Individual Compensation Reloaded: German Liability for Unlawful Acts in Bello

Written by

On 30 April, the Appeals Court of Cologne will rule on whether Germany has to pay compensation to victims of an airstrike in Afghanistan. Its judgment is likely to consolidate the new German approach to questions of compensation for armed activities which – given the increasing relevance of litigation about armed conflicts – merits a brief treatment.


In 2009, a German colonel ordered an airstrike against two fuel trucks that were stuck on a sandbank near the NATO camp in Kunduz/Afghanistan. Due to the tense situation in Kunduz, he assumed that the fuel or the trucks could be used for a bomb against ISAF units and thus represented an imminent threat. The airstrike caused the death of 142 individuals. Because many among the victims were civilians, it has become the most controversial modern operation involving the German Armed Forces (leading, amongst other things, to the resignation of a minister of government, criminal investigations and the establishment of a parliamentary investigation).

Seeking compensation for damages on the basis of domestic rules of governmental liability (Amtshaftung), victims filed a claim against the Federal Republic of Germany. In 2013, the Court of First Instance in Bonn rejected the claim (for details see my article in the JICJ). Although it held that governmental liability in principle applies to acts in bello, the Court concluded that the colonel did not breach his official duty to comply with international humanitarian law. A press release, summarizing the oral proceedings and the taking of evidence issued in March, indicates that the Cologne Appeals Court intends to uphold the result of the Court of First Instance.

As I have argued elsewhere (see JICJ article, at p 631-633), the legal assessment made by the Court of First Instance is questionable in several respects. Most importantly, it seems that the colonel did not comply with the customary rule encompassed in Art. 57 (2), a (i) AP I. He failed to do everything feasible to verify that the objectives of the attack were neither civilians nor civilian objects. Certainly, the level of precaution necessary depends on the specific circumstances of the attack. However, in this case the fact that trucks had been stuck for seven hours, and thus did not represent an imminent threat, was not sufficiently taken into account. The adoption of the first instance court’s assessment by the Court of Appeals would therefore be problematic.

While the two courts’ interpretation and application of rules of international humanitarian law is highly fact-dependent, a preliminary aspect is of more general relevance, and highlights the particular approach obtaining under German law: on what basis can Germany be held responsible, before domestic courts, for alleged violations of international humanitarian law?

Governmental liability in context

Under Section 839(2) of the Civil Code in conjunction with Article 34 of the German Basic Law German public authorities can be sued for damages caused by the violation of an official duty. However, only recently have German courts accepted that state conduct during armed conflicts can form the basis of claims for governmental liability under domestic German law. Matters are not entirely clear, though and much depends on the law applicable at the time of the alleged violation. For that reason, claims of Second World War victims have failed before German courts, while claims of war victims who suffered harm after the Reunification are more likely to succeed.

Germany has passed a number of domestic laws relating to the compensation of victims of the Nazi regime. However, claims of Auschwitz internees, former Italian military internees, and survivors of a massacre by a SS-unit in Distomo (Greece) who were not eligible for compensation under these laws, always failed. Neither the rules under international humanitarian law, nor the German law of governmental liability in force during the Third Reich were interpreted at the time of the violation (§ 56-69) as providing for an individual right against unlawful sovereign acts in bello. In its Distomo ruling of 2006, the Federal Constitutional Court did not decide definitely whether such a reading could be reconciled with the need for effective compliance with IHL. It left this question open because, in its assessment, the Greek victims that had brought proceedings could not invoke rules of governmental liability against Germany. At the time of the violation, liability was contingent on a reciprocal commitment on behalf of the victim’s home state which Greece had not made. From a victim’s perspective, attempts to obtain compensation for Nazi injustice thus can hardly be said to have been successful.

