Dutch Court Concludes Anfal Campaign Wasn’t Genocide

by Kevin Jon Heller

A little over a year ago, I noted that a Dutch court had sentenced Frans van Anraat, a chemicals dealer, to 15 years in prison for selling Iraq a key component of the mustard gas that Saddam used against the Kurds during the Anfal campaign. Last week, a Dutch appeals court upheld van Anraat’s conviction for complicity in war crimes, but rejected the prosecution’s claim that he should have been convicted of complicity in genocide instead:

An appeals court increased a Dutch businessman’s war crimes sentence from 15 to 17 years Wednesday for selling Saddam Hussein raw materials for poison gas attacks that killed thousands of Kurds, saying the trader was “driven by naked greed.”

The Hague Appeals Court upheld the complicity in war crimes conviction handed to Frans van Anraat in 2005 for selling chemicals to Saddam and agreed with trial judges that he was innocent of complicity in genocide.

The four-judge appeals panel said they increased the sentence because Van Anraat, 65, was motivated by greed and repeatedly sold chemicals knowing they were being turned into mustard gas. Seventeen years is an unusually long prison term by Dutch standards, where 20 years is the maximum unless a suspect is sentenced to life without possibility of parole, or convicted of a terrorist crime.

The appellate court’s decision to uphold van Anraat’s acquittal on the complicity in genocide charges seems sound, given that the trial court specifically concluded that van Anraat was not aware of Saddam’s plan to solve Iraq’s “Kurdish problem” through genocide. As the ICTR made clear in Akayesu, although complicity in genocide does not require the defendant to share the principal perpetrator’s specific intent to destroy the group as such, it does require proof that the defendant was aware of the genocidal plan itself:

[T]he intent of the accomplice is… to knowingly aid or abet one or more persons to commit the crime of genocide. Therefore… an accomplice to genocide need not necessarily possess the dolus specialis of genocide, namely the specific intent to destroy, in whole or in part, a national, ethnic, racial or religious group, as such. Thus, an accused is liable as an accomplice to genocide if he knowingly aided or abetted or instigated one or more persons in the commission of genocide, while knowing that such a person or persons were committing genocide, even though the accused himself did not have the specific intent to destroy, in whole or in part, a national, ethnical, racial or religious group, as such.

The appellate court, however, did not simply conclude that van Anraat was unaware of Saddam’s genocidal plan. It also held that the Anfal campaign did not amount to genocide:

In its written judgment, the appeals panel said there was insufficient evidence to establish that the poison gas attacks on Kurdish villages in 1987 and 1988 amounted to genocide.

That is, to say the least, a shocking — and indefensible — conclusion. The Anfal trial, which is still in progress, has been no more fair than the Dujail trial. As Human Rights Watch has ably documented, though, there is no question that the Anfal campaign itself, which killed between 50,000 and 100,000 Kurds, was genocidal both in intent and in effect:

The Anfal campaign was the culmination of a long-term strategy to solve what the government saw as its “Kurdish problem.” Since the Ba’ath Party coup in 1968, the Iraqi government had deemed the Kurds as a threat to the nation. Baghdad forced many Kurds to leave their homes and relocated them in the Kurdish “Autonomous Region.” It then “Arabized” the formerly Kurdish areas by enticing Arab tribespeople there with offers of relocation benefits. “Genocide in Iraq” shows that in the mid-1980s the government began to demarcate special areas within the Autonomous Region that it declared to be off-limits. The residents of these “prohibited zones” were, with very minor exceptions, Kurds who after the October 1987 census were defined as non-Iraqi nationals and traitors. In 1988, they were marked for destruction.

The “prohibited zones” were large areas that covered most of rural Iraqi-Kurdistan. The government had lost control of these regions because it had deployed so many troops to fight the war with Iran. By declaring these areas off-limits, Baghdad sought to regain control over them. Although the “prohibited zones” did not include all Iraqi Kurds, they were home almost exclusively to Kurds. The report shows that the vast majority of residents were civilian men, women and children. Many of the villages in the “prohibited zones” were rarely visited by Kurdish guerillas. The Ba’ath Party government simply treated all persons present in the zones alike without distinguishing between combatants and civilians.

