Does Libya Have to Surrender Saif to the ICC? (Answer: Yes)

Does Libya Have to Surrender Saif to the ICC? (Answer: Yes)

Most commentators have assumed — Julian included — that Libya has an obligation under the Rome Statute to surrender Saif Gaddafi to the ICC before it can challenge the admissibility of the case against him.  At The Multilateralist today, David Bosco quotes a UN diplomat who believes that Libya can challenge admissibility without first surrendering Saif:

[Y]esterday, an extremely well informed diplomat at the UN got in touch to point me to a provision in the statute that appears to anticipate the possibility a country holding onto the subject of an arrest warrant while that state argues that it can try the individual. The provision is Article 19(8):

Pending a ruling by the Court, the Prosecutor may seek authority from the Court…(c)  In cooperation with the relevant States, to prevent the absconding of persons in respect of whom the Prosecutor has already requested a warrant of arrest under article 58.

As the diplomat pointed out, this provision is nonsensical if a state must turn over anyone subject to an arrest warrant immediately. None of this means that the Libyan authorities can ignore the court; the ICC judges should, as a legal matter, still have the last word on where he will be tried. But it does mean that they may be within their rights to hold onto Saif while they convince the judges.

Although the Rome Statute is not the picture of clarity on the issue, I do not believe that the UN diplomat is interpreting Article 19(8)(c) correctly.  In my view, that provision envisions a situation in which a state challenges admissibility on the ground that it is investigating a suspect whom it does not yet have in custody.  We don’t normally talk about a suspect “absconding” from custody; “escape,” maybe, but not abscond.  Indeed, the two primary dictionary definitions of “abscond” are to “[l]eave hurriedly and secretly, typically to avoid detection or arrest,” and to “[f]ail to surrender oneself for custody at the appointed time.”  Moreover, if Article 19(8)(c) is designed to address suspects already in custody, why would it mention “the relevant States,” in the plural?  Doesn’t that envision the need for multilateral cooperation to apprehend a suspect whose whereabouts are unknown?  I think the provision would refer to the “relevant State” if it was worried about a suspect escaping from custody.

This interpretation of Article 19(8)(c), it is worth noting, is supported by what seems to me to be the far more relevant provision in the Rome Statute, Article 89 (emphasis mine):

1. The Court may transmit a request for the arrest and surrender of a person, together with the material supporting the request outlined in article 91, to any State on the territory of which that person may be found and shall request the cooperation of that State in the arrest and surrender of such a person. States Parties shall, in accordance with the provisions of this Part and the procedure under their national law, comply with requests for arrest and surrender.

2. Where the person sought for surrender brings a challenge before a national court on the basis of the principle of ne bis in idem as provided in article 20, the requested State shall immediately consult with the Court to determine if there has been a relevant ruling on admissibility. If the case is admissible, the requested State shall proceed with the execution of the request. If an admissibility ruling is pending, the requested State may postpone the execution of the request for surrender of the person until the Court makes a determination on admissibility.

Article 89(1) makes clear that States have an obligation to surrender a suspect for whom an arrest warrant has been issued.  (It refers specifically to States Parties, but there is no reason to believe that Security Council referrals function any differently.)  And Article 89(2) indicates that there is one — and only one — exception to that obligation: where a suspect has already been tried for a particular crime and has been either convicted or acquitted.  Expressio unius est exclusio alterius: if Article 89(2) was designed to permit a State to “postpone the execution of the request for surrender of the person until the Court makes a determination on admissibility” under Article 17(a) or (b) — i.e., prior to conviction or acquittal — it would have said so.

Readers, your thoughts?

UPDATE: It is also worth noting Article 95, which has to be one of the worst written provisions in the Rome Statute:

Where there is an admissibility challenge under consideration by the Court pursuant to article 18 or 19, the requested State may postpone the execution of a request under this Part pending a determination by the Court, unless the Court has specifically ordered that the Prosecutor may pursue the collection of such evidence pursuant to article 18 or 19.

The first part of Article 95 seems to imply that a state does not have to surrender a suspect if it is challenging admissibility.  The second part, however, seems to imply that Article 95 actually applies only to cooperation challenges that deal with the collection of evidence (“collection of such evidence”).  I have no idea how to reconcile those two aspects of Article 95, although I think the narrower reading is better — if Article 95 applies to all admissibility challenges, Article 89(2), which deals specifically with pending ne bis in idem admissibility challenges, is mere surplusage.  That seems unlikely to me.

That said, Article 89(2) literally deals only with ne bis in idem challenges brought by the suspect in national court (whatever that means), so perhaps the drafters intended to distinguish between personal and state challenges — personal challenges being subject to Article 89(2), state challenges being subject to Article 95.  But that doesn’t make much sense.  I don’t see why the two challenges should be treated any differently — and in any case, Article 95 does not apply only to state admissibility challenges.  Personal ne bis in idem challenges are also brought under Article 19.  So the surplusage problem remains.

