ECLI:NL:PHR:2018:1394

No. 17/02109 Hearing: 26 June 2018 Advocate General E.J. Hofstee Advisory opinion concerning: [defendant] CONTENTS Introduction (marginal numbers 1 - 9) I.1 The appealed judgment (marginal numbers 1 - 4) I.2 The course of the proceedings and the background to the case (marginal numbers 5 - 10) I.3 The production of evidence (marginal numbers 11 - 15) I.4 The scope for review in a cassation appeal (marginal numbers 16 - 19) The 1st through 3rd ground of appeal in cassation: denied requests to hear witnesses (marginal numers 20 - 72) II.1 Preliminary comments on grounds of appeal 1 through 3 (marginal numbers 20 - 29) II.2 The 1st ground of appeal in cassation (marginal numbers 30 - 45) II.3 The 2nd ground of appeal in cassation (marginal numbers 46 - 53) II.4 The 3rd ground of appeal in cassation (marginal numbers 54 - 72) The 4th through 6th ground of appeal in cassation: failure to respond to expressly substantiated pleas(marginal numbers 73 - 105) III.1 Preliminary comments on grounds of appeal in cassation 4 through 6 (marginal numbers 73 - 80)III.2 The 4th ground of appeal in cassation (marginal numbers 81 - 87) III.3 The 5th ground of appeal in cassation (marginal numbers 88 - 93) III.4 The 6th ground of appeal in cassation (marginal numbers 94 - 105) The 7th ground of appeal in cassation: the right to examine witnesses (marginal numbers 106 - 118) The 8th ground of appeal in cassation: collusion (marginal numbers 119 - 132) The 9th through 16th and 18th through 21st ground of appeal in cassation: incriminating witness statements (marginal numbers 133 - 266) VI.1 Preliminary comments on grounds of appeal in cassation 9 through 16 and 18 through 21 (marginal numbers 133 - 148) VI.2 The 9th ground of appeal in cassation (marginal numbers 149 - 154) VI.3 The 10th ground of appeal in cassation (marginal numbers 155 - 168) VI.4 The 11th ground of appeal in cassation (marginal numbers169 - 173)VI.5 The 12th ground of appeal in cassation (marginal numbers 174 - 180) VI.6 The 13th ground of appeal in cassation (marginal numbers 181 - 188) VI.7 The 14th ground of appeal in cassation (marginal numbers 189 - 193)VI.8 The 15th ground of appeal in cassation (marginal 194 - 213)VI.9 The 16th ground of appeal in cassation (marginal numbers 214 - 234)VI.10 The 18th ground of appeal in cassation (marginal numbers 235 - 245)VI.11 The 19th ground of appeal in cassation (marginal numbers 246 - 252)VI.12 The 20th ground of appeal in cassation (marginal numbers 253 - 258)VI.13 The 21st ground of appeal in cassation (marginal numbers 259 - 266) The 17th, 23rd and 24th ground of appeal in cassation: lawfulness of means of evidence (marginal numbers 267 - 285) VII.1 The 17th ground of appeal in cassation (marginal numbers 267 - 278) VII.2 The 23rd and 24th ground of appeal in cassation (marginal numbers 279 - 286) The 22nd ground op appeal in cassation: ‘denaturing’ statements of the defendant (marginal numbers 287- 292) The 25th through 28th ground of appeal in cassation: other expressly substantiated pleas (marginal numbers 293 - 327) IX.1 The 25th ground of appeal in cassation (marginal numbers 293 - 300) IX.2 The 26th ground of appeal in cassation (marginal numbers 301 - 309) IX.3 The 27th ground of appeal in cassation (marginal numbers 310 - 318) IX.4 The 28th ground of appeal in cassation (marginal numbers 319 - 327) The 29th ground of appeal in cassation: various other complaints (marginal numbers 328-353) The 30th ground of appeal in cassation: amnesty (marginal numbers. 354 - 376) Conclusion (marginal numbers 377 - 379) I. Introduction I.1 The appealed judgment The defendant was sentenced by judgment of ′s-Hertogenbosch Court of Appeal of 21 April 2017 to a term of imprisonment of nineteen years, with credit for time on remand, for - in brief - having aided and abetted the joint perpetration of war crimes and co-committing intentional breaches of the Sanctions Act 1977. The Court of Appeal classified the offences it found to have been proven as:Counts 1, 2 and 3 - The war crimes- at 1A, 2A and 3A, all being further alternative charges and in each case: ‘Complicity in co-committing violations of the laws and customs of war, while the offence results in death or involves rape, committed multiple times’; Counts 4 and 5 - The breaches of the Sanctions Act 1977- at 4: ‘Co-committing deliberate violations of a regulation laid down in article 2 ICW article 3 of the Sanctions Act 1977, committed multiple times’”; and - at 5: ‘Co-committing deliberate violations of a regulation laid down in article 2 of the Sanctions Act 1977, committed multiple times’. The Court of Appeal acquitted the defendant of the other offences with which he was charged. It did not consider that it had been lawfully and conclusively proven that the defendant had co-committed the war crimes referred to as count 1, count 2 and count 3 (1A, 2A and 3A, all of these being principal charges) or that he had incited others to commit them (1A, 2A and 3A, all of these being alternative charges). Under 1B, 2B and 3B the defendant was also charged, in brief, with having intentionally allowed war crimes to be committed by persons subordinate to him. The Court of Appeal also acquitted the defendant of these offences. 3. The Court of Appeal also ordered the arrest of the defendant. No grounds of appeal in cassation have been lodged against either the acquittals or the warrant of arrest. 4. The defendant has instituted an appeal in cassation (against the judgment in its entirety). On behalf of the defendant R.J. Baumgardt and I.N. Weski, both of whom are attorneys-at-law in Rotterdam, have lodged a statement listing 30 grounds of appeal in cassation. I.2 The course of the proceedings and the background to the case 5. This criminal case is attracting a great deal of public interest both in the Netherlands and abroad. And it is not just the media that have been covering the case. It has also been discussed in the criminal law literature on a number of occasions. This interest is due mainly to the rather exceptional nature of the charges, as it is not often that a European is charged in his own country for involvement in international war crimes committed outside the borders of Europe. 6. The appealed judgment of the Court of Appeal has been published on rechtspraak.nl. For the entire ruling on the evidence and the grounds on which the ruling was based, reference may therefore simply be made to that website. However, it is necessary to set out briefly here the long history of this case, as revealed in the exhibits. 7. The defendant is a Dutch national who, by his own account, has carried on business in Liberia on the west coast of Africa since the early 1980s. Among his activities there he managed a hotel known as [A] in the Liberian capital Monrovia and operated a logging business. From 1986 he owned a logging company known as [B] . When the first civil war broke out in Liberia he temporarily wound up his business activities there and moved to neighbouring Sierra Leone. Some time after the murder of the Liberian president Samuel Doe, the defendant returned to Liberia. His company [B] was later, from 1997 or 1998, named [C] . In 1999 the defendant, together with his Indonesian business partners, also established a logging business known as [D] . The defendant was president of [D] .

