Complementarity and the German Amicus Curiae Submission in the ICC Palestine Arrest Warrant Proceedings

In this post I will discuss the complementarity objections expressed by Germany and some other participants in the ongoing ICC arrest warrant proceedings in the Palestine situation. These objections only refer to the arrest warrant requests of 20 May 2024 concerning Prime Minister Netanyahu and Defence Minister Gallant. I will start with a critique of the lack of transparency regarding the amicus curiae procedure. I will then present some procedural considerations before discussing the actual complementarity issues. After summarizing the German position, I will offer a critical assessment, taking into account other relevant submissions. In a nutshell, while presenting some valuable and interesting arguments which deserve to be carefully weighed by the Pre-Trial Chamber (‘PTC’) I, on a more principled level, the German position is in line with the infamous Staatsräson doctrine representing Germany’s foreign policy towards Israel. There is strong, almost unconditional support, coming close to a primacy of politics over law, which entails interpreting international (criminal) law in a sort of most-favored-nation treatment of Israel.

Lack of Transparency

The usual practice at the ICC is to publish amicus curiae submissions – as, in principle, all fillings in line with the principle of publicity – immediately on the website, especially if they, as here, only deal with legal issues. Notwithstanding, the PTC I has initially classified all filings in these proceedings as secret (see UK Request for Leave, para. 29), and this has rightly been criticized by the UK as unnecessary which therefore requested a reclassification as public (ibid.). The PTC I granted this request (here, para. 7) and later decided that the actual submissions (“observations”) have to be filed publicly (here, Disposition). Thus, PTC I originally distinguished between the requests for leave, classifying them initially as secret, and the actual amicus curiae submissions that, from the start, had to be filed publicly. This lack of transparency and inconsistency seems weird and calls for an explanation. It is hoped that PTC I will reclassify all requests for leave as public as soon as possible.  

Germany has filed its original amicus curiae submission, as is apparent from its title, as “public” with the date of 6 August 2024, and it was received by the Court on 7 August (see top right of each page of the submission). It was then publicly available on the Court’s website on 9 August. Speculation about the content of the submission has been circulating since mid-July (see here and Talmon), but the Federal Foreign Office (as the responsible Ministry) did not inform the interested public of the above-mentioned procedural intricacies. It is also strange that the submission appears in the “Court Records” without any indication of the source – simply as “Observations pursuant to Rule 103(1) of the Rules of Procedure and Evidence”; only by downloading the actual submission, it becomes clear that it is from Germany. In the case of other submissions, the source usually appears already in the “Court Records” list (see, for example, the statements from the USA, the Israel Bar Association or Adil Ahmad Haque). Finally, it also needs to be explained why Germany reflected the situation name on the cover page as “Situation in Palestine” instead of “Situation in the State of Palestine” (as in the submission from 2020). Does Germany want to make its non-recognition of a Palestinian State explicit?

Procedural Considerations

Procedurally, a Chamber “may” admit such a submission “at any stage of the proceedings” if it considers this to be “desirable for the proper determination of the case” (Rule 103(1) of the ICC Rules of Procedure and Evidence, ‘RPE’). In our context, however, two limitations must be taken into account. First, complementarity challenges under Articles 18 and 19 of the ICC Statute (‘ICCS’) can only be brought by a State that “would normally exercise jurisdiction” or “which has jurisdiction” with regard to the respective situation or case (Articles 18(1), 19(2)(b) ICCS). Thus, strictly speaking, only such a State has a right to intervene on the basis of complementarity arguments, at least if it intends to challenge admissibility (the counterargument would then be that a State submitting amicus curiae observations does not challenge admissibility but just present its views on the matter). Secondly, there is only limited room for complementarity considerations within the framework of arrest warrant proceedings. Article 58(1) ICCS requires “reasonable grounds to believe that the person has committed a crime within the jurisdiction of the Court” and that the arrest “appears necessary”. Complementarity is, in principle, not an issue here, but it may be addressed by a PTC proprio motu (for an excellent early discussion, see El-Zeidy). We will return to a PTC’s limited discretion in that regard below when assessing the German complementarity objections.

