On July 11, 2017, Amnesty International published “At Any Cost: The Civilian Catastrophe in West Mosul, Iraq,” alleging “repeated violations of international humanitarian law, some of which may amount to war crimes” by both ISIS and “Iraqi government and US-led coalition forces” in the first half of 2017 (pg. 5). Even if “IS tactics and violations created particular challenges,” “Iraqi government and US-led coalition forces failed to adequately adapt their tactics to these challenges – as required by international humanitarian law” (pg. 6).

According to Amnesty, pro-government forces must “Assume the presence of civilians in every structure when engaging IS fighters” (pg. 46). Amnesty demands an adjustment in tactics, including restrictions on “the use of explosive weapons with wide area effects – including artillery, mortars and IRAMs” (pg. 7). Based on the discussion of specific incidents (more on this below), Amnesty seems to rule out all airstrikes.

In other words, if pro-government forces in Iraq had adhered to Amnesty’s version of the laws of armed conflict (LOAC), they could not have effectively fought in west Mosul. Instead, Amnesty’s rules would force a slow, brutal, bloody ground campaign, which would clearly increase civilian casualties and coalition troop losses.

For years, especially in publications on the 2009 and 2014 Gaza conflicts, Amnesty has been advancing a similar argument regarding Israeli operations in urban zones, albeit with much more condemnatory language than it uses here. Because Hamas in Gaza and Hezbollah in Lebanon deliberately co-locate their forces in heavily populated areas, in particular in mosques, hospitals, and family residences, Israel must fight with one hand tied behind its back. All Israeli strikes are automatically labeled “disproportionate” and “indiscriminate.” Precautions implemented far and above legal obligations by Israel to warn civilians, such as leaflets, SMS messages, and “knock on the roof,” are twisted by Amnesty into an indication of Israel’s criminality (and Hamas’ protection of civilians).

Moral and Legal Inversion

Amnesty’s absurd approach inverts morality and distorts LOAC beyond recognition: it rewards deliberate violators of international law (ISIS in Iraq, Hamas in Gaza) with no respect whatsoever for civilian life as if they operate within the bounds of rule of law. Amnesty therefore shifts the moral and legal obligation entirely to the one party that considers itself constrained by LOAC and civility.

Amnesty also erases the carefully developed LOAC principles by applying an amateurish “effects-based” analysis – a standard that is not applied by courts, militaries, or legal experts.

As explained by Charlie Dunlap (former USAF JAG) in a post on Lawfire: “Amnesty International goes too far in concluding that simply because civilian casualties occurred, even in significant numbers, that the law of war was necessarily violated. As Amnesty International should know, the law – even in this context – carries a presumption of innocence and typically demands evidence of the information and resources reasonably available to the commander (or other attacker) prior to the attack, as well his or her intent at that point, before ascribing criminal liability. I didn’t see much of that in the report” (emphasis in original). Crucially, “Amnesty International may think pro-government forces did not take all feasible precautions, but they are in no real position to say what was or was not, from the commanders’ perspective, feasible under the complex circumstances of this offensive.” (Amnesty’s distortion of the law, as reflected in Dunlap’s assessment, is even more pronounced in its publications on Israel – as documented by NGO Monitor [here, here, here, here, here, here, here].)

An illustrative example in the Mosul report is Amnesty’s reference to a March 17 incident where a strike against two snipers triggered “significant amount of secondary explosive materials in the structure,” which ISIS had “deliberately staged” there. Because over 100 civilians died in the resulting explosions, Amnesty suggests that even though “pro-government forces appear to have struck their intended military target, their use of unsuitable weapons or failure to take other necessary precautions resulted in needless loss of civilian lives” (pg. 25). Yet, Amnesty has no idea what precautions were used, what intelligence was available to the pro-government forces prior to the strike, or if there was a technical malfunction. Indeed, a military investigation revealed that pro-government forces did not know about the explosive materials and the secondary explosions were unexpected, and hence, the strike was tragic, but not illegal. Under the law, it is ISIS that bears responsibility for the civilian casualties for deliberately embedding weaponry in a civilian area that they knew was likely to cause significant harm when they drew fire to their position.