Charles Taylor, with whom the defendant had already been in contact before the first Liberian civil war (because Taylor was a member of the Doe government), had become president of Liberia in 1997. As the owner or president of two large logging companies in which Charles Taylor had a financial interest, the defendant had fairly close ties with Taylor and his regime. 8. 1999 saw the outbreak of the second Liberian civil war, about which the Court of Appeal found, inter alia, as follows. For some five years there was an almost continuous armed struggle between Liberian combined armed forces led by Charles Taylor and a rebel group known as Liberians United for Reconciliation and Democracy (LURD). This latter group was supported by another rebel group known as the Movement for Democracy in Liberia (MODEL). LURD is thought to have also received support from the government of Guinea, which borders Liberia to the north. Fighting also took place in Sierra Leone, which borders Liberia to the north-west. The second Liberian civil war followed a cyclical pattern, with mixed results on both sides. According to the evidence on which the Court of Appeal based its judgment, the Liberian combined armed forces committed war crimes at various places between 2000 and 2002 during this second Liberian civil war. Count 1 relates to war crimes committed in Guéckédou, a town in Guinea near the border with Liberia. Count 2, namely involvement in war crimes, relates to offences committed in or near Voinjama. Count 3 relates to war crimes committed in or near Kolahun. Voinjama and Kolahun are towns in northern Liberia, situated in Lofa County near the border with Guinea. The Court of Appeal explained at length in its appealed judgment (at J.1 to J.4) what atrocities had been committed in these towns and the neighbouring villages. The war crimes committed (on multiple occasions) by Liberian combined armed forces, which were held to have been proven by the Court of Appeal, consisted (inter alia) of: - intentionally shooting and killing civilians; - subjecting civilians to a barrage of fire; - setting fire to houses in which civilians were present; - decapitating civilians; - smashing babies’ skulls; - throwing live babies into a well; - driving civilians from their homes, locking them up and then throwing a hand grenade inside; - raping women and children; - torturing civilians; and - plundering the possessions of citizens. 9. An investigation was started against the defendant in the Netherlands in the spring of 2004. This was in response to a report issued in December 2000 by a Panel of Experts set up by the Security Council of the United Nations (hereinafter UN) and a 2003 report by Global Witness. These reports indicated that the defendant and [D] had been heavily involved in illegal arms trading in the region. From the outset, however, the defendant has categorically denied having been involved in importing arms into Liberia or contributing to the armed struggle in any other way. Defence counsel, who has represented the defendant before all the courts of fact, has argued his defence on this basis and has pointed throughout to what she sees as deficiencies in the criminal investigation and to inconsistencies and discrepancies in and the possible unreliability of the evidentiary material in this case. She has submitted in respect of all counts of the indictment that the case brought by the Public Prosecution Service is inadmissible or that the defendant should be acquitted. 10. In 2006 The Hague District Court sentenced the defendant to a term of imprisonment of eight years for joint perpetration of intentional breaches of the Sanctions Act 1977 (counts 4 and 5). The District Court held that the defendant’s criminal involvement in the war crimes with which he had been charged had not been proven. The Hague Court of Appeal acquitted the defendant on all counts in 2008. An appeal in cassation lodged against this judgment by the Public Prosecution Service was held by the Supreme Court in 2010 to be well-founded because the Court of Appeal had failed to give adequate reasons for refusing the request made at the hearing to have the examining magistrate question witnesses A03 and A04 pursuant to article 226a of the Code of Criminal Procedure. The Supreme Court set aside the judgment of The Hague Court of Appeal and referred the case to ′s-Hertogenbosch Court of Appeal, which - as already noted - gave judgment on 21 April 2017. The relatively protracted nature of the appeal proceedings was due to various factors, including the length and complexity of the investigation, the many aspects of the case which the Public Prosecution Service and the defence wished to have investigated, the international element of the investigation and the consequent dependence on the willingness of foreign authorities to cooperate, as well as the Ebola crisis which broke out in West Africa at the time of the trial. I.3 The production of evidence 11. The judicial findings of fact were included by the Court of Appeal in the appealed judgment after part N.1. 11. For an adequate assessment of the grounds of appeal in cassation and a proper understanding of this advisory opinion, it is important to know what facts and circumstances found by the Court of Appeal to have been established together form the basis for its view that the defendant intentionally aided and abetted the commission of the war crimes by Liberian government troops in and around Guéckédou, Voinjama and Kolahun. As regards the defendant's involvement in these war crimes, the Court of Appeal held as follows in the appealed judgment: ‘L. Considerations regarding the evidence for the defendant’s involvement in the war crimes. Regarding the role of the defendant or his involvement in the abovementioned war crimes, on the basis of all the foregoing considerations and from what is demonstrated in the evidence, it has been established that: - in the proven periods, he was the owner of [C] and president of [D] ; - in he period from 2000 until 2003 inclusive, with his company [D] , he was actively involved in the import of weapons with the [ship 1] in the port of Buchanan and also supplied weapons to (the regime of) Charles Taylor (see H. through H.6); - he provided employees of the companies [D] and [C] to unload the weapons and ammunition, but also to transport these weapons to the Loop in Buchanan, the Executive Mansion and White Flower in Monrovia and Camp Bomi Wood near Tubmanburg . These include places where those weapons were stored and/or distributed among defendant's employees and/or the Liberian combined armed forces (see H. to H.6 inclusive). Those weapons and ammunition were also transported to the front lines by [D] / [C] personnel. - he provided employees of the companies [D] and [C] for the armed conflict in Lofa County and Guéckédou. These employees often used the weapons which had been distributed to them by these companies (see J.1 to 4 inclusive); - he allowed that [co-perpetrator 2], [co-perpetrator 3] and [co-perpetrator1], being persons who were closely related to the defendant and those companies, but also to (the regime of ) Charles Taylor, played important roles in the armed conflict in Lofa County and Guéckédou, and also allowed them to order the employees of these companies to participate in the armed conflict in Lofa County and Guéckédou (see under J.1 to 4 inclusive); - he apparently made [C] camp Bomi Wood available for the storage of weapons and ammunition from the [ship 1]. This place was also used to distribute those weapons and ammunition among the commanders at the frontlines of the armed conflict, including by [co-perpetrator1] (see under J.1 to 4 inclusive); - he visited that camp multiple times and encouraged warriors to fight; - he was in the immediate vicinity when the command was given to use heavy weapons in the attack on Guéckédou.

The defendant expressly promised the soldiers that they were allowed to loot (“the soldiers had to complete the mission and (…..) promised that they could keep everything that they would find afterwards”).’ 13. These considerations from the second indent onwards contain that the defendant was actively involved in importing arms on the Antarctic Mariner, the vessel owned by [D] (referred to in the case documents as AM), through the port of Buchanan. Not only was this active involvement in the importation of arms one of the main reasons for holding that the defendant had aided and abetted the joint perpetration of war crimes, but it was also the basis for his criminal conviction for having perpetrated intentional violations of the UN arms embargoes. As regards the delivery of arms and ammunition by the Antarctic Mariner, the Court of Appeal held as follows in the appealed judgment at H.4 (footnotes omitted): ‘H.4 Delivery of weapons and ammunition by the [ship 1] With regard to the delivery of weapons and ammunition by the [ship 1]: From 2000 [D] is the owner of the ship the [ship 1]. - Between 2000 and 2003, goods were transported via Serway Shipping (previously: Global Star Shipping Company) by the [ship 1] from/for [D] and [C] between Liberia and Asia. For this transport the ship used the port of Buchanan in Liberia. - The overview in the logbook of the [ship 1] shows that in the period from 2000 to April 2003 the ship entered the port of Buchanan eight times to unload goods. In this respect the court considers that, according to the underlying data16 of this overview, contrary to the data in the logbook, the ship did not enter the port of Buchanan on 21 April 2003, but on 25 April 2003. In conclusion, the ship entered the port of Buchanan in the months of August 2000, January 2001, June 2001, November 2001, February 2002, August 2002, December 2002 and April 2003. - The witness statements used by the court as evidence show a consistent image of a more or less fixed procedure which was followed when the [ship 1] entered the port of Buchanan with weapons and ammunition on board (this cannot be established with certainty regarding the delivery of August 2002, see below under H 4.2). (i) High-ranking persons of [D] and Taylor’s government would come to the port of Buchanan. The same persons would not always be present, but often the defendant and [co-perpetrator 1] were among them. (ii) [D] security officers were instructed to close the port and ensure that only the high-ranked persons, [D] security personnel and [D] employees who were involved in unloading the weapons, would be present. (iii) On several occasions the defendant himself inspected shipments of weapons and ammunition. (iv) On several occasions the defendant himself directly gave the order to unload the weapons and ammunition. (v) Subsequently, the high-ranking persons would disappear from the port, after which various witnesses, including persons belonging to the security of [D] , boarded the ship and unloaded the weapons and ammunition. (vi) Unloading weapons and ammunition only took place during late evening and night, because the [D] leadership did not want to disclose the nature of the goods on board of the ship. However, the unloading of regular “commercial” goods did take place during the day. (vii) The weapons and ammunition were packed in wooden crates/boxes (“wooden crates/cartons/boxes”) and containers. (viii) Several witnesses saw weapons and/or ammunition in those containers and wooden crates/boxes. Those witnesses testified about RPG’s, AK 47’s and large machine guns (GMG’s) and the corresponding ammunition. In addition, witnesses saw inscriptions on the containers and wooden crates/boxes that indicated the presence of the aforementioned weapons (for example, "AMMO", "ak47 rifle" "RPG"). Some witnesses saw weapons in crates/boxes, because those crates/boxes had been broken during transport or unloading. (ix) The weapons were loaded on trucks, logging trucks (including low loaders) and/or pickups ("trucks"). The containers and crates were covered with nets or tarpaulins. - Furthermore, those testimonies give a general picture of the places where the arms and ammunition brought to the port of Buchanan by the [ship 1] were subsequently stored and distributed: (x) Some of the weapons and ammunition were directly distributed to the security guards of [D] in the port of Buchanan. (xi) Escorted by the Anti Terrorist Unit (ATU), the ‘trucks’ transported the containers and crates/boxes with arms and ammunition. Some of the weapons and ammunition were brought to the ‘Loop’. On the premises of ‘The Loop’ there were villas which were used by the management of [D] . This is where the weapons and ammunition were stored and distributed to the security guards of [D] . (xii) Weapons and ammunition were also transported to Monrovia, both to White Flower, a residence of Charles Taylor in Congotown, and to the Executive Mansion in Monrovia. In (the direct vicinity of) both residences arms and ammunition were stored and distributed to government troops to be used (among others) for the armed battle at the frontlines. (xiii) Other parts of the shipment were subsequently transported to Bomi Wood, a small ‘base camp’ of [C] close to Tubmanburg, where again weapons were distributed among the government forces, which were subsequently brought to the frontlines. 14. The Court of Appeal based its judicial findings of fact and the other findings and considerations of a factual nature on the substance of the 89 items of evidence listed in the annexe to the appealed judgment. 14. The main reason that ′s-Hertogenbosch Court of Appeal reached such a different conclusion from that previously reached by The Hague Court of Appeal was that it assessed the evidentiary material differently, particularly the witness testimony. The items of evidence include 33 incriminating statements (nos. 29-56 and 80-84) made by a total of 21 witnesses in the presence of the police or an examining magistrate. It is apparent from the evidence, for example, that the defendant played a leading role in [D] and [C] , had close ties with Taylor and his regime, used the ‘we’ form when talking about the government troops and identified with those troops, and that war crimes were committed by government troops, that the Antarctic Mariner was available to the defendant and his companies, and that this vessel was moored in the port of Buchanan a number of times each year. However, that the vessel was used to import arms, that the defendant played an active role in importing them and, for example, was present in the port and carried out inspections and that staff of [D] and [C] had been sent to the front with his knowledge and approval, which has always been denied by the defence, can only really be concluded from the witness statements referred to above. It has been noted more generally in the literature that often little evidence other than witness testimony is available in criminal cases involving international war crimes, while in precisely this type of case it is difficult to assess the reliability of the witness testimony. This has been no different in the present case. The criminal proceedings have throughout revolved mainly around one question: whether or not the incriminating witness statements are sufficiently reliable to serve as proof of such serious accusations. The ′s-Hertogenbosch Court of Appeal found with regard to the reliability of the witness testimony that it had ‘no reason whatsoever to doubt the accuracy and reliability of [witness] statements, as far as used for the judicial finding of fact.’ The Hague Court of Appeal, by contrast, viewed the same evidence as an ‘irreconcilable impediment’ to forming a conclusive opinion since ‘it would not be appropriate to disregard anything that these witnesses stated which was inconsistent, impossible or in any other way hardly likely, and to carefully select odds and ends that might support the finding of fact.'.

