Hague court says it is serious about gender-based crimes. But it is yet to secure a conviction.
Ever since the International Criminal Court (ICC) began bringing suspected war criminals to The Hague, it has been criticised for not taking crimes involving sexual violence seriously enough.
Ahead of a four-day international summit organised by the British government to tackle sexual violence in conflict, the ICC's prosecutor, Fatou Bensouda, has published a new policy paper on how her office will investigate such offences.
The June 10-13 London summit seeks to focus the efforts of governments, the United Nations and civil society towards ending sexual violence and rape in conflict.
Since the ICC's creation in 2002, it has been expected to include crimes of sexual violence in the cases it prosecutes. Under the court's founding Rome Statute, sexual slavery, enforced prostitution, forced pregnancy and enforced sterilisation were set out as prosecutable offences for the first time under international law.
Bensouda's policy paper restates that commitment, even though 12 years into its operations, the court is still to convict anyone of rape or other gender-based crimes.
The ICC's first conviction in March 2012 saw Congolese warlord Thomas Lubanga Dyilo found guilty of enlisting and conscripting child soldiers in the eastern Democratic Republic of Congo (DRC). He has appealed against that judgement.
The charges against him did not include rape or other forms of sexual abuse, even though rights groups documented many such acts in relation to his rebel militia, the Union of Congolese Patriots.
In the early years, the critics of the Office of the Prosecutor (OTP) claimed that it was so keen to get cases under way quickly that this came at the expense of thorough investigations into the full range of crimes that had taken place in a given country.
In the Lubanga case, rights groups said the OTP should have charged the suspect with gender-based crimes.
In an IWPR investigation into prosecutions of sexual violence, a former ICC investigator said the OTP did originally looked at crimes of sexual violence as well as other abuses like mass killings and torture. However, the former investigator said, the OTP chose subsequently to focus its case entirely on the offences of conscripting, enlisting and using child soldiers in hostilities. (The report was ICC Investigative Strategy Under Fire.)
In Uganda, another country where ICC prosecutors have pursued investigations, only two senior commanders of the rebel Lord's Resistance Army have been charged with crimes of sexual violence despite indications that these occurred on a routine basis.
In subsequent cases, the OTP has included the charge of rape in the indictment of former DRC vice-president Jean-Pierre Bemba, who is on trial for war crimes and crimes against humanity committed in the Central African Republic. Sudan's president, Omar al-Bashir, is charged with rape allegedly committed by his forces in Darfur.
Prosecutors have also charged the current president of Kenya, Uhuru Kenyatta, with responsibility for rapes that occurred in the electoral violence which unfolded following a disputed election result in 2007. However, the OTP has faced criticism in Kenya for not providing enough evidence to link the defendant to rapes that occurred in all parts of the country.
Bensouda's policy paper on tackling sexual violence, unveiled in The Hague on June 6, states that the OTP will consider such crimes in all its investigations.
"Within the scope of our mandate we hope to apply a gender analysis to all of the crimes that fall within our jurisdiction," Bensouda told journalists in The Hague at the policy launch. "Sexual and gender-based crimes are some of the most serious crimes [that we deal with] and wherever we find evidence to charge for these crimes, we will do so."
Bensouda emphasised that many of the cases now before the court do include a gender component.
"We have learnt the lessons, and we are building on those lessons to make sure that we are more efficient and effective in the investigation and prosecution of these crimes," she said.
Sceptical observers question whether this renewed will to act will translate into convictions for acts of sexual violence. Much will depend on the level of evidence that OTP investigators can collect and how its prosecutors present the charges to judges.
FRAMING THE CHARGES
The most obvious failure of the OTP to secure a conviction for gender-based crimes was in the case against rebel commanders Mathieu Ngudjolo and Germain Katanga for a 2003 attack on the village of Bogoro in eastern DRC.
Both Ngudjolo and Katanga were charged with crimes of sexual violence, including rape and sexual slavery. They were initially charged together, but the case was subsequently split into two.