… In official pronouncements, national boundaries dissolved and age-old ethnic identities were recycled. Government propaganda alleged that “the Kurds” had allied themselves with “the enemy Persians” against “the Arabs.”

The treatment of those who were loyal to the government was no different, which shows that the official policy was based not on political loyalty but on ethnicity. Even pro-government tribes and/or members of the pro-government Kurdish militia, the National Defense Battalions, were warned that they and their families would not be spared if they chose to remain in their villages in the “prohibited zones.”


The pretext for the policy was the presence of the Kurdish guerilla organizations, which had been using some of these sectors as bases for their insurgency. But what had begun as a counter-insurgency effort against rebels became a murderous campaign against a distinct part of the Kurdish population. “Genocide in Iraq” shows that the repression against the Kurds of the “prohibited zones” escalated from economic embargo to air and artillery bombardments to the systematic slaughter of everyone present.

Interestingly, the Dutch appellate court admitted that “at the very least, there were powerful indications that the Iraqi leaders had genocidal intent.” Why the court still refused to conclude that the Anfal campaign involved a genocidal plan is thus even more mysterious.

If we have any Dutch readers who know more about the appellate court’s decision, I hope they will respond to this post in the comments.

UPDATE: Reader AnneJ has kindly provided the following translation of the critical sentence in the decision:

The Court determines that a final judicial decision on the very important question that is highly relevant from an international perspective, whether certain acts in the indictment by certain people amount to genocide, must be based on a more solid foundation than was presented to the Court.

Concluding that genocide was not adequately proved may be slightly better than concluding that genocide did not take place — but not much better. The appellate court did not have to upset the trial court’s finding that genocide occurred in order to affirm van Anraat’s acquittal on the complicity charges, because the trial court held that, even if genocide did occur, van Anraat was unaware of the genocidal plan. The fact that the appellate court nevertheless went out of its way to contradict the trial court thus seems to indicate that, in fact, it did not believe that genocide took place — not simply that the prosecutor failed to prove that it did (itself questionable, given the trial court’s decision).

My thanks to AnneJ for her help!

UPDATE 2: It has been suggested in the comments that the trial court and the appellate court could have reasonably disagreed about whether the Anfal campaign was genocidal, because “the trial court made the assessment that Anfal was genocide… on the basis of a single report, without examining witnesses, engaging in any sort of comprehensive investigation into Saddam & Co.’s genocidal intent etc.” That description of the trial court’s efforts is incorrect. Although the court did, in fact, take into account a report on the situation in Iraq written by the U.N.’s Special Rapporteur of the Commission on Human Rights — including, it is important to note, specifically considering a number of the original Iraqi documents contained in the report — it also took into account literally dozens of (1) statements made in court by live witnesses, (2) witness statements taken by the court’s examining magistrate, (3) witness statements taking by the Dutch national police, and (4) witness statements taken by Belgian investigating officers. That testimony documented every aspect of the Anfal campaign, including the nature of Kurdishness, the regime’s hatred of the Kurds, and the specific genocidal acts the regime committed during the campaign. I thus stand behind my contention that the appellate court did not simply disagree — reasonably or otherwise — with the trial court about the quality of prosecution’s proof, but actually believed (and so held) that the Anfal campaign did not amount to genocide.

(To add insult to injury, Dutch courts use the lowest standard of proof of any civil-law country in Western Europe: whereas the other civilian countries use either an “intimate conviction” or a “reasonable doubt” standard, the Netherlands uses a “gained the conviction” standard. See my essay on the Iraqi High Tribunal here for a survey. A number of civil-law scholars have criticized the Dutch standard for being too low.)


13 Responses

  1. The quote you refer to is actually from Human Rights Watch, not “Human Rights.”

  2. Guess I’ll have to look into that one.

  3. Done: The Appeals Court didn’t say that the Andal Campaign wasn’t genocide.

    I quote (translation mine): “The Court determines that a final judicial decision on the very important question that is highly relevant from an international perspective, whether certain acts in the indictment by certain people amount to genocide, must be based on a more solid foundation than was presented to the Court.”

