of the Johns Hopkins Bloomberg School of Public Health, published in The Lancet). The higher figure involves serious implausibilities, as IBC's critique suggests (for a careful and serious consideration of these issues, see two openDemocracy articles by Michel Thieren: " Deaths in Iraq: how many, and why it matters" [18 October 2006] and "Deaths in Iraq: the numbers game, revisited" [11 January 2008]).

In any event, and independently of particular contexts, a cautious attitude to numbers is generally advised in these matters. The true death-toll incurred by major events tends to vary greatly from initial estimates: the 11 September 2001 attacks are an example, with an apparently convincing early number of 7,000 victims fell to 2,974 when a proper count was possible. But the lower figure does nothing to diminish the horror of the event - and this is equally true of Iraq, where no one can doubt that there has been an appallingly high loss of life as a direct and indirect result of the US-led invasion. It can be said with near-certainty that at least a 100,000 civilians in Iraq have died, directly and indirectly, as a result of armed conflict since 2003; and perhaps 20,000 in Afghanistan since 2001.

The human impact

In Iraq, the tolls of death and injury on all sides have in 2008-09 been falling from the catastrophically high levels of 2006-07. At the same time, many are still dying from car-bombs and local incidents; and a new International Crisis Group report highlights the explosive potential of the ethnicised political conflict over Kirkuk (see "Iraq and the Kurds: Trouble Along the Trigger Line", 8 July 2009).

In Afghanistan, a rising trend of violence was evident even before the recent escalation of the war. Human Rights Watch has estimated that US and Nato attacks in January-July 2008 roughly equalled the number of British soldiers who had died since 2001 (see "Afghanistan: Civilian Deaths From Airstrikes", 7 September 2008).

But "mere" body-counting is not the only proper measure in such discussions. The  human impact of war is inacalculable. Many more soldiers and civilians are injured than killed, and even more suffer mental trauma; civilians very rarely are guaranteed the high standards of medical care that western soldiers receive. Indeed, the civilian experience of war is of a huge landscape of harm that includes severe disruption across many areas of life - including displacement, forced migration, loss of employment, family separation, and material impoverishment.

Another figure reveals the scope of the harm in Iraq even more than the body-count: 4.8 million - the United Nations refugee agency (UNHCR's) estimate of the combined number of Iraqi refugees (1.9 million); internally-displaced people (IDPs) (2.7 million); and returned refugees (0.2 million). True, over a million of the displaced result from Saddam Hussein's campaigns, but the vast majority have been forced from their homes since 2003. Moreover, it is misleading to attribute displacement simply to "the war"; for the invasion and subsequent fighting probably cost "only" around 200,000 Iraqis their homes. The vast majority have lost their homes due to deliberate policies of exclusion: they include the Kurds forced out by Saddam's Arabisation programme (before 2003), the Shi'a attacked by al-Qaida in Iraq and other Sunni-based militia, and the Sunnis targeted by Shi'a armed groups (since 2003).

This is the major difference with Afghanistan, where both death-tolls and the numbers of IDPs (230,000) are much lower than in Iraq. There are also huge numbers of Afghan refugees, 3.3 million, though most left the country as a result of the pre-2001 wars.

Paul Rogers's weekly column has tracked the Afghan war since September 2001. Among the recent articles:

"Afghanistan's Vietnam portent" (17 April 2008)

"Afghanistan in an amorphous war" (19 June 2008)

"Afghanistan: state of siege" (10 July 2008)

"Afghanistan: on the cliff-edge" (28 August 2008)

"Afghanistan: the dynamic and the risk" (9 October 2008)

"Iraq's gift to Afghanistan" (20 November 2008)

"Afghanistan's twisting path" (9 July 2009)

"Afghanistan's lost decade" (16 July 2009)

The Barack Obama doctrine, echoed by Gordon Brown and his government colleagues in Britain, sees Afghanistan as the "good" war and the Taliban as the "evil", terrorist (and with reference to al-Qaida, terrorist-hosting) enemy. Yet for all the brutality of the Taliban pre-2001 rule - including massacres of the Hazara people - their new insurgency has not been accompanied by the campaigns targeting certain population groups which characterised post-2003 Iraq.

The three-sided conflict

In fact what is referred to as "the" war in Iraq was more the concatenation of three main lines of conflict: the United States-led coalition (including the Iraqi government) vs the Sunni-based "resistance" and  at times Shi'a-based armed groups; Sunni-based vs Shi'a-based armed groups; and each of these sets of groups vs their respective "enemy" civilian populations. The militias' conflict can be seen as a low-grade civil war, but its worst feature is that it has involved the third dimension - violence against civilians belonging to the "other" side.