1.4 Scope for review in a cassation appeal 16. In view of the above, it is hardly surprising that in these cassation proceedings the defence has once again taken issue with the investigation of the facts by the Court of Appeal, its selection and evaluation of the available evidentiary material and to what extent it considered or failed to consider the reliability of the witness testimony. With the exception of the 30th ground of appeal in cassation, all the grounds of appeal raise objections to the judicial findings of fact and/or the Court of Appeal’s decisions refusing investigation requests. In view of the history of the case, it comes as no surprise that the grounds of appeal in cassation repeatedly test the boundaries of what evidentiary decisions can be reviewed in cassation proceedings. 17. Nonetheless, there is little if any latitude in cassation proceedings for reviewing whether the Court of Appeal was right to convict. In its judgment of 19 December 2017 in the cases jointly known as the Breda Six, the Supreme Court once again concisely summed up the limitations to which it is subject when reviewing evidentiary decisions (and declared that this applied by analogy to cassation proceedings following a renewed conviction after a retrial). '[..] in cassation proceedings [..] it is [not] possible to examine whether the court [..] which has held the charges to be proven on the basis of its factual assessment of the evidentiary material was right to have formed that opinion or whether the facts and circumstances held to have been established by the court [ ...] in its explanation of its assessment of the evidence are correct. This is also true of the conclusions of a factual nature which the court has drawn from the facts and circumstances contained in the evidence. [...] in cassation proceedings, the only criterion by which such findings and conclusions can be assessed is comprehensibility (cf. Supreme Court 18 September 2001, ECLI:NL:HR:2001:AD3530). In the case of a conviction, the court must select from the available evidentiary material, within the limits set by the law, whatever it considers useful in terms of reliability and discard whatever it considers to be of no value. This decision on the selection and assessment of evidence, for which no reasoning need be given (save in exceptional cases), cannot be successfully challenged in cassation proceedings. The introduction of an obligation to state reasons in article 359, paragraph 2, second sentence of the Code of Criminal Procedure in 2005 has not altered the basic principle that the selection and assessment of the available evidentiary material is a matter for the courts of fact. This provision does mean that the court will have to give reasons for its decision in a number of cases. This is so, for example, if a detailed position has been argued by or on behalf of the defendant in relation to the evidentiary material in the case. However, this obligation to provide reasons does not mean that where the court does not accept such a position it has to go into every detail of the argument. (Cf. Supreme Court 11 April 2006, ECLI:NL:HR:2006:AU9130, NJ 2006/393.)’ 18. I will not go into any more detail about the general restrictions which the law of cassation procedure imposes in this respect. At this point it is merely important to remember that they exist and that they are of major significance in assessing the proposed grounds for cassation in the present case. I will discuss the more specific restrictions on review in cassation proceedings where they are relevant to consideration of the grounds of appeal in cassation. 19. The drafters of the statement of grounds of appeal in cassation have organised and structured their statement in such a way that the grounds are grouped together by subject. Wherever possible, this advisory opinion will deal with the grounds of appeal and submissions in the order they have proposed. Only occasionally have I felt it necessary to depart from this order where there is a connection between certain grounds of appeal in cassation. [...] XI. The 30th ground of appeal in cassation: amnesty 354. The 30th ground of appeal in cassation raises various issues with the Court of Appeal’s finding (and the reasons given for it) that the granting of amnesty to the defendant in Liberia did not prevent the admissibility of his prosecution in the Netherlands by the Public Prosecution Service. 355. The official record of the hearing at which the substantive aspects of the case were heard on appeal shows, among other things, that defence counsel put the defence's case in court on 17 February 2017 in accordance with the memorandum of oral pleading lodged by her with the Court of Appeal and deemed to have been incorporated in the record. In her oral pleadings, defence counsel submitted, among other things, that an Amnesty Act was approved in Liberia on 7 August 2003 under which all persons were granted amnesty in respect of both civil and criminal proceedings for acts and crimes committed by them during the Liberian civil war between December 1989 and August 2003. She argued that this Amnesty Act prevented prosecution of the defendant. 355. In its judgment, the Court of Appeal summarised and rejected this defence as follows (the footnotes included in the judgment have been renumbered by me). ‘Admissibility of the case brought by the Public Prosecution Service A. Liberian amnesty scheme Defence counsel has argued – in brief – that the case brought by the Public Prosecution Service should be declared inadmissible in respect of all charges, since an “Act to grant immunity from both civil and criminal proceedings against all persons within the jurisdiction of the Republic of Liberia for acts and crimes committed during the civil war from December 1989 to August 2003” dated 7 August 2003 (below: the Liberian amnesty scheme) is in force. As this amnesty scheme is directly applicable to the crimes with which the defendant has been charged, defence counsel argues that the defendant should not have been prosecuted (or that his prosecution should have been discontinued) by the Public Prosecution Service, or in any event that the same conclusion can be drawn on the basis of the principle of equality of treatment, the principle of protection of legitimate expectations and/or the absence of jurisdiction. A.1 The Liberian amnesty scheme and the right to prosecute The Court of Appeal notes as follows in relation to the Liberian amnesty scheme. During her oral pleadings, defence counsel submitted the above-mentioned Liberian amnesty scheme. Defence counsel attached to it an affidavit of attestation dated 14 February 2017. These documents show that the Liberian amnesty scheme was approved by the then President of Liberia on 7 August 2003 and published by the Liberian Ministry of Foreign Affairs on 8 August 2003. The affidavit was attached to demonstrate that the Liberian amnesty scheme is valid under Liberian law and has not been formally revoked in the intervening period. As Charles Taylor resigned as President of Liberia on 11 August 2003, he approved this amnesty scheme shortly before leaving office. The Comprehensive Peace Agreement (CPA) took effect soon after the publication of the Liberian amnesty scheme, namely on 18 August 2003, following negotiations in Ghana between the government (GOL) and its opponents (LURD and MODEL). This peace agreement also recommended the creation of the National Transitional Government of Liberia (NTGL).

This transitional government was broadly composed of representatives of the parties to the conflict, political parties, civil society organisations and the 15 counties (districts) of Liberia. Article XXXIV of the CPA provides that the NTGL: “shall give consideration to a recommendation for general amnesty to all persons and parties engaged or involved in military activities during the Liberian civil conflict that is the subject of this Agreement.” Article XXXV (c) of the CPA provides that: “For the avoidance of doubt, relevant provisions of the Constitution, statutes and other laws of Liberia which are inconsistent with the provisions of this Agreement are also hereby suspended”. Article XXXV (e) of the CPA provides that: “All suspended provisions of the Constitution, statutes and other laws of Liberia, affected as a result of this agreement, shall be deemed to be restored with the inauguration of the elected Government by January 2006. All legal obligations of the transitional government shall be inherited by the elected government.” Article XIII of the CPA of 18 August 2003 provided for the establishment of a Truth and Reconciliation Commission to address issues of impunity. The Truth and Reconciliation Commission (TRC) was established by Act of 10 June 2005. Point (g) of Article VII (Functions and Powers) of the Truth and Reconciliation Commission Act (TRC Act) stated as follows: “(...) provided that amnesty or exoneration shall not apply to violations of international humanitarian law and crimes against humanity in conformity with international laws and standards.” (all underlining above added by the Court of Appeal) Article II of the TRC Act contains definitions of “human rights violations” and “violations of international humanitarian law”. From the provisions of article VII of the TRC Act quoted above, the Court of Appeal infers that the Transitional Government (NTGL) apparently decided not to introduce a general amnesty. In 2009, the Truth and Reconciliation Commission published a report containing recommendations on amnesty and prosecution of offences committed during the civil war. These were offences to which the Liberian amnesty scheme cited by the defence would also apply. A.1.1 On the basis of the facts and circumstances described above at A.1. the Court of Appeal concludes as follows. The Liberian amnesty scheme was approved by Charles Taylor, the then President of Liberia, on 7 August 2003, which was one of the last days of his presidency. On 11 August 2003, i.e. four days after giving his approval, Taylor transferred power to the then Vice President Moses Blah and left or fled from Liberia. When the amnesty scheme was introduced during the then rule of Charles Taylor, peace talks were already being held in Ghana and all parties involved were thus discussing how and in what circumstances amnesty would be granted. It is apparent from the Comprehensive Peace Agreement (CPA), which came into force several days later, that the parties agreed that the Transitional Government (NTGL) should consider introducing a general amnesty scheme. It also explicitly provided for the establishment of a Truth and Reconciliation Commission (TRC), which would be expressly authorised to make recommendations regarding amnesty or prosecution. Article VII, section 26, point g of the TRC Act explicitly provides that recommendations for amnesty would not apply to “international crimes”, such as the war crimes with which the defendant is charged.[noot van AVT: de Engelse term 'international crimes' wordt niet gebruikt in onderdeel g. De citaat is: 'violations of international humanitarian law and crimes against humanity in conformity with international laws and standards'.] No reference is made to the Liberian amnesty scheme mentioned by the defence either in the CPA or the TRC Act or in a subsequently published TRC report. A.1.2 The Court of Appeal notes first of all that there is nothing in the case file or the proceedings at the court hearings to suggest that the Liberian amnesty scheme was invalid or was formally withdrawn or formally rendered inoperative at some point. However, from the manner in which the Liberian amnesty scheme was established and the choices subsequently made and steps taken by (among others) the Transitional Government (NTGL) and the Truth and Reconciliation Commission (TRC), the Court of Appeal concludes, together with the Advocates General, that although the scheme may have been in force at a certain time, it did not have (or no longer had) formal legal force owing to the conclusion of the CPA and the establishment of the Truth and Reconciliation Commission (TRC). After all, the power to make recommendations to the government about prosecution or amnesty with regard to crimes committed during the second civil war passed to the Truth and Reconciliation Commission (TRC). Moreover, recommendations to grant amnesty for war crimes or crimes against humanity were expressly excluded in the TRC Act. These findings are not changed by defence counsel’s submissions that the Supreme Court of Liberia held in a judgment that the TRC had exceeded its powers in certain respects, that a commission member of the TRC had explicitly mentioned the Liberian amnesty scheme in a dissenting opinion on that judgment, that there had been (or was still) a debate about the scope of the TRC’s powers and that two commission members did not sign the TRC report dated 30 June 2009, whatever the merits of these submissions may be. A.1.3 The Court of Appeal also notes, for the record, that even if it has to be assumed that the Liberian amnesty scheme had formal legal force (for a short period) and applied to the defendant as well, amnesty or entitlement to amnesty for war crimes would be contrary to international law and should therefore be excluded. The Court of Appeal believes that under international law, for example articles 2 and 3 of the Convention for the Protection of Human Rights and Fundamental Freedoms (ECHR), there is a positive obligation to institute an effective (criminal) investigation in cases where there is a suspicion of war crimes or crimes against humanity and, if necessary, to prosecute such crimes. Granting amnesty for war crimes and crimes against humanity is therefore not compatible with international law. An amnesty such as this one, which would exclude a prosecution brought in consequence of the above-mentioned positive obligations, would therefore constitute a violation of the ECHR (except in special circumstances, of which the Court of Appeal has not seen any evidence in this case). Accordingly, the Court of Appeal considers that, on the basis of international law, a national amnesty scheme which excludes prosecution for war crimes or crimes against humanity is incompatible with the obligation under international law to institute prosecution for such crimes. Any such amnesty scheme is therefore not applicable. Indeed, this is also a direct consequence of UN Security Council Resolution 1674 (2006), which emphasises the responsibility of States to comply with their obligation to end impunity and to prosecute those responsible for war crimes and crimes against humanity.