    The Appeals Court, for reasons not specified, relied on the report of the Special Rapporteur to the Commission on Human Rights. The Court states that the report clearly indicates that there was a genocidal intent, but that the report in itself does not directly proof a genocide up to a sufficient standard. The report states that it was based on 14 tons of paper, but apparantly the District Attorney only supplied the report.

    It is up to the Disctrict Attorney to proof the indicted facts, not to the Court. Still dissapointing, but not as dissapointing as saying that no genocide occurred.

  4. here is the link to the Court of Appeal’s press release.

    A full translation of the verdict is pending, apparently.

  5. Don’t know if that was mentioned here already, but the dutch translation of decision of the Court in first instance can be found here.

    Although I agree that this decison is somewhat dissapointing one must take into account that the Court of Appeals has treated this case de novo (as far as I know). Clearly the Court of Appeal disagreerd with the Court of fist instance whether or an actual genocide was actually proved.

  6. Kevin,

    I think Anne’s last point is very important – in most civil law systems, and for all that I know in the Dutch one as well, the appellate court is conducting a full review of all of the facts of the case, in essence conducting a trial de novo, and is not deferring to the trial court in its assessment of the facts. So, if the Court of Appeals felt that genocide was not conclusively established by the prosecution, it had no choice but to overturn the trial court’s finding.

  7. Marko and Anne,

    I confess that I’m not quite sure what point you are making. Are you saying that there is nothing problematic about the appellate court’s holding regarding whether genocide occurred during the Anfal campaign? Both the trial court and the appellate court saw the same evidence, yet whereas the trial court concluded that genocide was “conclusively proven,” the appellate court did not. I find that troubling. True, it may be that the appellate court is simply rebuking the prosecutor for not doing a good enough job of proving the genocide — but again, the prosecutor obviously did a good enough job to satisfy the trial court. So I continue to believe the only rational inference is that the appellate court was hostile to the idea that the extermination of the Kurds constituted genocide.

  8. Kevin,

    Well, the point that I (and, I hope, Anne) was trying to make is the nature of factual determinations made by the Dutch Court of Appeals. As you well know, at the international level (and generally in common law systems) the standard of review adopted by appellate courts is whether a reasonable trial chamber could have established the existence of a crime beyond a reasonable doubt. Now, even this standard presupposes that two reasonable triers of fact could reach two different conclusions, but that the appellate court will indeed defer to the determination made by the trial chamber unless no reasonable trier of fact could have made it.

    In civil law systems there is no such standard of review. An appellate court can simply think that the trial court got it wrong in the first instance, and reverse their factual findings. That seems to be what happened here, as the trial court made the assessment that Anfal was genocide (which I think it was, btw) on the basis of a single report, without examining witnesses, engaging in any sort of comprehensive investigation into Saddam &Co.’s genocidal intent etc. So, even if we say that the trial court made a reasonable determination that genocide was committed in the Anfal campaign, the appellate court could just as reasonably say that insufficient evidence was presented to reach that conclusion.

  9. Marko says it more eloquently than I could… So I defer to Marko.

  10. Professor Heller,

    is there any authority for your statement that the Dutch criminal standard does not require proof beyond a reasonable doubt? Could it not be argued that the judge cannot have ‘gained the conviction’ if he/she continues to have reasonable doubts? As a matter of language and psychology, how can you be convinced if you are also left in (reasonable) doubt?

    I suspect (I don’t know) that the formulation based on the conviction of the judge is no more than the civil law’s way of describing the ‘beyond reasonable doubt’ test of the common law. As you note in your article, French law speaks only of the ‘intimate conviction’ of the judge, not of the strength of the evidence in any objective sense. Similarly, the German Code of Criminal Procedure only speaks of the ‘free conviction’ of the judge in assessing the evidence put before the court; the expression ‘beyond reasonable doubt’ was only used in the cases to describe the statutory standard, not to add to it.