Elsewhere, Amnesty accuses pro-government forces of an indiscriminate attack that killed three civilians but no ISIS members (pg. 28), yet quotes eyewitnesses who say there were combatants in the vicinity. In another instance, Amnesty complains about a strike where the (legitimate) target left the building, but provides no evidence that pro-government forces should have known this.

East vs. West: Miscalculating Proportionality

As with Gaza, Amnesty obfuscates the meaning of “proportionality.” According to LOAC, a strike is illegal if it is “expected to cause” civilian damage “which would be excessive” to the anticipated “military advantage.” As paraphrased by Dunlap, “the law anticipates and, indeed, tolerates civilian losses in combat so long as they are not excessive in relation to the concrete and direct military advantage anticipated.”

As we’ve seen, Amnesty proceeds from a colloquial meaning that erases the concepts of “expected” and “anticipated.”

There are other factors at play in the proportionality calculus, including some that Amnesty presents in Section 3 of its report and then proceeds to ignore in Section 4 when accusing pro-government forces of war crimes.

According to the report, “The use of artillery, mortars and particularly IRAMs (Improvised Rocket Assisted Munitions),” which Amnesty believes to be illegal in urban combat situations, “did not feature as heavily in east Mosul as it did in the west Mosul operation.”

Why not?

First of all, “The biggest factor affecting the tactical change from east to west Mosul appears to have been lack of available manpower. The best trained and equipped fighting force available for urban warfare in Mosul were Iraqi Special Operations Forces (ISOP) I and II….Having been so severely depleted by unexpected levels of IS resistance in east Mosul, the Iraqi military had to rely on other fighting forces in its rush to retake west Mosul in the first half of 2017… Much of the shortfall was made up by the Federal Police, whose members are believed to have received [only] between two and six weeks of pre-deployment training.” (pg. 13, emphasis added).

In other words, the pro-government forces did not have the necessary elite fighters for combat in west Mosul. Why? Because, the “troops spearheading the east Mosul operation were expected to fight their way into IS-controlled neighbourhoods without reliance on close air support,” precisely the type of tactics that Amnesty advocates. Therefore, the elite troops “paid a heavy price for this tactic… deaths and injuries, in east Mosul range from 4,000 to 6,000 of a total fighting force of 8,000 troops” (pg. 11).

The significant loss of troops and unexpected ISIS tactics are exactly the type of complexities that factor into the proportionality analysis as the battle shifted to west Mosul. Yet, Amnesty ignores them in its allegations of “war crimes.” Amnesty’s faulty conclusions also demonstrate its fundamental lack of expertise. While Amnesty appears to recognize operational realities (such as the mass loss of troops in east Mosul), it is unable to process how dynamic events on the ground impact proportionality and tactical assessments (the loss of troops necessitated new battle tactics in west Mosul and changed the proportionality calculus).

Moreover, Amnesty erases the fact that a far greater number of civilians were protected by the actions of pro-government forces than would have been had they restricted themselves to Amnesty’s rules. As Dunlap notes:

Liberating a city like Mosul, which was once home to over 650,000 people, from the horrors of ISIS is vitally important ‘military advantage anticipated,’ especially to those civilians still trapped there.

People may differ about what is or is not “excessive” in a given instance, but I am sad to say that under the circumstances, the deaths of the 3,706 civilians (between 19 February and 19 June 2017) which Amnesty International cites do not strike me as inarguably excessive in light of the priority that must be given to save what could be many tens or even hundreds of thousands of civilians who were likely still in Mosul under ISIS’s thumb at the beginning of the campaign.

In Mosul pro-government forces faced a situation in which any delay in their advance would leave civilians in the grip of ISIS.” (emphasis added)

What Starts in Israel Doesn’t End in Israel

As with its reports on Israel, Amnesty’s publication on Mosul is legally and morally flawed. Although its condemnations of Israel are much harsher in tone, Amnesty’s Mosul claims demonstrate that outlandish NGO campaigns targeting Israel do not stay with Israel. These NGOs invariably also target and try to limit the ability of Western democracies to combat asymmetric warfare.

Ultimately, Amnesty weakens the framework of international law, perverts morality beyond recognition, and, in the end, will lead to far greater civilian harm and never-ending conflict.