In the present case, the United Kingdom (still under a Conservative government) originally submitted such an application on 10 June 2024 (see here the version made public per order of Pre-Trial Chamber [‘PTC’] I on 27 June 2024) and PTC I granted the application on 27 June 2024 with a filing deadline of 12 July 2024 (see here). This triggered a flood of statements (see Court Records), although the UK itself (now under a Labour government) later withdrew from the proceedings (see here). Since the Chamber did not limit the subject matter of the submissions to the question of jurisdiction (in connection with the Oslo Peace Accords), as originally raised by the UK, numerous submissions went beyond this. Also, Germany, which requested its participation (“request for leave”) on the last day (!) of the above-mentioned deadline (12 July 2024), did not deal at all with the question of jurisdiction (unlike in 2020, see critically Schabas, para. 12 ff.). The Chamber granted leave on 22 July 2024 with a deadline of 6 August 2024 (see Submission, para. 2-3). Germany then submitted on exactly that day.

In the Court’s previous practice, no external (amicus curiae) statements were permitted in the context of an arrest warrant application procedure (see Vasiliev with references; see also here, para. 6 regarding the suspect’s lack of standing). A key reason for this is the ex parte nature of these proceedings (see, e.g., here, para. 6, 18 and here, para. 9) and, probably, the procedural delay caused by such statements. In the present case (critically also Roth and Vasiliev), the PTC not only had to extend the filing deadlines for post-UK requests to leave (and some, like Germany, have exhausted the deadlines granted), but it is now also obliged to process hundreds of pages of statements and at least give the prosecution the opportunity to reply (Rule 103(2) RPE; done per Order of 9 August until 26 August, para. 8). Recall that usually, the period between the application and the issuance of an arrest warrant amounts to several weeks, for example in the case of Al-Werfalli about two weeks (see here), in the case of Saif Al-Islam Gaddafi just under six weeks (here), in the case of the rebel leader Joseph Kony around eight weeks (here), and the case of Russian President Putin only three weeks (here). The case of former Sudanese President Omar Al-Bashir, where the period was eight months (see the application here and the issuance here), is the exception to the rule.

The (German) Complementarity Objections

In a nutshell, the German Federal Government argues – invoking the principle of complementarity that governs the relationship between the ICC and national criminal justice systems (Articles 17-19 ICCS) – that Israel must be given a real opportunity and more time to pursue criminal investigations and prosecutions itself. Four arguments are presented:

(1) The complementarity test must take into account whether the State in question is a Rechtsstaat (complies with the rule of law) and has an independent judicial system that investigates and tries statutory crimes:

“… the Court should take into account whether the State is committed to the rule of law, whether it has a robust and independent legal system and whether that system is actively examining, investigating and reviewing a wide range of issues and allegations relating to potential violations of international humanitarian law.” (para. 9).

(2) Furthermore, a State that is currently engaged in an armed conflict should be treated with greater leniency:

“… where a State – such as Israel – is subject to an ongoing armed attack and faces serious threats from additional actors, this State should be given an appropriate and genuine opportunity to put its accountability mechanisms into action before the Prosecutor may request warrants for arrest under Article 58 of the Statute.“ (para. 10).

(3) When applying the complementarity principle, it must be ensured – in the sense of the primacy of the territorial State – that this State is given an appropriate and genuine opportunity to present its national investigative and legal protection mechanisms, especially if, like Israel, it has made clear its willingness to cooperate with the ICC:

“In other words, where a State is willing to cooperate with the Prosecutor in a given situation – and it is  our  understanding  that  Israel  had indicated willingness to do so in the situation at hand – Article 17 should be interpreted, based on the  principle of good faith, as required under Article 31(1) of the Vienna Convention on the Law of Treaties, with a view to ensuring that this State receives an appropriate and genuine opportunity to present its domestic investigation and legal review mechanisms with regard to the allegations at hand.” (para. 12)

(4) Article 18 ICCS, which regulates the complementarity mechanism for situations (as opposed to specific cases regulated by Article 19), must be interpreted “in a substantive sense” to mean that a State must be given a new or second opportunity to invoke the complementarity objection in the event of a fundamental change of a factual situation that requires a “new investigation” (as was the case with the Hamas attack of 7 October 2023):

“When an initial investigation is subject to significant change over time due to a fundamental change in the factual situation – thus making it, in substance, a new investigation – the State concerned should anew be given an appropriate and genuine opportunity to inform the Court about its accountability mechanisms. With regard to the Situation in Palestine … Germany is of the view that the attack by Hamas brought about such fundamental change in the situation that a new notification was required which would have given the State concerned the procedural opportunity to request that the Prosecutor defer to the State’s investigation.” (para. 14)