    Disclosing, if you will, my reasons for doubting whether there is actually any difference between the civil law formulation and the common law formula, I should say that I have recently read an English case, where Lord Lane CJ said this:

    ‘the tribunal should apply the criminal standard of proof, that is to say proof to the point where they feel sure that the charges are proved or, to put it another way, proof beyond reasonable doubt’ (In re a Solicitor [1993] QB 69, 91).

    The Lord Chief Justice here appears to be saying that the conviction test and the ‘beyond reasonable doubt’ test are the same.

    Also, if there is a difference, this might actually work to the benefit of defendants in civil law jurisdictions: in the civil law tradition, the judge must be convinced of guilt, whereas at common law (and under the US Constitution – In re Winship, 397 U.S. 358 (1970)), the court must find that there are no reasonable doubts. Thus, if there is a difference between being convinced and not holding any reasonable doubts, the situation is this: a Continental European judge cannot convict if he remains unconvinced, even if there are no reasonable doubts (I still think this sounds nonsensical), whereas his/her colleague in the UK and the US (etc.) will only ask the latter question, and proceed to convict.

    I would be grateful for an explanation. The above is hardly more than a gut feeling.

  11. Tobias,

    Thanks for your (typically) challenging and insightful question. I will need a day or two to get you the cites — but I think it is safe to say that, at the semantic level, a simple “gained the conviction” standard is lower than either the “intimate conviction” or “beyond a reasonable doubt” standard. One can, I think, easily be convinced that something is true yet acknowledge the possibility — even the reasonable possibility — that you are wrong. For example, I’m convinced that my beloved Chicago Bears are going to make a return trip to the Super Bowl this football season, because they are still the best team in the NFC. I would not be surprised, though, if they didn’t: too many things can happen over the course of a season. I’m still convinced that they will — but I’m not intimately convinced, nor am I convinced beyond a reasonable doubt.

    That said, I acknowledge that my interpretation of “gained the conviction” could be an idiosyncratic one. Indeed, semantics aside, there is a decent amount of work in cognitive psychology that suggests not only that there is no significant difference in most people’s minds between the “clear and convincing” civil (US) standard and the “beyond a reasonable doubt” standard, but also that a decent percentage of people actually think that the “clear and convincing” standard is higher than the “beyond a reasonable doubt” standard. So you may well be right that, in practice, a judge will no more easily “gain the conviction” than he will be convinced “beyond a reasonable doubt.”

    Any thoughts?

  12. Prof. Heller,

    thank you for your reply.

    I am still partial to the view that being convinced and being in no reasonable doubt comes to the same thing. I acknowledge, however, that this may depend on whether you read the word ‘conviction’ in the statutes with any particular emphasis. (The German criminal cases suggest that you should, and that you would not change the statutory meaning in doing so.)

    With the greatest respect to you and to the Chicago Bears, I doubt whether you can really be ‘convinced’ that they will succeed again. I assume (possibly due to my near-perfect ignorance of American Football) that your assessment there contains a slight subjective element, so the term ‘strong belief’ would seem to be more in point than the term ‘conviction’.

    [As an aside: more fundamentally, can you really be ‘convinced’ of future events? You can certainly be convinced of the existence (the present existence) of a good chance or high risk of some future event, but probably not of the event itself, right? This would imply a distinction between the degree of the chance/risk, and the degree of your conviction of that chance/risk (which distinction admittedly makes the concept of a standard of proof in assessments of risks or chances ‘not particularly helpful’: see R (N) v. Mental Health Review Tribunal (Northern Region) [2005] EWCA Civ 1605, [2006] QB 468, paras. 93-94, 99).]