In adjudging on damages claims based on conduct during more recent armed conflicts, German courts have taken a different approach. Victims of a NATO airstrike on a bridge in Varvarin (Kosovo) sued Germany for simply accepting that the bridge was a valid military target for NATO, even though no member of the German Armed Forces participated in the actual airstrike. At all levels of jurisdiction these claims were dismissed. However, in 2005, the Appeals Court in Cologne accepted in principle that state conduct during armed conflicts could form the basis of claims for governmental liability (§ 107-122). The court held that, in order to be effective, the law of war needs to be backed up by an enforcement mechanism that sanctions breaches. In reaching this result (which for German courts was a decisive watershed) the Appeals Court in the Varvarin case drew on general considerations about the interaction between international law and the German legal order. It noted that the provisions of the German Basic law, which places human dignity and fundamental rights at its centre, govern the conduct of state agents abroad. International legal obligations binding upon Germany – including those deriving from international humanitarian law – permeate the domestic legal order. Under Art. 25 of the Basic Law international customary law is directly applicable within the German legal order. The increasing protection of the individual through the 1977 Additional Protocols, the (international) prosecution of human rights crimes, as well as the extraterritorial application of European Convention on Human Rights cannot be ignored. All this meant that, in principle, domestic law principles governing governmental liability could apply. None of this eventually was of help for the claimants in the Varvarin case. According to the Appeals Court of Cologne, the Varvarin bombing was not attributable to Germany, hence the Court rejected the claim on its merits. Based on the same legal reasoning, both the Federal Court of Justice and the Federal Constitutional Court ultimately stated that they would not need to rule on the applicability of governmental liability. Instead, the Federal Constitutional Court referred to the on-going debate (§ 59) among German legal scholars regarding the application of governmental liability to acts in bello. By referring to this discussion, the Court allowed for a possible shift in the interpretation of governmental liability law in favour of victims of armed conflicts. This new approach was then taken in the Kunduz proceedings. 


The current litigation about the airstrikes in Kunduz has to be seen against this background. These airstrikes provide German courts with an opportunity to clarify their approach. In its 2013 judgment in the Kunduz case, the Court of First Instance adopted a modern approach by applying governmental liability to acts of state in bello. As is clear from the press release, it is most likely that the upcoming decision of the Appeals Court will further consolidate this trend even if it rejects the claim on its merits. This would indeed complete the opening up of German domestic rules on governmental liability to acts of war.

This state practice certainly involves restrictions on the conduct of the German Armed Forces abroad and the margin of foreign policy. One thus may argue that by taking such an approach, German courts do not act in the interest of Germany if we think of NATO operations and other constellations of shared responsibility. However, other NATO member states like the Netherlands take the same approach. Ultimately, if – such as in Varvarin – German troops do not participate in the concrete organization and execution of an operation, attribution of (mis)conduct is most unlikely. This is because liability requires misconduct of a German state agent (see also Federal Court of Justice in the Varvarin case, § 38-39). It remains to be seen whether the Constitutional Court will ultimately have to rule on the Kunduz case. It then will need to address the question whether the application of governmental liability to acts in bello contravenes the state’s interest in handling foreign affairs.

Print Friendly, PDF & Email


No tags available

Leave a Comment

Comments for this post are closed


Aurel Sari says

April 29, 2015


Thanks for sharing this. In respect of precautions, you write that ‘the fact that trucks had been stuck for seven hours, and thus did not represent an imminent threat, was not sufficiently taken into account.’ I do not understand why you stress imminence in this context and would be grateful for an explanation.

From an AP I precautions perspective, the obligation imposed on Col Klein was to ensure that the targets constituted military objectives within the meaning of Article 52(2) AP I. In so far as the trucks and the fuel are concerned, it seems that they were considered to satisfy the definition of a military objective because their purpose, in other words their intended use against ISAF. Imminence does not play a role here. In so far as the ‘insurgents’ are concerned, it appears that they were considered to either carry out a continuous combat function. Again, imminence plays no role.

From an ROE perspective, it is reasonable to conclude that the hijacking of the tankers constituted a hostile act, which was still ongoing. At the very least, the activities around the tankers would reasonably constitute hostile intent. Again, imminence does not play a role.

Where imminence does come into the picture is the use of two F-15s to carry out the attack. If I’ve got my facts right, it seems that no suitable assets were available to Col Klein to neutralise the tankers. He therefore declared a ‘troops in contact’ incident, which enabled him to call upon close air support by the F-15s to counter an ‘imminent threat’ against friendly forces. In reality, the tankers posed no imminent threat to any friendly forces: the Col was simply a bit creative to get his hands on the necessary assets. However, this is completely irrelevant from the precautions point of view. Similarly, the fact that the F-15 crew expressed doubts whether the planned strike fell within their ROE may well be down to the fact that their ROEs might have been limited to self-defence, whereas the ROE applicable to Col Klein were not. But all of this is an ROE and ‘internal’ matter, not a precuations in attack issue.