In reaching this finding, the Court of Appeal has taken into consideration that the amnesty scheme in question was not created as part of the peace process, but was instead drawn up by the then government of Charles Taylor, under whose rule the crimes with which the defendant has been charged occurred, namely shortly before Mr Taylor’s more or less forced departure from Liberia. This is why the Court of Appeal believes that, even if it has to be assumed that the Liberian amnesty scheme had formal legal force (whether for a short period or otherwise) and also that the scheme is (or would have been) applicable to the defendant in Liberia, this does not mean that the Public Prosecution Service in the Netherlands has lost the right to prosecute or continue to prosecute the defendant. A.2 The Liberian amnesty scheme and the principles of equality of treatment and protection of legitimate expectations As regards the argument based on the principles of equality of treatment and protection of legitimate expectations, the Court of Appeal holds as follows. Under the discretionary principle laid down in article 167, paragraph 1 of the Code of Criminal Procedure, it is up to the Public Prosecution Service to decide independently whether prosecution should take place following a criminal investigation. According to current case law, the decision to prosecute lends itself to substantive judicial review only to a very limited extent, since the sole ground on which a prosecution brought by the Public Prosecution Service can be declared inadmissible is that instituting or continuing the prosecution would be incompatible with the principles of due process (in so far as relevant here, the principles of equality of treatment and protection of legitimate expectations) as no member of the Public Prosecution Service could reasonably have concluded that prosecution or continued prosecution would serve any interest protected by enforcement under the criminal law. As this criterion necessitates judicial restraint, the reasoning given for any decision to hold that the case brought by the Public Prosecution Service is inadmissible must meet stringent requirements. This involves weighing the Public Prosecution Service’s position on the interest served by the present criminal prosecution against the circumstances submitted by the defence, which are intended to show that the decision to prosecute is contrary to the principles of equality of treatment and protection of legitimate expectations, as argued in this case. A.2.1 Violation of the principle of equality of treatment occurs only where equal cases are treated unequally and there is no reasonable and objective justification for this unequal treatment. The Court of Appeal considers that it has not been argued or proven that Dutch citizens accused of crimes such as those with which the defendant has been charged have not been prosecuted on account of an amnesty granted under the Liberian amnesty scheme (or an entitlement to such an amnesty). The Court of Appeal therefore considers that the existence of equal cases involving the defendant and one or more unidentified third parties has not been demonstrated. A.2.2 As regards the submission based on the principle of protection of legitimate expectations, the Court of Appeal holds that such a submission can succeed only if the prosecution has been instituted or continued after statements (or comparable acts) by or attributable to the Public Prosecution Service have given the defendant a legitimate expectation that he will not be prosecuted or that the prosecution will be dropped. However, as a general rule a legitimate expectation cannot be founded on statements made or acts performed by officials in the Netherlands who have no authority regarding the decision on whether or not to prosecute (or continue a prosecution). In the Court of Appeal’s opinion, it has not been shown that definite undertakings were at some point given to the defendant that he would not be prosecuted for the offences with which he was charged. The Court of Appeal also considers, in view of what has been held above at A to A.1.3, that the defendant could not have derived a legitimate expectation from the Liberian amnesty scheme that he would not be prosecuted by the Dutch Public Prosecution Service or that any prosecution would be dropped. A.3 Absence of jurisdiction regarding war crimes With regard to having jurisdiction in relation to the alleged war crimes, the court considers that the prosecution of the defendant for these crimes is founded on articles 8 and 9 of the Wartime Offences Act (WOS). (WOS). To begin with, article 94 of the Constitution states that statutory provisions in force within the Kingdom of the Netherlands cannot be applied if this applicability is incompatible with any binding provisions and decisions of international law organisations. This provision states that the court needs to verify statutory provisions against treaties and decisions of international law organisations, but that the court may not verify those against unwritten international law. It is in the interest of all peoples that war crimes, wherever committed and by whomever, are combated and prosecuted. In any case, the four Geneva Conventions of 12 August 1949, are based on this principle (also see the considerations under K. through K.4). Since the entry into force of these treaties, acting in violation of common article 3 of these treaties constitutes crimes as defined in article 8 of the WOS and in this case, based on article 3 of the WOS, the Dutch judiciary is granted so-called universal jurisdiction. The court is of the opinion that this also expressly concerns the crimes defined in article 9 WOS. A.4 In view of the above the court concludes that with respect to the war crimes as charged, jurisdiction follows from the provisions laid down in articles 8 and 9 of the WOS and additionally concludes that in spite of the Liberian Amnesty Scheme, the Dutch Prosecution Service has not lost its right to (further) prosecute this defendant of Dutch nationality for crimes made punishable by Dutch legislation in force (during the period as charged by the Dutch Prosecution Service). A.5 Absence of jurisdiction regarding violation of the Sanctions Act of 1977 The defence has further argued that the Dutch Prosecution Service should be declared inadmissible in its prosecution with regard to counts 4 and 5 (acts in violation of the Liberian Sanctions Regulations 2001 and 2002, hereafter: the Sanctions Act violations) allegedly committed in Liberia, since jurisdiction to prosecute these crimes is lacking. After all, article 5 ICW 7 of the Dutch Criminal Code requires double criminality is this respect. Because the alleged violations of the Sanctions Act 1977 are also governed by the Liberian Amnesty Scheme, this requirement is not met, according to the defence. A.5.1 The court considers that the defence ignores the provision given in article 13 of the Sanctions Act 1977, since this article states that for violations of this act, an active nationality principle is in force. The article provides a more extensive regulation than article 7 of the Dutch Criminal Code. The purpose of the extension of the applicability is to prevent Dutch nationals from evading the rules imposed by the Sanctions Act 1977 by committing the prohibited acts in a foreign country. Article 13 of the Sanctions Act 1977 does not include the requirement that the crime should be punishable by virtue of the law of the country where the crime was committed.