    I am also not quite sure that the word ‘intimate’ in ‘intimate conviction’ (‘intime conviction‘, in Articles 353, 304, 427 of the French Code of Criminal Procedure) really qualifies the strength of the conviction. Instead, I suspect that the phrase is directed at a feature of medieval civil law that the common law has, to the best of my knowlegde, never known: the former source of law used to provide for fixed rules on the weight of types of evidence; thus, three witnesses would be enough to establish a fact, no matter what the judge personally made of them (this is only an example; I don’t know if I have accurately described this particular rule). It is against this background that I would see the ‘intime conviction’ phrase: under this rule, the judge is free to form his or her own opinion of the totality of the evidence adduced. The phrase thus does not mean ‘strong conviction’, it means ‘personal conviction’.

    I feel that this reading is closer to the natural meaning of the word ‘intimate’/’intime: the word generally means something a lot like ‘close’ (which isn’t a million miles from ‘personal’ or ‘internal’), but it is less likely to mean ‘certain’ or ‘strong’. Besides, I derive some support for my suggested interpretation of the phrase from the statutes themselves:

    Of the cited (French) provisions, Article 304 in particular emphasizes that the law does not prescribe what weight the judges will have to give certain pieces or kinds of evidence, and summarizes this in the final question: ‘do you have the intimate conviction’? Article 353, for its part, mentions the presumption of innocence and the rule that the accused must have the benefit of the doubt, but this seems to have been put side by side with the ‘intime conviction’ part, not as another illustration of the latter phrase.

    Similarly, German law uses the term ‘the free conviction of the judge’ (§ 261 of the Code of Criminal Procedure, § 286 of the Code of Civil Procedure and § 108 of the Administrative Courts Act), and not a phrase relating to the strength of the judicial conviction.

    (Incidentally, if this point is correct, then the – frankly, astonishing – result would be that the criminal standard of proof beyond reasonable doubt does not in fact find expression anywhere in the German Code of Criminal Procedure; it would then be simply a general principle of criminal law and, of course, it would be a consequence of the presumption of innocence – which, likewise, appears nowhere in express terms in German law, except in Articles 6(2) ECHR and 14(2) ICCPR), but has been authoritatively read into the constitutional provision on the rule of law.)

    Finally, one further short remark on your point about the practical difference between the various standards: as you may know, English law does not know the intermediate standard applied under the Fifth Amendment (e.g., in Addington v. Texas, 441 U.S. 418 (1979)). It does, however, apply the one civil standard of proof on the preponderance of probabilities in a manner that has regard to the severity of the charge and of the consequences of the finding. This leads to much the same results as the US approach (see on all of this the case cited above, at paras. 60, 62-67). In cases where the charges were serious, and the consequences of the relevant finding of fact were severe, the courts have occasionally said that the civil standard, in its application to such cases, was ‘for all practical purposes (…) indistinguishable from the criminal standard’ (B v. Chief Constable of Avon and Somerset Constabulary [2001] 1 WLR 340, para. 31 (per Lord Bingham of Cornhill CJ); Gough v. Chief Constable of the Derbyshire Constabulary [2002] EWCA Civ 351, [2002] QB 1213, para. 90; R (McCann v. Manchester Crown Court [2002] UKHL 39, [2003] 1 AC 787, para. 37 (per Lord Steyn), para. 83 (per Lord Hope of Craighead); Campbell v. Hamlet [2005] UKPC 19, paras. 17-22).

    This is, of course, a bit of an abidication of principle, but it tends to show that the standards are not easily distinguished. It may even show, as Morris LJ put it in Hornal v. Neuberger Products Ltd. [1957] 1 QB 247, 266, that ‘the life of the law is not logic but experience’ (adopted in N‘s case, cited above, at paras. 39, 68).

    I still maintain that logic has a role to play…

  13. Sorry, the links to the French code don’t work. If you are interested, you should go to http://www.legifrance.gouv.fr, then to Les codes, on to CODE DE PROCEDURE PENALE, and then enter the number of the article in the search engine.

    Alternatively, you could have a look at the English version by clicking the Union Jack on the start page of legifrance, and making your way from there. The translation of ‘intime conviction‘, incidentally, is ‘inwardly convinced’ or ‘innermost conviction’. But this is obviously not authoritative.

Trackbacks and Pingbacks

  1. There are no trackbacks or pingbacks associated with this post at this time.