Critical Assessment

Before dealing with the four arguments presented by Germany it should be recalled that a PTC is, in principle, as already said above, entitled to address admissibility proprio motu in arrest warrant proceedings, but it has a limited discretion in that regard. As held by the Appeals Chamber in the DRC situation already in 2006 (para. 52):

“… the Pre-Trial Chamber should exercise its discretion only when it is appropriate in the circumstances of the case … Such circumstances may include instances where a case is based on the established jurisprudence of the Court, uncontested facts that render a case clearly inadmissible or an ostensible cause impelling the exercise of proprio motu review.”

None of these (three) circumstances is in any way linked to the German arguments.

1. The Rule of Law Argument

Germany does not pretend that this argument has a basis in Article 17 ICCS. In fact, the principle of complementarity is neutral vis-à-vis the political system (democratic or not) of the respective State and its rule of law credentials or deficits. It is exclusively or at least primarily concerned with the States’ duty and willingness/ability to bring the perpetrators of international crimes to justice by investigating, prosecuting and trying them (Art. 17(1)(a)-(c) ICCS). The due process reference in para. 2 is thus not to be understood as a fair trial requirement stricto sensu but as an expression of the respective State’s genuineness in bringing those responsible to justice. While the Court has to undertake an assessment of the quality of domestic justice within the framework of the unwillingness test (Article 17 (2)), considering human rights standards (see also Article 21(3) ICCS), a possible due process violation shall not be decisive for the admissibility determination (Schabas/El Zeidy, in Ambos, ICC Commentary 2022, Art. 17 mn. 75). In short, Art. 17 is about admissibility, not due process (Ambos, Treatise ICL III 2016, 313).

To be sure, a “robust and independent legal system” may be considered a prerequisite of effective criminal investigations but its mere existence, in abstracto, does not suffice to comply with the obligations under the complementarity principle. It is indeed upon the State to demonstrate that this “system is actively examining, investigating and reviewing a wide range of issues and allegations relating to potential violations of international humanitarian law.” (Germany, para. 9). A general deference to ‘advanced’ legal systems would unfairly privilege these systems as compared to less stable and/or democratic systems which still could, however, comply with the complementarity requirements. While Israel has, in principle and despite counter-reform efforts of this government, an independent justice system (see, generally, Israel Bar Association, para. 2, 4 ff.; more specifically, Centre for Israel and Jewish Affairs, para. 22; Cohen/Shany) and, perhaps more importantly in this context, a solid military justice system (cf. NATO High Level Military Group [HLMG], para. 25 ff. [“… IDF military justice and accountability mechanism …  consistent with the highest standards of our own armed forces.”]), the key question in the complementarity context is, especially at the case-related stage of arrest warrant proceedings, whether this system concretely investigates the alleged crimes (persons/conduct). We will return to this question below (3.).

2. The Armed Conflict Argument

This argument has also been put forward by the just mentioned HLMG (para. 24: “unrealistic to expect such action [investigation etc.] during the midst of a military campaign of this magnitude …”) and by the Democratic Republic of the Congo (DRC) (para. 22: invoking, similar to Germany, a “considération substantielle” especially with regard to an “état d’hostilités”). Yet, it did not come up in other situations where (non-international) armed conflicts take place (including in the DRC!). The most obvious situation is perhaps the Russia/Ukraine war, where the argument has not been invoked by/in favour of either the aggressor or the victim State. In fact, Israel itself does not argue that it remains inactive with regard to its IHL or ICL obligations. In the ICJ proceedings in South Africa vs. Israel, the latter is actively engaged and stated several times that it “is taking steps to try and contend with the massive complexity that such a situation [Gaza] presents.” (see here, para. 19). As to the ICC proceedings Israel showed great surprise as to the arrest warrant application since it claims to actively cooperate with the Prosecutor (see Cohen/Shany). The Israel Bar Association explicitly mentions in its amicus curiae submission, “significant law enforcement actions are taken while Israel remains in a state of war” (para. 26; similarly, Rubinstein et al., para. 12; Touro Institute, para. 28).