It was deemed appropriate not to include that requirement of double criminality since, in case of sanctions, this often involves crimes committed in a country against which the sanctions are directed and which are not punishable in that country. The above implies that regarding the question of jurisdiction, only the Dutch nationality of the defendant is relevant. A.6 In this respect the court notes, perhaps unnecessarily, that the fact that an amnesty scheme is announced does not imply that the crimes which fall under this scheme would not be considered to be punishable crimes in Liberia anymore, but merely that the persons accused of such crimes will not be prosecuted. Therefore, in order to determine jurisdiction of The Netherlands and the possible double criminality in the context of that jurisdiction, such a scheme is not relevant. A.7 Conclusion In view of everything that has been held above, the Court of Appeal dismisses the defences put forward by defence counsel that are based on the Liberian amnesty scheme and intended to obtain a declaration that the case brought by the Public Prosecution Service is inadmissible.’ 357. All things considered, the Court of Appeal’s rejection of the claim by the defence that the prosecution's case is inadmissible on account of an amnesty can be divided into three parts. At A.1.2 the Court of Appeal has first held, in brief, that although the amnesty legislation in question may admittedly have been lawfully enacted, it has to be assumed that it had no (or no longer had) formal legal effect as a result of the peace agreement and the establishment of the TRC. Second, at A.1.3, the Court of Appeal has held ‘for the record’ that under international law there is a positive obligation to prosecute where there is a suspicion of serious international crimes, including war crimes, and that the granting of amnesty to the defendant is therefore not compatible with international law. This incompatibility with obligations under international law therefore means that the amnesty law is not applicable. Third, the Court of Appeal has considered the questions of whether an unwritten principle of law prevents prosecution of the defendant (at A.2) or whether there is no jurisdiction in respect of the offences (at A.3 to A.6). Both these questions have been answered in the negative by the Court of Appeal. 357. The ground of appeal in cassation, which contains submissions on points of law and reasoning, challenges above all the first two of these three decisions. However, I will view the ground of appeal in cassation and the decision which it challenges from a rather broader perspective and focus on how a possible amnesty may affect the criminal proceedings in the Netherlands. The theme of amnesty in general and the status to be accorded in Dutch criminal proceedings to a foreign amnesty in particular have - as far as I know - never previously been considered in Supreme Court judgments, although these subjects could potentially crop up more frequently in criminal cases given the prevalence of amnesty schemes, especially in connection with the termination of national or international armed conflicts. 359. The possibility of granting amnesty has been expressly recognised by the legislature in the Dutch Constitution. Until 1983 the subject of amnesty was dealt with in article 77 of the Constitution. Now it is concisely regulated in article 122, paragraph 2 of the Constitution: ‘Amnesty is granted by or pursuant to Act of Parliament.’ The legal consequence of granting amnesty was described by the legislature in relation to the 1983 Constitution in the following terms: ‘The granting of amnesty removes the criminal character of offences that have been committed.’ In doing so it also touched on an important substantive difference between amnesty and pardon: ‘A pardon involves reducing or commuting a sentence. The starting point for the decision is the convicted person or persons. The criminal nature of the act that has been committed continues to exist. Prosecution of other perpetrators of the act remains possible since they too have committed a criminal offence. In the case of amnesty the issue is approached from the perspective of objective law. Certain offences are held - in retrospect - not to be punishable. What has occurred has no criminal law consequences. Prosecutions are discontinued and new prosecutions are not instituted. Amnesty applies in principle to any perpetrators - known or unknown - of the offence eligible for amnesty. Pardon affects the sentence received by a convicted person, whereas amnesty nullifies the criminal nature of the act. The nature of these two legal concepts therefore differs.’ In view of the last part of this passage, the scope for amnesty in the Netherlands can best be compared to a ground for inadmissibility of the prosecution, but it should be noted that convicted persons can also benefit from an amnesty scheme. Although amnesty extinguishes the criminal nature of the act, this does not - or in any event need not - mean that in retrospect the conduct in question is justified or should be considered excusable. Amnesty is an opinion of the legislature about whether it is expedient to attach criminal law consequences to acts that are intrinsically prohibited. An amnesty does not detract from the extent to which an offence warrants punishment; instead it eliminates the possibility of imposing punishment. 360. Amnesty is a concept that is dormant in the Netherlands. As far as I am aware, the last occasion on which amnesty was granted in the European territory of the Kingdom of the Netherlands dates back to 1941. The subject therefore receives no real attention in the main textbooks on substantive criminal law and the law of criminal procedure. Only in the context of legal assistance provided to or received from foreign states is the subject of amnesty addressed somewhat more frequently and in greater detail. This is hardly surprising since a glance beyond the Dutch border reveals that amnesty plays a much greater role in the international community. Examples known in the Netherlands are the Indemnity Act enacted in South Africa and the Surinamese Amnesty Act of 1992, which was changed to such an extent under the regime of President Bouterse in 2012 that it now also relates to human rights violations committed in Suriname between 1980 and 1985, including the events of December 1982. As amnesty is granted on such varied grounds and conditions and for such varied purposes worldwide, it is hard to define it adequately and precisely. In general, however, amnesty always involves a legal measure or decision whose primary function is to remove the prospect and/or consequences of (criminal) liability for one or more individuals or groups of persons in respect of certain (criminal) offences. The precise scope of an amnesty depends on the relevant scheme and must therefore be viewed from the perspective of the law of the state granting it. Nonetheless, the content and tenor of foreign amnesty legislation cannot be determined autonomously in cassation proceedings since the court of fact’s interpretation of the amnesty law of a foreign state cannot then be challenged on points of law. This is because the Supreme Court is precluded from interpreting the law of foreign states under article 79 of the Judiciary (Organisation) Act. The court of fact’s judgment can be set aside on account of irrationality only if there are manifest errors in its interpretation. 361. The realisation that the most serious international crimes must not go unpunished is a major reason for the existence of international criminal law. Therefore attention paid in international criminal law literature to the legal concept of amnesty focuses in particular on whether it can be lawfully granted for crimes of this kind in the light of written and unwritten international law, and whether international courts and tribunals are bound by amnesty schemes.

No completely clear and unqualified answer is yet forthcoming. While the importance of holding perpetrators to account is recognised in the literature, the fact that an amnesty scheme may serve another real and legitimate interest is not overlooked. This justification of amnesty, even for the gravest crimes, is of a fairly pragmatic nature, namely that excluding application of the criminal law may be desirable both to foster a willingness to cooperate in laying down arms and to promote the reconstruction of a society during and after a transitional period. It is therefore generally accepted that international law cannot simply disregard amnesty arrangements outright in all circumstances. On the other hand, the view that not all amnesty arrangements can be justified has broad support. For example, self-amnesty adopted by a regime seeking to evade future criminal prosecution shortly before a change of power would be hard to justify. This notion seems to have been reflected by the Court of Appeal in the present case, where it explicitly notes that the amnesty scheme referred to in the ground of appeal did not come about in the course of a peace process, but was drawn up by the then government of Charles Taylor, under whose rule the offences occurred, shortly before his more or less forced departure from Liberia. 362. In my view, however, the question of whether an amnesty scheme should not be applied, either on account of the circumstances in which and the conditions on which it was established or on account of the nature and gravity of the crimes committed or both, does not arise in Dutch criminal proceedings until it has been determined whether the amnesty scheme can, in principle, be applied in the relevant case. After all, whether a given amnesty scheme is admissible in law in a specific case or is in conflict with international law becomes relevant only once it has been established that the scheme could frustrate criminal proceedings in the Netherlands. 362. It can be inferred from the system of the law and, above all, from articles 9 and 167 of the Code of Criminal Procedure that the Public Prosecution Service’s task of prosecuting crime extends, in principle, to all criminal offences. Corstens & Borgers rightly note that Title 8 of Book 1 of the Criminal Code accordingly frames this as a negative, namely the ‘lapse of the right to prosecute’, but that the criminal jurisdiction provisions have been formulated positively and can therefore justify referring in positive terms to the conditions for liability to prosecution. With the exception of the provisions that prevent Dutch criminal law and criminal procedure law from having jurisdiction over all offences worldwide, no matter where they are committed, discharge from liability to prosecution is therefore always based on special grounds. In so far as international law and unwritten legal principles affect whether offences can be prosecuted, they can be defined not so much as grounds for liability to prosecution (positive) but as grounds for exclusion of prosecution (negative). The Dutch system of liability to prosecution can therefore be summarised by saying that if jurisdiction over a criminal offence exists in the Netherlands the liability to prosecution is, in principle, a given, unless a general or special ground for inadmissibility applies. 364. First of all, therefore, it has to be asked whether an amnesty granted in a foreign state reduces or eliminates the scope for exercising jurisdiction. For Dutch criminal jurisdiction, a requirement of double criminality applies in some cases. Can it still be said that a prosecuted offence is punishable under the law of the state in whose territory it was committed if amnesty has been granted for it in that state? As far as I know, the Supreme Court has not yet decided in a general sense whether the requirement of double criminality in the statutory provisions on jurisdiction should be interpreted in abstract or concrete terms. However, the Supreme Court has previously held that the scope of article 5, paragraph 1, 2° (old) of the Criminal Code means that the fact that a prosecution has become time-barred abroad does not mean that the requirement of double criminality is no longer satisfied. It is usually assumed, more generally as well, that an abstract, overall test of whether the conduct in question is covered by the statutory definition of an offence is sufficient. Consideration is at most given in this connection to possible justifications for the offence that apply abroad, together with the elements of the offence. The text of the Code also points in the direction of an abstract test since it merely requires that the offence is punishable under the law of the state in which it is committed. Generally speaking, amnesty cannot be said to justify the conduct concerned in retrospect, but instead merely protects the person concerned from its possible legal consequences. Moreover, in the case of the double criminality requirement in the provisions on jurisdiction, the issue is whether double criminality existed at the time of the offence, whereas amnesty subsequently nullifies the criminal liability. A supplementary system-based argument for the position that amnesty is separate here from the double criminality of the offence can be derived from the state of the law on legal assistance. Although double criminality admittedly has a somewhat different function in relation to the rules on jurisdiction and its content is therefore not identical, it nevertheless seems important to me that under the law on extradition and surrender amnesty is a separate ground for refusal, together with the absence of double criminality. Article 3 (1) of Council Framework Decision 2002/584/JHA on the European arrest warrant and the surrender procedures between member states and article 4 of the Second Additional Protocol to the European Convention on Extradition both contain a provision making it mandatory to refuse surrender or extradition if an amnesty has been declared in the member state of execution or the requested state, as the case may be. If amnesty were to mean an absence of double criminality, there would be no need for the Second Additional Protocol to the European Convention on Extradition to provide specifically for amnesty to be a ground for refusal, since article 2 of that Convention requires double criminality. The Framework Decision attaches importance to the distinction between ‘list offences’, for which the member state of execution is not permitted to verify the existence of double criminality, and ‘non-list offences’, for which surrender may be made dependent on double criminality. This system is based on the principle of mutual recognition between the EU member states. If amnesty were to be regarded as a legal concept that undermines the double criminality of the offence, it would be a poor fit with this system since it is a - mandatory - ground for refusal. As article 3 (1) of Framework Decision 2002/584 /JHA nonetheless makes it mandatory to refuse surrender on account of an amnesty arrangement, this suggests that amnesty should be regarded not as a category of case in which (double) criminality is absent, but rather as an independent decision of the state concerned in respect of the criminal proceedings against the person claimed. As such, the decision is, in principle, also entitled to recognition from the other member states. I conclude from this that, even in a European context, decisions to grant amnesty, which are often of a political nature, are also not treated for the purposes of legal assistance as a circumstance removing the double criminality of the offence. In my view, it must therefore be assumed that the granting of an amnesty by a foreign state in whose territory an offence has been committed does not mean that the offence is not punishable under the law of that state within the meaning of article 5, paragraph 1 of the Criminal Code and article 7, paragraph 1 of the Criminal Code. 365.