This began with attacks on Shi'a religious commemorations in 2004, and escalated by 2006-07 into mutual attempts by each set of armed groups to expel "other" populations from the areas they controlled. This violence, involving targeted killing which terrorised populations into flight, brought a genocidal dimension to the conflict in Iraq. Indeed, most of the deaths and injuries in the country have resulted from this targeted militia violence. Even The Lancet study suggested that by 2005-06 coalition violence accounted for only around a quarter of civilian deaths (and in turn around half of these were produced by air-strikes).

At a conference of the International Network of Genocide Scholars in Sheffield in January 2009, I was approached by an American student who had served as a soldier in Iraq; he was distressed by a suggestion made in one of the sessions that the United States was responsible for genocide in that country. My response was to assure him that - whatever war crimes the US had committed in invading Iraq and exposing its civilians to fallout from air-strikes and other violence - there was no reason to think that the US had aimed to destroy the Iraqi population or any group within it, which would have constituted genocide.

But this answer does not exhaust the question of US responsibility for genocide in Iraq. In invading Iraq, the George W Bush administration provoked a bitter conflict among Iraqi armed as well as political groups over long-term control of the Iraqi state. The evolving armed resistance based on the minority Sunni population - elements of which had benefited from the Saddam Hussein regime, and which more broadly saw itself as a loser when elections gave power to parties of the Shi'a majority - was always likely to be directed against the majority Shi'a population as well as the occupier. The background of the genocidal record of Saddam's Ba'athism created an ever-present potential for murderous revenge and further violence; it was reinforced in the post-2003 resistance by al-Qaidism with its murderous attitudes to Christians, Jews and Shi'a Muslims. All this makes clear that new genocidal violence against the non-Sunni population was an inherent danger of the post-invasion conflict in Iraq.

At an early stage of the Iraq conflict, I suggested that the key problem of the "new western way of war" was how it exposed the civilian population in the war-zones to much greater risks than those taken by western soldiers (let alone their political leaders). The moral deficiency of this "risk-transfer war", I argued, could be seen in the preference for high-altitude air-strikes which kept air-crews safe while inevitably killing civilians on the ground; and in the "shoot-first, ask-later" approach of American soldiers at roadblocks (see The New Western Way of War: Risk-Transfer War and its Crisis in Iraq [Polity, 2005]) .

Now, six years on, it is clear that this was only the beginning of the risks to which the US invasion exposed civilians. Even greater in its potential for harm was the murderous civil war that the invasion opened up, and which for several years the US seemed powerless to control. Nor was this an entirely unpredictable consequence: on the contrary, the post-2003 "resistance" attacks on the Shi'a population repeated the Saddam regime's brutal repression of the Shi'a and Kurdish insurrections, provoked by the US victory in the Gulf war of 1991.

Thus the dangers of western military interventions for civilians include not only the direct harm to which they are exposed by US bombing (the chief complaint of Afghans today). They also include the indirect risk of provoking genocidal violence. So far, the new Afghan insurgency, based mainly in the Pashtun-majority areas from which the Taliban draw support, has not had this outcome: but if it succeeds in extending the war into other parts of Afghanistan, these issues could well return.

It is also, as suggested above, too soon to conclude that the anti-group violence has subsided in Iraq. The United States and British governments bear huge responsibilities for the dangers to which civilians remain exposed there and in Afghanistan, as well as for the plight of millions displaced by the violence of these years. As the conflict and insecurity in both lands continue, the primary agents of the "new western way of war" must be made to look their victims in the face.

Among openDemocracy's articles on Afghanistan's war:

Marco Niada, "Afghanistan: no time to lose" (21 April 2004)

Hamish Nixon, "Afghanistan's election world" (13 September 2005)

Irfan Husain, "Kabul vs Islamabad: a war of words" (16 March 2006)

Marco Niada, "Afghanistan: the last chance" (12 July 2006)

Fred Halliday, "The Malvinas and Afghanistan: unburied ghosts" (3 May 2007)

Antonio Giustozzi, "The resurgence of the neo-Taliban" (15 December 2007)

Kanchan Lakshman, "India in Afghanistan: a presence under pressure" (11 July 2008)

Shaun Gregory, "The Pakistan army and the Afghanistan war" (25 November 2008)

Antonio Giustozzi, "The neo-Taliban: a year on" (11 December 2008)

Mariano Aguirre, "Barack Obama and Afghanistan: a closer look" (8 April 2009)

Plus regular reports on terrorism.openDemocracy