Ngudjolo was acquitted of all charges in December 2012. In March this year, judges convicted Katanga of war crimes including pillage and murder - but they acquitted him of all charges relating to sexual violence.
Judges concluded unanimously that crimes of sexual violence were not part of the common purpose of the attack on Bogoro.
So what went wrong?
During the Katanga trial, prosecutors tried to show that rape and sexual violence - as well as pillaging, looting and murder - took place as a result of a common plan directed by the defendant.
They failed to do so, largely because the evidence linking Katanga to crimes of a sexual nature was deemed insufficient.
Clair Duffy, a senior legal advisor at the International Bar Association's ICC programme, based in The Hague, believes that prosecutors did not place enough emphasis on sexual violence as a method that Katanga employed in the attack on the village.
"Part of the problem is how the charges are framed," she said. "Sometimes, that plan can be framed too tightly, so conceptually the judges consider killing as the main tool used to wipe people out. If the prosecution states that there was a common plan to wipe out a village, then this needs to be framed widely enough to include all kinds of criminal acts that were engaged in to meet that objective."
Duffy argues the importance of paying particular attention to sexual violence, because it could be all too easy for the judges to see such crimes as an unfortunate by-product of conflict, rather than as something orchestrated and planned.
Other legal experts have told IWPR that as a legal point, the question of whether sexual violence is a component of a planned attack or simply a consequence of armed violence is an unfortunate distinction to draw.
Mariana Goetz, deputy director at Redress, a UK-based human rights organisation, points out that in both scenarios, sexual violence in conflict is against international law.
"It is not terribly helpful to make a distinction between sexual violence as a weapon of war and sexual violence that is a by-product of war, because this appears to diminish the severity of the crime when it happens in conflict but is not necessarily part of a wider military plan," she said. "Under humanitarian law, sexual violence in conflict is still a war crime and just as heinous, even if it is not used as a weapon of war."
In the Katanga case, Goetz argues that the OTP could have selected a different mode of legal liability for the charges, one that better reflected the reality of the crimes, and that might well have led to a conviction. Rather than charging cases of sexual violence as part of a planned attack, prosecutors could have charged them under the legal framework known as "command responsibility". Under that form of liability, prosecutors would have had to show that Katanga knew about the crimes but did not take steps to prevent them happening or punish his subordinates for committing them. They would not have had to demonstrate that crimes of sexual violence were a pre-planned tactic.
Goetz said prosecutors could have tried to show that "although such acts were not part of the militia's planned criminal conduct, Katanga should have been in control of his troops, and as their leader would be responsible for conduct that he should have foreseen or known about."
APPLYING BROADER CHARGES
The OTP says that in future cases, it plans to do this and prosecute sexual violence under broader modes of liability, including command responsibility.
The new policy paper states the OTP will increasingly aim to hold military leaders "accountable not only where they intended the specific conduct or consequence of sexual and gender-based crimes, but also where they knew or should have known about... the commission of such crimes and failed to take all necessary [measures]".
There are some important test cases in the pipeline.
In June 2013, judges declined to confirm charges against the former president of Ivory Coast, Laurent Gbagbo, on the grounds that there was not enough evidence to show a sufficient link between Gbagbo and the national armed forces. They instructed the prosecution to come up with more evidence.
The charges against Gbagbo include allegations of rape and sexual violence committed by the armed forces.
As well as submitting additional evidence, the prosecution responded by extending the mode of liability so that Gbagbo could also be prosecuted for not being in full command of his troops.
The OTP has also used this mode of liability against Congolese militia leader Bosco Ntganda as well as in the Bemba case.
Alex Whiting, a professor at Harvard Law School, believes that one of the challenges facing ICC prosecutors is that pre-trial judges have required them to charge suspects under a single mode of liability, rather than several.
Judges have subsequently used their powers to change modes of liability in order to more accurately reflect the nature of the crimes. In the Katanga case, judges made such a change, but not to the broader "command responsibility" mode that many argue could have secured a conviction.