Even if it were necessary to come to a different conclusion about this in a general sense, this would not result in a lack of jurisdiction in the present case. In the appealed judgment, the Court of Appeal has concluded at A.4 that the Netherlands has universal jurisdiction over the charge of aiding and abetting the joint perpetration of war crimes, pursuant to section 3, subsection 1, 1° of the Wartime Offences Act. At A.5.1 the Court of Appeal has held that no double criminality is required for offences 4 and 5 under the provisions of section 13 of the Sanctions Act 1977. Both views seem to me to be correct. This issue has not been raised in the cassation proceedings. 365. As noted by the Court of Appeal, once the legislature has established jurisdiction by declaring Dutch criminal law to be applicable to an offence, this applicability is not, in principle, limited by the law of a foreign state. If this were otherwise, the sovereignty of the Dutch state would be seriously and unacceptably jeopardised. Article 8d of the Criminal Code provides that once jurisdiction has been established, it may be limited only by ‘the exceptions recognised in international law’. This provision does not give the courts the power – notwithstanding the prohibition under article 94 of the Constitution on assessing whether statutory regulations are in conformity with unwritten international law – to assess whether Dutch rules on jurisdiction are in conformity with customary international law in the broadest sense, but amounts to ‘no more than a statutory recognition of immunity from jurisdiction derived from international law.’ In the light of these two provisions, the Dutch courts have only a modest role in determining how the Dutch legislature has established jurisdiction. 367. The immunities from jurisdiction to which article 8d of the Criminal Code relates are hard to ascertain in international law. There is no need to consider their precise scope. All that is of importance here is that amnesty differs in essential respects from the immunities recognised under international law and cannot therefore be regarded as an obstacle to the exercise of Dutch criminal jurisdiction under article 8d of the Criminal Code. I would mention just a few differences that seem to me to explain and justify the differing consequences under criminal procedure law between the two legal concepts. Immunities preclude the liability of certain entities, offices and acts from the outset. Immunity is claimed by the state, international organisations, heads of state, foreign ministers, ambassadors and foreign armies and warships. The rationale of immunities generally relates to the sovereignty and mutual independence of states and/or to the contribution they make to interstate legal relations. An important reason for a state to recognise in advance the immunity of dignitaries of foreign states and thus to accept a substantial violation of sovereignty within its own territory is to be able to rest assured that its own representatives and authorities, where eligible, will in turn also be safe from prosecution abroad. Amnesty, by contrast, is granted in retrospect and, by its very nature, is not limited to the holders of certain offices. In principle, anyone can be unilaterally protected by their national state from prosecution (or the possibility of prosecution). The element of reciprocity which characterises immunities under internal law is absent. 368. Against this background, it comes as no surprise at all that ‘amnesties do not have the status of state immunity in customary international law, nor the same basis in treaty law.’ It is generally assumed that the granting of amnesty is a legal decision of an individual state, which does not affect the sovereignty of other states and whose legal consequences are thus confined to within the legal order of the amnesty-granting state. The academic literature is consentient about this. For example, Cassesse et al. write as follows: ‘The current status of international practice, in particular its inconsistency combined with the more and more widespread opinio iuris in the international community that international crimes should be punished, could be conceptualized as follows. Subject to what has been said above with regard to terrorism and what is stated below with regard to genocide and crimes against humanity, there is not yet any general obligation for states to refrain from enacting amnesty laws on these crimes. Consequently, if a state passes any such law, it does not breach a customary rule. Nonetheless, if the courts of another state having in custody persons accused of international crimes decide to prosecute them, although in their national state they would benefit from an amnesty law, such courts would not act contrary to general international law, in particular to the principle of respect for the sovereign prerogatives of other states.’ O’Keefe also distinguishes between the questions of whether a state that grants amnesty thereby violates an international obligation and whether a rule of international law obliges a state to refrain from prosecuting a person who has been granted amnesty by a foreign state. As regards the second question, he states that: ‘International law does not prohibit the prosecution of an international crime statute-barred or amnestied in one state by any foreign state with concurrent prescriptive jurisdiction over that crime. [...] As for municipal law, it is self-evident that the courts of one state are not bound by the legislation of another state.’ Other commentators reach the same conclusion. For example, F. Williams states: ‘An amnesty does not of itself protect a person who has committed an international crime from proceedings in other states exercising universal jurisdiction or from the jurisdiction of international tribunals.’ P.D. Duyx too says there is little or no discussion about this: ‘The notion that amnesties have no extraterritorial effect now has fairly widespread acceptance.’ 369. In reaching this conclusion the various authors do not base themselves solely on the opinions expressed in the literature; international case law also supports this view. Here it is sufficient to refer thereto. However, I should like to single out for special mention the judgment of the European Court of Human Rights (ECtHR) in the case of Ould Dah v. France, because - like the present case - it involved the prosecution, in a state which is party to the European Convention on Human Rights (ECHR), of a defendant covered by an amnesty scheme in the country where the offences were committed. The applicant had been an officer in the Mauritanian army and was convicted by a French criminal court, exercising universal jurisdiction, for his part in acts of torture committed in that capacity. Ould Dah was sentenced in France to ten years’ imprisonment, although he was covered in his own country by an amnesty law enacted in 1993. He complained to the ECtHR of a breach of article 7 ECHR and submitted that he was entitled to invoke in France the amnesty that had been granted to him in Mauritania. The ECtHR viewed the matter differently. It reasoned, first, that torture should not go unpunished and that there are international obligations to prosecute and punish that crime.

It then went on to examine the extraterritorial effect of amnesty: ‘In addition, the Court notes that international law does not preclude a person who has benefited from an amnesty before being tried in his or her originating State from being tried by another State, as can be seen for example from Article 17 of the Statute of the International Criminal Court, which does not list this situation among the grounds for dismissing a case as inadmissible. [...] Having regard to the foregoing, the Court considers, in the present case, that the Mauritanian amnesty law was not capable in itself of precluding the application of French law by the French courts that examined the case by virtue of their universal jurisdiction and that the judgment rendered by the French courts was well founded.’ The application was declared inadmissible. 370. In the decision of the ECtHR, the fact that an amnesty has been granted abroad does not, in principle, preclude the institution of a prosecution in a state that is party to the ECHR. All in all, there is no doubt that the granting of an amnesty in a foreign state does not in itself extinguish Dutch jurisdiction and that there is no general international obligation not to institute a prosecution in the Netherlands on account of such an amnesty. 370. As explained above, the finding that an amnesty scheme does not obstruct the exercise of Dutch jurisdiction means that the Public Prosecution Service has, in principle, a right to prosecute unless a ground for immunity from prosecution exists. In my view, the only statutory ground for exclusion of liability to prosecution which could even remotely be considered applicable is that set out in article 68, paragraph 2 of the Criminal Code. This provision applies the ne bis in idem principle to the judgments of foreign criminal courts. It does so – since many years – in a rather far-reaching way A foreign criminal judgment that can be equated under Dutch law with an acquittal, a discharge from prosecution on a point of law, a judicial pardon or a conviction followed by the full execution of the penalty, a pardon or debarring of the execution of the penalty due to expiry of the period of limitation prevents prosecution of the defendant in the Netherlands, regardless of the country in which or the regime under which the judgment was given. Similarly, article 54 of the Convention implementing the Schengen Agreement (CISA) provides that a person whose trial has been finally disposed of in one Contracting Party may not be prosecuted in another Contracting Party for the same acts provided that, if a penalty has been imposed, it has been enforced, is actually in the process of being enforced or can no longer be enforced under the laws of the sentencing Contracting Party. The question whether an amnesty scheme constitutes a circumstance preventing enforcement of a penalty under the laws of the sentencing Contracting Party admits of various answers. A case can be made for answering this question in the affirmative. Furthermore, in my view, the possibility cannot be entirely excluded that an amnesty granted abroad must, in certain circumstances, be treated as equivalent to the granting of a pardon for the purposes of article 68, paragraph 2 of the Criminal Code. However, the present case is not one in which a conviction has not been enforced (or not fully enforced) as a consequence of an amnesty. After all, criminal proceedings were never instituted in Liberia. I see no reason to equate the amnesty with an acquittal, a discharge from prosecution on a point of law or a judicial pardon. Nor, therefore, does article 68, paragraph 2 of the Criminal Code prevent the prosecution of a defendant who has been granted an amnesty abroad, provided in any event that he has not also already been tried in that country. 