At any rate, while the specific context in a given situation must always be taken into account, there is simply no legal basis for a general armed conflict exception or reservation within the framework of complementarity proceedings. In fact, such an exception would be fundamentally inconsistent with the idea of a preventive effect of ICL investigations. Just recall that the ICTY was established during an ongoing conflict in the belief that its establishment “will contribute to ensuring that such violations are halted and effectively redressed.” (S/RES 827 (1993), Preamble). In other words, acknowledging the fact that an armed conflict creates particular challenges to the workings of criminal justice (and to many other State services) does not mean that these services are or should be suspended altogether.

3. Primacy of Territorial State in Case of Willingness to Cooperate

This argument has two sub-claims. The general one is that it follows from the general primacy of domestic jurisdictions under the complementarity regime that certain deference is to be given to a State which has “indicated willingness” to cooperate, i.e., that “this State receives an appropriate and genuine opportunity” to present its investigative activity. This claim finds, in principle, support in the new 2024 Policy on Complementarity and Cooperation of the Office of the Prosecutor (OTP), which stresses the role of States in the prosecution of international crimes and the importance of a constructive relationship of cooperation between them and the OTP (for a good discussion Labuda). As a kind of minimum standard of such a cooperative complementarity model – which, noted in passing, draws on the earlier ideas of “partnership”, “vigilance” and “positive complementarity” (see here and Ambos, Treatise ICL III 2016, 327 ff.) – one may indeed infer an obligation of the Prosecutor to give States enough time to explore all domestic options and take into account the specific circumstances, e.g. the existence of an armed conflict (in this vein DRC, para. 16 indicating in total six “normes minimales”).

However, this sub-claim is (too) general and premised on the respective State’s willingness to cooperate. In fact, it is intrinsically linked to the specific sub-claim, namely that “Israel had indicated willingness” to cooperate. In this vein, the Israel Bar Association affirms that Israeli law enforcement authorities have demonstrated a willingness to cooperate with non-Israeli authorities (para. 27). Given that any cooperation at this stage of proceedings is confidential, it is difficult, if not impossible, to verify this claim. There are rumours that Israel’s Attorney General Office was in close contact with the OTP with a view to sharing information about the different domestic investigations and an official OTP visit was scheduled the day the arrest warrant application was issued (Touro Institute, para. 27). Further, the US claims (submission, para. 9,16) that Israel submitted a formal letter requesting a deferral that the OTP ignored. If such a formal request indeed existed (and one accepts the argument of a new situation, thereto below 4.), the Prosecutor would have been obliged to “defer to the State’s investigation” and request an authorization from the PTC to continue its investigation (Article 18(2) ICCS; also USA, para. 25). This writer has not seen this letter but people who have seen it have characterized it as just a boilerplate “please defer because we have a functioning judiciary”. According to these same people, the letter doesn’t mention Article 18, a new investigation or any concrete suspects. If that is true, the letter would not comply with Article 18(2) ICCS, which requires that the respective State, quite concretely and precisely, informs “the Court that it is investigating or has investigated its nationals or others within its jurisdiction with respect to criminal acts which may constitute crimes referred to in article 5 and which relate to the information provided in the notification to States.”

This brings us to the State’s burden to provide specific information as to investigations carried out in a given situation (generally on the State’s burden of proof Kenya Appeal Judgment, para. 62) – which should not be confused (as apparently done by Germany) with a State’s general willingness to cooperate and/or unspecific promises to investigate/prosecute. In line with the ICC’s two steps test (here, para. 147 ff., Ambos, Treatise ICL III 2016, 296), only if (first step) there is some investigative activity on the part of the respective State (in line with Article 17(1)(a)-(c) ICCS) the OTP must (second step) defer to the domestic proceedings unless the unwillingness/inability exception (Article 17(2), (3)) applies. Once an investigation has advanced from a situation to a concrete case – as specifically in the case of arrest warrant applications – the “same person/same conduct” test applies, i.e., “the national investigation must cover the same individual and substantially the same conduct as alleged in the proceedings before the Court.” (Kenya Appeal Judgment, para. 40; critically DRC, para. 17 ff.). While the “same person” requirement is straightforward, the “same conduct” is more normative, looking at concrete incidents which “sufficiently mirror[s]” or “overlap” with the ones investigated by the OTP (Gaddafi Appeal Judgment, para. 72, 73; confirming the “mirroring” test Venezuela I Appeal Judgment, para. 10, 255 and passim). In other words, Israel would have to show that it investigates Prime Minister Netanyahu and Defence Minister Gallant (same persons) for the allegations set forth in the arrest warrant applications (same conduct), i.e., especially the war crime of starvation of a civilian population (Article 8(2)(b)(xxv) ICCS).