This approach has been controversial. In the Katanga case, both the defence and one of the judges argued that making a change of this kind violated the rights of the defendant. (Congolese Rebel's Conviction Only a Qualified Success has more on the trial.)
As a result of this controversy, Whiting argues that judges will now be more open to the OTP presenting charges containing multiple modes of liability at the start of a case.
"That [changing modes of liability during a case] hasn't really worked out, so now pre-trial chambers will be much more open to allowing alternative modes of liability from the very beginning," Whiting said. "I think it is in everybody's interests - the prosecution, the defence, the judges - to allow for alternative modes of liability from the beginning, and then let the case unfold that way."
FINDING THE EVIDENCE
While charging sexual crimes under different modes of liability could bring the OTP some success, legal experts add a note of caution, pointing out that securing a conviction mainly comes down to gathering strong evidence.
In the Kenyatta case, judges forced prosecutors to drop rape charges in relation to attacks in two locations, Naivasha and Kibera, because they did not have enough evidence to link him directly to the crimes. The judges ruled that crimes could not be attributed to the accused "under any mode of liability".
The problem with crimes of a sexual nature is that it is often hard to gather evidence, or that the abuse comes to investigators' attention too late.
Goetz from Redress says the OTP should be working harder to uncover evidence of sexual and gender-based crimes. She believes that currently, too much is left to chance.
"The first challenge in successfully prosecuting crimes of sexual violence is to go beyond the mindset that you expect to find the evidence waiting for you like a corpse in a murder case," Goetz said. "There are many reasons why people might not want to come forward. And even when they do, they might tell you about all the other things that were happening - the murder of their husband, the looting of their home - without talking about sexual violence."
Prosecutor Bensouda has pledged that all future investigations will pay more attention to the possibility that sexual abuse took place. She will also consider the kind of evidence she presents.
In the past, the OTP has placed a lot of emphasis on witness testimony, a strategy that the prosecutor accepts has limitations.
The OTP's investigations in Kenya have been hamstrung by the intimidation of witnesses. Sources close to the court have cited this as the reason why rape charges were not brought in another case, against Kenyan deputy president William Ruto and former broadcaster Joshua Arap Sang. (See ICC Kenya Probe "Hampered by Intimidation".)
Following the difficulties it faced in its Kenya investigations, the OTP released a new investigations strategy last October which focused on going beyond witness testimony. Speaking in The Hague last week, Bensouda said the renewed focus on sexual and gender-based crimes was part of that approach.
"We have decided to change that strategy by not only looking at witness statements but also documentary evidence such as hospital records and using forensic investigation strategies as a new way of collecting the evidence that we need," she said.
Bensouda hopes that refocusing investigations in this way will help bring about successful prosecutions when cases come to trial.
There will still be obstacles, though. In particular, victims are often reluctant to report crimes of sexual violence, which means that the physical signs of abuse can disappear before investigators begin their work.
"The sooner the survivor of a crime comes forward, the easier it is to collect evidence," said Karen Naimer, director of the programme on sexual violence in conflict zones at Physicians for Human Rights, a United States-based NGO. "But victims of rape and sexual violence are often stigmatised by their community, and so may only come forward much later, if they come forward at all."
Gathering the right evidence can help in cases where a direct link can be made between the accused and the crimes that took place on the ground, by demonstrating a systematic pattern to attacks. But under the command responsibility framework, it may not always be easy to link crimes to perpetrators.
"It will generally be easier for the prosecution to mount a case where there are positive acts by an individual," Duffy said. "That might be his physical proximity to the commission of the crimes - for example, if he's sighted at a place where rapes have been openly committed. It could also be a specific utterance from the accused, either encouraging the commission of rape or directly ordering it.
"The real issue comes in many cases where you don't have either of those things, but you still have the commission of rape by those who are subordinate to the accused or acting together with him."
Blake Evans-Pritchard is an IWPR contributor in The Hague. Simon Jennings, IWPR's Africa Editor, also contributed to this report.