372. In my view, the mere fact that amnesty has been granted to the defendant abroad cannot mean that his prosecution in the Netherlands is in breach of a principle of unwritten law. When assessing the defence’s argument based on the principle of legitimate expectations and the principle of equality of treatment, the Court of Appeal has rightly presupposed (at A.2), first, that the decision to prosecute lends itself to substantive judicial review only to a very limited extent, since it is only in exceptional cases that a prosecution brought by the Public Prosecution Service can be declared inadmissible on the ground that instituting or continuing the prosecution would be incompatible with the principles of due process and, second, that where a court finds that exceptional circumstances do exist and that the prosecution brought by the Public Prosecution Service must therefore be declared inadmissible its reasoning must meet stringent requirements. The Court of Appeal’s finding that the principle of legitimate expectations is not violated by an amnesty granted abroad is based on the view that only promises attributable to the Public Prosecution Service can give such a legitimate expectation of non-prosecution (or discontinuation of prosecution) that a subsequent prosecution will fail for this reason (at A.2.2). This view is correct. An amnesty granted by a foreign state is in any event not attributable to the Dutch Public Prosecution Service. However, those who have framed this ground of appeal in cassation still attach some value to a submission that was also made before the Court of Appeal, namely that the Netherlands endorsed and/or actively contributed to the introduction of the Liberian amnesty scheme. However, government bodies that operate in fields that have nothing whatever to do with investigation and prosecution cannot make a promise (or apparent promise) on behalf of the Public Prosecution Service that no prosecution will be brought. As the scope for review of the prosecution decision is very limited, the mere fact that a prosecution takes place in spite of an amnesty granted abroad cannot, without additional and extraordinary circumstances, conflict with one of the other principles of due process. 373. All of this leads me to conclude that an amnesty granted abroad cannot in itself prevent a prosecution in the Netherlands. Things may perhaps be different where there are special circumstances, for example where the Public Prosecution Service has categorically undertaken to respect the foreign amnesty law and hence to dispense with prosecution or where application of a foreign amnesty scheme is mandatory by virtue of a provision of a treaty binding on all persons or of a resolution of an international institution. Special circumstances of this kind have not been invoked either on appeal or in cassation, nor was their existence plausibly demonstrated during the proceedings in court. 373. Given the frequently political nature of amnesty and the justifications given for it, the conclusion that a foreign amnesty does not, in principle, prevent prosecution in the Netherlands seems perfectly reasonable. Notwithstanding the previous findings, amnesty arrangements can be an effective instrument in resolving international conflicts. In theory, therefore, a situation could arise in which the prospect of a prosecution in the Netherlands could jeopardise the peace process in a foreign state to such an extent that the importance of achieving peace must take priority over the exercise of the right of prosecution. However, this is a political decision and does not fall within the remit of a criminal court. Although an amnesty does not in itself affect the right of the Public Prosecution Service to bring a prosecution, the legislature is, in principle, free to pass an amnesty law relating to offences committed abroad. The public prosecutor may also take a foreign amnesty scheme into account when deciding whether or not to refrain from prosecuting, either at the direction of the Minister of Justice or otherwise. 375. In my view, this conclusion that a foreign amnesty arrangement does not, as a rule, bar a Dutch prosecution seals the fate of the 30th ground of appeal in cassation.

The complaints made in the ground of appeal and its explanatory notes relate to the Court of Appeal’s finding that the amnesty scheme is no longer valid under Liberian law and/or its finding that the application of the Liberian amnesty scheme in the present case would be contrary to international law. The submissions are therefore all based on the view that if the Liberian amnesty scheme is in force and not contrary to international law it can be applied in the present Dutch criminal case and can therefore be an obstacle to declaring that the prosecution instituted by the Public Prosecution Service is admissible. As I have said, I consider this interpretation of the law to be incorrect. 375. The 30th ground of appeal fails. XII. Conclusion 377. I will now conclude this advisory opinion. On the basis of the above, I consider that all 30 grounds of appeal in cassation fail. The first 29 grounds of appeal mainly take issue with the Court of Appeal’s judicial findings of fact, its selection and assessment of the evidentiary material (and the reasons given for this), and/or its decisions not to allow further examination of witnesses. Questions of law which the Supreme Court has not previously had an opportunity to answer are not raised in them or are raised only indirectly. In my view, therefore, these complaints do not raise questions of law that need to be answered in the interests of the uniform interpretation and application of the law or the development of the law. If the Supreme Court agrees with me that these complaints cannot result in cassation, it can, in my view, clearly reject grounds of appeal 1-6, 8-25 and 28 and the second complaint in ground of appeal 29 by confining itself to referral to section 81, subsection 1 of the Judiciary (Organisation) Act. Grounds of appeal 7, 26 and 27 and the first and third complaints in ground of appeal 29 could perhaps also be rejected on the grounds set out in section 81, subsection 1 of the Judiciary (Organisation) Act. The 30th ground of appeal is not, in my view, suitable for rejection in this way. 377. I have not found any grounds on which the Supreme Court should exercise its power to set aside the appealed judgment ex proprio motu. 377. I therefore recommend that the appeal in cassation be dismissed. For the Procurator General at the Supreme Court of the Netherlands Advocate General To this extent the charges relate to section 9 (old) of the Wartime Offences Act, which until 2001 read as follows: ‘Any person who intentionally allows an offence as referred to in the previous section to be committed by a subordinate is liable to the same sentence as that carried by the offences referred to in that section.’ As regards the interpretation of this provision and how this must be viewed in relation to international law, particularly the doctrine of command responsibility, see: Supreme Court, 8 November 2011, ECLI:NL:HR:2011:BR6598, NJ 2012/202, with note by Keijzer. For some articles published in Dutch since the appealed judgment, see: W. Huisman, De koopman en de dominee. Multinationals, mensenrechten en misdrijven, DD 2017, 70; C. Ryngaert, Nederlandse wapenhandelaar aansprakelijk gesteld voor medeplichtigheid oorlogsmisdrijven, Ars Aequi 2017, 833; H.G. van der Wilt, Tussen idealisme en realisme: Vervolging van internationale misdrijven in Nederland anno 2018, DD 2018, 21. Cf. W. Huisman, De koopman en de dominee. Multinationals, mensenrechten en misdrijven, DD 2017, 70, para. 1: ‘Although as yet only two individual businessmen have been convicted (i.e. the defendant and a Dutch national previously convicted by final and unappealable judgment of aiding and abetting war crimes in Iraq), the Netherlands is leading the way worldwide in prosecuting companies and business people for involvement in serious violations of human rights. Criminal proceedings against companies are also pending in some other countries, for example France and Sweden, but these cases have not yet resulted in conviction.’ ’s-Hertogenbosch Court of Appeal 21 April 2017, ECLI:NL:GHSHE:2017:1760 (available in English at ECLI:NL:GHSHE:2017:2650). The Hague District Court, 7 June 2006, ECLI:NL:RBSGR:2006:AX7098 (available in English at: ECLI:NL:RBSGR:2006:AY5160). The Hague Court of Appeal, 10 March 2008, ECLI:NL:GHSGR:2008:BC6068, NJ 2008/469, with note by Keijzer (available in English at ECLI:NL:GHSGR:2008:BC7373). The Supreme Court, 20 April 2010, ECLI:NL:HR:2010:BK8132, NJ 2011/576, with note by: Klip. Cf. Court of Appeal’s sentencing grounds, at Q of the appealed judgment. Cf. S. Huiberts-Van Dijk & M. Simmer, Opsporing & internationale misdrijven: de zoektocht naar documenten, Strafblad 2014, pp. 374-379. See: L. van den Herik, ‘The difficulties of exercising extraterritorial criminal jurisdiction: The acquittal of a Dutch businessman for crimes committed in Liberia’, International Criminal Law Review 2009, pp. 211-226; N. Combs, Fact-finding Without Facts – The Uncertain Evidentiary Foundations of International Criminal Convictions, Cambridge: Cambridge University Press, 2010; A.L. Smeulers, ‘Betrouwbaarheid van getuigenbewijs in WIM-zaken’, Strafblad 2014, pp. 354-364; G.N. Best, Fair and Accurate Fact Finding in Dutch Atrocity Crime Cases (diss. University of Amsterdam), Amsterdam 2016; S.L. Schot, ‘De kwetsbaarheid van verklaringen binnen de opsporing en vervolging van internationale misdrijven’, Strafblad 2017, pp. 551-555. Cf. the similar reasoning by former Advocate General W.H. Vellinga in his advisory opinion (at 17 and 18) before the Supreme Court judgment of 22 February 2005, ECLI:NL:HR:2005:AR5714. Supreme Court 19 December 2017, ECLI:NL:HR:2017: 3189. The subject is considered in detail in A.J.A. van Dorst, Cassatie in strafzaken, 8th edition, Deventer: Kluwer 2015, p. 181 ff. See also the above-mentioned opinion of former Advocate General W.H. Vellinga before the Supreme Court judgment of 22 February 2005, ECLI:NL:HR:2005:AR5714. Parliamentary Papers Senate 1978/79, 14006, 34, p. 12. Parliamentary Papers House of Representatives 1977, 14006, 9, p. 3. For example, the question of an amnesty granted abroad was raised before the courts of fact in the case of an Afghan general who was prosecuted in the Netherlands for crimes that included the joint perpetration of torture. See The Hague District Court, 25 June 2007, ECLI:NL:RBSGR:2007:BA7877 and The Hague Court of Appeal of 16 July 2009, ECLI:NL:GHSGR:2009:BJ2796. Parliamentary Papers House of Representatives 1979/80, 16 162, no. 3, p. 24. Parliamentary Papers House of Representatives 1979/80, 16 162, no. 3, pp. 24-25. The following passage appears in Parliamentary Papers House of Representatives 1980/81, 15 965, no. 5, p. 8: ‘leaving aside that the granting of amnesty in the Netherlands is more theory than practice.’ Tax Amnesty Decree of 10 March 1941, Bulletin of Acts and Decrees S. 408. Amnesty was last granted in 1914, in connection with the mobilisation following (in brief) the outbreak of the First World War, ‘to all those persons who [...] have been guilty of desertion from the armed forces of the Netherlands [...] before 1 August 1914 if they have reported to the mobilised forces of the Netherlands or report to them before 1 November 1914 [...]', Act of 6 August 1914, Bulletin of Acts and Decrees 1914, 376. It should also be noted that in the period when Indonesia was seceding from the Kingdom of the Netherlands amnesty was granted for crimes that were a consequence of the political conflict in that country. For more information about amnesty, see O.E. van Schravendijk, Leemten in het gratierecht (diss. Leiden), Deventer: Kluwer 1968, pp. 166 ff. and pp. 340-347. Publications that have dealt with amnesty are: A.J.A. van Dorst, Het Europees aanhoudingsbevel, in: G.J.M. Corstens et al.