To comply with the investigation requirement, it is not sufficient to only open an investigation but to take concrete steps to advance this investigation (Kenya Appeal Judgment, para. 40: “investigative steps … actually taken”; for a discussion, see Heller: “‘tangible, concrete and progressive’ investigative steps aimed at eventual prosecution”). Thus, the key question then is whether it is conceivable that Israel undertakes an investigation into a policy of starvation in Gaza that emanates from its leadership. While there may be some investigative activity (see e.g. DRC, para. 21), this is very unlikely given Israel’s poor track record in investigating and prosecuting crimes committed against Palestinians (crit. here, here and here; also Chile/Mexico, para. 34: “no information that … Israel has started any prosecution …”) and the current enforcement problems with regard to the investigation of the alleged torture of Palestinian detainees. I note in passing that this raises ability concerns within the meaning of Article 17(3) ICCS: If Israel is not able to properly investigate these allegations due to right-wing protests, this may amount to a “collapse or unavailability of its national justice system” (cf. Ambos, Treatise ICL III 2016, 317 ff.; affirming ability, however, from a military perspective see HLMG, para. 28 [“We do not believe there is a credible basis to conclude Israel lacks the ability or will to implement national investigatory and judicial processes that are comparable to other countries and their militaries.”]).

Perhaps Israel may employ alternative investigation mechanisms falling short of a criminal investigation stricto sensu, e.g., a commission of inquiry, but then the question arises of whether such mechanisms comply with the investigation requirement of Art. 17(1) ICCS. While this is not a wholly untenable position (in favour Cohen/Shany), the ultimate objective of such alternative mechanisms must necessarily be a criminal prosecution (see Ambos, Treatise ICL III 2016, 303-4; similarly Cohen/Shany [Commission of Inquiry „should be explicitly authorized … to make recommendations to the state prosecution regarding follow-up action required in order to establish the potential criminal responsibility of specific suspects.”]). For this reason, mere administrative or constitutional proceedings, such as the current petition regarding the humanitarian situation in Gaza by civil society organisations to Israel’s Supreme Court (see Israel Bar Association, para. 29 ff.), fall short of what is required by complementarity.   

4. Fundamental Change of the Original „Situation“ within the meaning of Article 18 ICCS?

So far, the ICC case law and practice has broadly interpreted the term “situation” in Article 18. The term is originally mentioned in Article 14 but nowhere defined in the Statute. It refers to “a description of facts, defined by space and time, which circumscribe the prevailing circumstances at the time (‘conflict scenario’) …” (Chaitidou, in Ambos, ICC Commentary 2022, Art. 14 mn. 29). The Court defined the concept by reference to the “territorial, temporal and possibly personal parameters” (ibid. mn. 31 with references in fn. 119). The “situational linkage” jurisprudence, focusing on the alleged link of new crimes to a “situation of crisis” (ibid. mn. 37-8), even broadened the situation concept further. As a consequence, the OTP has only exceptionally opened new situations referring to a territorial area already under situation examination. Just recall that even the Russian invasion of Ukraine in February 2022 did not entail the opening of a new situation, but the Prosecutor linked its (new) March 2022 investigation to the original situation going back to 21 November 2013 (here). In fact, the OTP only opened new situations with regard to the Central African Republic (CAR II) and Venezuela (Venezuela II). As to the former situation then Prosecutor Bensouda saw a relevant distinction as to CAR I “both with regard to the armed groups involved, and the political context of the events.” (here, para. 9). As to Venezuela, the distinction between the two situations is fundamental in that Venezuela I refers to alleged crimes of the Maduro government, while Venezuela II to alleged crimes against this government by the US.