I’d be interested to hear your thoughts on this.


Elisabeth V. Henn says

May 1, 2015


Thank you for your valuable remarks.

You are right by saying that RoE are internal matters and that imminence is irrelevant with regard to precautionary measures.

However, the point I want to make is another one. If I am not mistaken, you do not understand why I refer to an imminent threat. The term of imminence is misleading in this context. When saying that trucks ‘did not represent an imminent threat’ I meant that there was no (imminent) time pressure.
The obligation to do everything feasible to verify whether the objectives to be attacked are neither civilians nor civilian objects is context-specific. A bundle of aspects such as weather conditions, stress and time pressure need to be considered. Hence, if an unit is taken under fire the standard of precaution required is lower. If the situation does not change for hours – which was the case in Kunduz(the truck had been stuck) – the standard is higher.


Aurel Sari says

May 1, 2015


I have indeed misunderstood what you had in mind, thank you for clarifying it. However, your clarification raises a different set of issues.

First, you suggest that the ‘standard of precaution’ may be higher or lower, depending on the factual situation. I do not think that is the right way of putting it. The legal standard is exactly the same, irrespective of the situation. It is the measures that need to be taken to satisfy the standard which differ, depending on the prevailing circumstances. This may seem like a semantic or even pedantic quibble, but I think it is an important point. It is not the case that there is an infinite number of legal standards of precaution corresponding to an infinite variety of factual scenarios, but rather a single legal standard requiring whatever action is appropriate in each particular situation.

Second, you seem to suggest that the correct timeframe against which to assess whether or not all feasible precautions were taken is the seven hours during which the tankers were stuck on the sandbank. I disagree. The correct timeframe to look at is the time during which the attack was planned and carried out. This starts when Col Klein entered the operations centre after the location of the tankers was identified and ends with the air strike. According to the information available in the public domain, Col Klein entered the operations centre between 00:15 and 00:25 hours and the strike took place at 01:49. That is significantly less time than seven hours and constitutes the right timeframe for assessing the application of Art 57 AP I.

Third, by ordering the strike to take place at 01:49, Col Klein had to satisfy himself that he took all feasible precautions by that point in time. Of course, delaying the strike in order to take additional precautions was one of the measures available to him to comply with Art 57 AP I. Whether or not taking more time was a feasible course of action depended partly on the level of certainty he had already attained in classifying the targets and partly whether delaying the strike was practical in the light of the military advantage he was seeking to gain. That military advantage was to destroy the tankers. On previous occasions, tankers had been used by the Taliban against friendly forces. Troops under Col Klein’s command had been engaged and suffered casualties the day before the incident. Specific intelligence was said to exist which suggested that the two tankers were to be used in an attack on Col Klein’s position. In the light of all of this, the tankers clearly posed a substantial threat that had to be addressed to protect his forces. Regarding the time element, what matters is not the fact that the tankers had been stuck for several hours, but that significant efforts were made to get them going again, including through the use of additional vehicles. While there could be no certainty from Col Klein’s position when exactly the tankers would be mobile again, he could be more than reasonably certain that they were going to get moving and would do so sooner rather than later. You suggest that ‘there was no (imminent) time pressure’. I disagree: there was a huge amount of time pressure. In this respect, it is also important to bear in mind what means were available to Col Klein to achieve his intended objective. He already lost access to the aircraft which located the tankers and the two F-15s were not going to be available to him forever either.

Fourth, whether or not Col Klein did ‘everything feasible to verify that the objectives to be attacked [were] neither civilians nor civilian objects’ as required by Art 57(2)(a)(i) AP I depends in part, I would suggest, on whether he took advantage of measures that were practically available to him during the relevant time period and which could have materially contributed to increase the level of certainty as to the status of the objectives. The information available in the public domain suggests that some such measures were available, although the reasons given for not adopting them need to be taken into account before arriving at any conclusions.
One of the reasons why I feel that this is such an important case is because there is an unusual amount of information available which enables us to assess the legal aspects of the strike with far greater accuracy, and fairness, than in many other instances. However, we need to be looking at the right facts.