Europeanisering van het Nederlands recht. Opstellen aangeboden aan Mr. W.E.Haak, Deventer: Kluwer 2004, pp. 406-424, p. 412; J.W. Ouwerkerk, Quid Pro Quo? A comparative law perspective on the mutual recognition of judicial decisions in criminal matters (diss. Tilburg), Cambridge/Antwerp/Portland: Intersentia 2011, p. 106; and V.H. Glerum, De weigeringsgronden bij uitlevering en overlevering. Een vergelijking en kritische evaluatie in het licht van het beginsel van wederzijdse erkenning (diss. Vrije University) Nijmegen: WLP 2013, pp. 582-589. Indemnity Act 35 of 17 May 1990, GG 12470. Also relevant is the Further Indemnity Act 151 of 9 November 1992, GG 14401. For a detailed account of the substance and context of this legislation, see A. Ellian, Een onderzoek naar de Waarheids- en Verzoeningscommissie van Zuid-Afrika (diss. Tilburg), Nijmegen: WLP 2003, particularly chapters 3 and 4 (2). Act of 5 April 2012 amending the Act of 19 August 1992 regulating the granting of amnesty to persons who have committed certain offences specified herein in the period from 1 January 1985 to the date of entry into force of this Act, Bulletin of Acts and Decrees of the Republic of Suriname 2012, no. 49. For more information about this legislation, see A. Ellian, President Bouterse en zijn amnesia. Suriname tussen waarheid en straffeloosheid, NJB 2012, issue 24, pp. 1645-1651. M. Freeman, Necessary Evils. Amnesties and the Search for Justice, Cambridge: Cambridge University Press 2009, particularly pp. 12-17. Supreme Court, 1 December 2009, ECLI:NL:HR:2009:BJ2779, NJ 2010/118, with note by Wortmann, and Supreme Court, 22 December 2015, ECLI:NL:HR:2015:3688, NJ 2016/63. On this point and for more information about the background to and reason for this limitation of review in cassation, see: A.J.A. van Dorst, Cassatie in strafzaken, 8th edition, Deventer: Kluwer 2015, pp. 183-184. See A. Cassesse et al., Cassesse’s International Criminal Law, Oxford: Oxford University Press 2013, p. 3. See for example: J. Dugard, Dealing with Crimes of a Past Regime. Is Amnesty still an Option?, Leiden Journal of International Law 1999, pp. 1001-1015; A. O’Shea, Amnesty for Crime in International Law and Practice, The Hague: Kluwer 2002; M. Pensky, Amnesty on trial: impunity, accountability, and the norms of international law, Ethics & Global Politics 2008, pp. 1-40; M. Freeman, Necessary Evils. Amnesties and the Search for Justice, Cambridge: Cambridge University Press 2009; R. van Eijl, Dealing with crimes of the past: no more amnesty? An inquiry into the lawfulness of amnesty for human rights atrocities under international law, University of Utrecht 2009; V. Vriezen, Amnesty justified? The need for a case-by-case approach in the interests of human rights (diss. Tilburg), Cambridge/Antwerp/Portland: Intersentia 2011; L. Mallinder, Amnesties, in: W.A. Schabas & N. Bernaz (eds.), Routledge Handbook of International Criminal Law, London/New York: Routledge 2011, pp. 419-434. G.J.M. Corstens, Het Nederlands strafprocesrecht, edited by M.J. Borgers, 8th edition, Deventer: Kluwer 2014, p. 183. See, above all, the jurisdiction provisions based on the passive personality principle (article 5, paragraph 1 Criminal Code) and the active personality principle (article 7, paragraph 1 Criminal Code). Supreme Court, 18 December 2001, ECLI:NL:HR:2001:AD5211, NJ 2002/300. A comparable position was taken by then Advocate General Langemeijer in his advisory opinion before the Supreme Court judgment of 27 March 1973, ECLI:NL:HR:1973:AB6110, NJ 1973/248, in which he advised that, in the case of an offence that could only be prosecuted under foreign (in this case German) law in response to a complaint, the absence of a complaint was not an obstacle to jurisdiction based on article 5 of the Criminal Code. Cf. J. de Hullu, Materieel strafrecht. Over algemene leerstukken van strafrechtelijke aansprakelijkheid naar Nederlands recht, 8th edition, Deventer: Kluwer 2015, p. 149 and Noyon/Langemeijer/Remmelink, Het Wetboek van Strafrecht, article 5 of the Criminal Code, note 12. In a similar vein, see F. Williams, Immunities and Amnesties, in: N. Keijzer and E. van Sliedregt (eds.), The European Arrest Warrant in Practice, The Hague: T.M.C. Asser Press 2009, pp. 157-178, p. 173 and marginal number 359 above. See J. de Hullu, Materieel strafrecht. Over algemene leerstukken van strafrechtelijke aansprakelijkheid naar Nederlands recht, 8th edition, Deventer: Kluwer 2015, p. 149. Cf. – with further references – V.H. Glerum, De weigeringsgronden bij uitlevering en overlevering. Een vergelijking en kritische evaluatie in het licht van het beginsel van wederzijdse erkenning (diss. Vrije University) Nijmegen: WLP 2013, pp. 585, who also concludes that the framers of the convention did not consider that amnesty negated the criminality of the offence. See article 2 (2) and article 2 (4) respectively of the Framework Decision. For what is therefore a critical view of the ground for refusing surrender, see: V.H. Glerum, De weigeringsgronden bij uitlevering en overlevering. Een vergelijking en kritische evaluatie in het licht van het beginsel van wederzijdse erkenning (diss. Vrije Universiteit) Nijmegen: WLP 2013, pp. 587-588. Cf. in this connection the protection afforded by the ne bis in idem principle in article 3 (2) of Framework Decision 2002/584/JHA. See also section 13a of the General Legislative Provisions Act. Cf. Supreme Court, 18 September 2001, ECLI:NL:HR:2001:AB1471, NJ 2002/559 (consideration 4.4.1.), with note by Reijntjes, and Supreme Court, 8 July 2008, ECLI:NL:HR:2008:BC7418, NJ 2011/91, with note by Klip. Supreme Court 8 July 2008, ECLI:NL:HR:2008:BC7418, NJ 2011/91, with note by Klip. On this point and for information on the judicial scope for assessing the jurisdiction arrangement for conformity with international law, see R. van Elst, Rechtsmacht, in: R. van Elst and E. van Sliedregt (eds.), Handboek Internationaal Strafrecht. Internationaal Strafrecht vanuit Nederlands Perspectief, Deventer: Wolters Kluwer 2015, pp. 73-161, p. 88. See also R. Prouvèze, Immunities, in: W.A. Schabas & N. Bernaz (eds.), Routledge Handbook on International Criminal Law, London/New York: Routledge 2011, pp. 355-367, especially pp. 355-356, and R. van Elst, Rechtsmacht, in: R. van Elst & E. van Sliedregt (eds.), Handboek Internationaal Strafrecht. Internationaal strafrecht vanuit Nederlands perspectief, Deventer: Wolters Kluwer 2015, pp. 73-161, p. 130. Both Prouvèze and Van Elst provide a useful survey of the scope of the immunities recognised in international law. Cf. section 16 of the International Crimes Act, which provides a limited codification of a few immunities. As regards the difference between immunity and amnesty in relation to armed opposition groups, see also my advisory opinions in the Tamil Tigers cases (e.g. ECLI:NL:PHR:2016:967, points 64-67), citing E. Crawford, The Treatment of Combatants and Insurgents Under the Law of Armed Conflict, Oxford: Oxford University Press 2010, p. 79, and S. Sivakumaran, The Law of Non-International Armed Conflict, Oxford: Oxford University Press 2012, p. 520. F. Williams, Immunities and Amnesties, in: N. Keijzer and E. van Sliedregt (eds.), The European Arrest Warrant in Practice, The Hague: T.M.C. Asser Press 2009, pp. 157-178, p. 175. For a similar view, in addition to the positions quoted below, see also A. O’Shea, Pinochet and Beyond: The International Implications of Amnesty, South African Journal on Human Rights 2000, pp. 642-668, especially p. 668 and V. Vriezen Amnesty justified? The need for a case-by-case approach in the interests of human rights (diss. Tilburg), Cambridge/Antwerp/Portland: Intersentia 2011, p. 115. A. Cassesse et al., Cassesse’s International Criminal Law, Oxford: Oxford University Press 2013, p. 312.