What now follows from this for the Palestine investigation? It was formally initiated, and the relevant States were notified in March 2021 (here). The one-month period (Article 18(2) ICCS) passed without a complementarity objection by Israel or any other State. Is it now possible to reclassify the October 7 Hamas attack as a new Palestine situation? Does it represent “a fundamental change in the factual situation” as argued by Germany and would this justify or even require the opening of a new situation? The original investigation covers Statute crimes allegedly committed in Palestine, i.e. West Bank (including East Jerusalem) and Gaza, since 13 June 2014. On 20 December 2019, then Prosecutor Bensouda concluded on the basis of the preliminary examination (on the distinction to a formal investigation, see Ambos, Treatise ICL III 2016, 335 ff.) that there was a reasonable basis to believe that war crimes have been committed in Palestine (see here). Bensouda referred in particular to possible war crimes by the Israel Defence Forces (IDF) in the context of the 2014 Gaza hostilities, the 2018 Gaza border fence demonstrations and the occupation of the West Bank (para. 2, 4, 5). She also referred to war crimes committed by Hamas and other Palestinian armed groups (PAGs) (para. 3). She stressed that these crimes are “illustrative only” (para. 9).

Against this background, it is difficult to argue that the 7 October attack brought about a change of such a fundamental nature that a new situation within the meaning of Art. 18 had to be opened. The post-7 October investigation still concerns the same parties, namely Hamas/other PAGs and the IDF. While the nature of the Hamas attack was certainly different in scope and impact than previous attacks, it was still an attack concerning the same adversary (Israeli civilians and combatants), objects (civilian and military ones) and the same territory (mainland Israel); similarly, the Israeli reaction, while much more prolonged and intense, corresponds to earlier strikes reacting to Hamas (rocket) attacks. From this perspective, the original Palestine/Israel situation has not changed, at least not in terms of the relevant “situation of crisis” giving rise to possible Statute crimes (for a different view, possibly inspiring the German position, Shany/Cohen here and here, para. 24 [“major differences between the factual patterns and categories of suspects”]; in a similar vein, USA, para. 16 ff. [“entirely new focus”, para. 19]; Rubinstein et al., para. 11 ff.; The Hague Initiative for International Cooperation, para. 30 ff.; similarly, DRC, para. 16, 20 ff., 33; in stronger terms Touro Institute, para. 26 ff. [“violates … complementarity”] and Centre for Israel and Jewish Affairs, para. 14 ff. [“pre-emptive, preclusive, and prejudicial nature of the process”], but “an entirely new set of allegations”, para. 27, certainly does not suffice). If PTC I adopted the ”fundamental change” argument, it would have to reconsider the Ukraine situation. For if an attack by a non-State actor like that of 7 October gave rise to a new situation, a full-fledged military invasion by a State would, a fortiori, do so.

While the “fundamental change” argument is of a more substantial nature, it is rather formal or procedural to argue, as the USA (para. 22) and DRC (para. 28 ff.) do, that the collective State Party referrals of 17 November 2023 and 18 January 2024 warranted a subsequent notification by the Prosecutor. At first sight, Article 18(1) ICCS seems to support this view since it requires the Prosecutor to “notify all States Parties and those States which … would normally exercise jurisdiction over the crimes concerned” when a situation has been referred under Article 13(a) ICCS, i.e. by one or more State Parties. Yet, the gist of the matter seems to be whether any State referral triggers the notification obligation – in the sense of an “article 18 dialogue through a formal and sufficiently detailed communication to relevant States” with a view to the one month response period in Article 18(2) ICCS (USA, para. 22) – or whether such a notification obligation only arises when a State refers a wholly new situation or at least a situation which implies a fundamental change of an already existing situation. The example of the (second) DRC self-referral of May 2023 speaks – contrary to the DRC submission (para. 30) – more in favour of this latter view since, in this case, the Prosecutor did not notify States but only the ICC Presidency. Also, in the Ukraine situation the massive collective State referrals starting on 1 March 2022 did not have any impact on the determination of the (original) situation. At any rate, this further argument is important enough to be dealt with by PTC I.   

A last reflection on the relationship between Articles 18 and 19 ICCS. Irrespective of a possible second chance under Article 18(2), at the stage of arrest warrant proceedings, we have a case against one or several suspects and Article 19 applies. The admissibility can then still be challenged, either by the suspect/accused (Article 19(2)(a)) or a State concerned (Article 19(2)(b), (c)). Thus, a restrictive interpretation of Article 18 does in no way limit the suspect’s/State’s right to challenge the admissibility of the concrete case and this right exists at least until the commencement of the trial (Article 19(4)).

I thank several colleagues for commenting on a draft of this piece.