Print Friendly, PDF & Email
International Criminal Law, International Human Rights Law, Middle East, Organizations
Notify of
Tamás Hoffmann
Tamás Hoffmann

“It refers specifically to States Parties, but there is no reason to believe that Security Council referrals function any differently.”

I beg to differ. I believe that the very rationale of these provisions to ensure that States Parties, who are unwilling or unable to prosecute the suspects (or have referred the situation to the ICC themselves, i.e. relegated their competence to the Court) will not prevent ensuing trials.
In other words, the concept is firmly grounded in the notion of complementarity.

In case of Security Council referrals, the country in question is not a State Party and the Court did not determine before the start of the proceedings that it is unable or unwilling to prosecute the alleged perpetrators. Hence I would argue that if the country can make a prima facie convincing case that it can stage a fair trial then the basic assumption of the SC referral becomes moot…

Mark Kersten

Very interesting. I’m no lawyer, but I do have a quick question on the subject: what is the jurisprudence on this? I know that an admissibility challenge was brought to the ICC in the case of Joseph Kony (who, of course has/had not been detained). My understanding is that ICC Pre-Trial Chamber II ruled on the basis of admissibility in the Kony case pursuant to Article 19.

Mark Kersten

Also, just to muddy the waters a bit more, at this point it’s not clear that Saif al-Islam is in “custody” of Libya and it isn’t even known whether Senussi was detained at all. Saif is in the custody of Zintan rebels, not the NTC and not the government. Is it possible (or just dubious) to argue that Article 19 still applies because the Libyan government, which would lodge the admissibility challenge with the ICC, is not actually in custody of Saif al-Islam or Senussi and therefore “absconding” is still a possibility?

Jens David Ohlin

I’ll have more to say on this in a moment, but my reading of Rome Statute Article 89(2) is that it deals only with defendants who challenge jurisdiction, but in this case, it would be Libya who would be challenging the ICC’s jurisdiction.  Under the Rome Statue, either the defendant or the state can challenge on grounds of the complementarity principle.
Also, we should distinguish between two issues.  One, is Libya required to turn over the defendant simpliciter?  Two, is Libya prevented from challenging the ICC’s jurisdiction before they turn over the defendant?  The second question is really a question about the necessary preconditions for filing a challenge on grounds of complementarity.  Could a PTC deny Libya’s motion on the grounds that the court will not even reach the merits of Libya’s argument until the ICC has custody of the defendant?  Is there a rule or case a PTC could cite for this argument?  I haven’t heard one yet, though I have an open mind.

Jens David Ohlin

Good update.  Thinking…

Jens David Ohlin

Your question as to why is a good one, and I need to consult the commentaries (cassese, triffterer, schabas), which I don’t have in front of me right now.
But here’s my reading of just the provisions in question.  Article 89(1) establishes the duty to surrender the defendant.  Article 95 then carves out an exception to that rule, in cases where there is an admissibility challenge.  The second clause of Article 95 then carves back in the original duty to surrender if the Court so orders it (notwithstanding the pending admissibility challenge), but this carve-in only applies to the collection of evidence, not the surrender of a defendant.  Since the carve-in does not apply, where are left with the carve out, which means the Article 89(1) duty does not apply.

Jonathan Birchall

The ICC’s statement today seems to make it clear that the court believes the requirement to cooperate includes includes surrender of Saif or others, as originally requested:
In accordance with Resolution 1970, adopted unanimously by the United Nations Security Council on 26 February 2011, the Libyan authorities have the obligation to cooperate fully with the Court. On 5 July 2011, a request for cooperation with regard to the surrender of the suspect was notified, together with the warrant of arrest, to the Libyan authorities.

Jens David Ohlin

I am inclined to read Article 89(2) as dealing with defendant challenges to admissibility, and Article 95 as dealing with state objections to requests.  Now, of course, the question is why the reference in Article 95 to both Articles 18 and 19, which collectively deal with both defendant challenges and state challenges?  That’s because a state could conceivably object to a request under Article 95 because the state is objecting to the court’s jurisdiction, or because the defendant is objecting to the court’s jurisdiction. The key thing here is that there is a difference between objecting to the court’s jurisdiction and objecting to the request (for evidence or surrender or whatever).  The two are related but not the same.
For what it’s worth, here is my general analysis of the situation over at LieberCode.
By the way, I’m not entirely certain of any of this, and Kevin’s preferred view may indeed be the correct one, so if anyone has a better idea of explaining all of the relevant provisions of the Rome Statute into a coherent whole, I am all ears.


[…] any admissibility challenges. For a really good exchange of views, see posts by Kevin Jon Heller at Opinio Juris and by the equally excellent Jens David Ohlin (who teaches at Cornell Law School) at his new […]


1. To begin with i would argue that the suspense effect of challenges to admissibility is an inherent general principle of the statute. It is explicitly expressed not only with regard to the prosecutor (Art. 19 (7)) but also with regard to requests directed at states, specifically arrest warrants as provisioned in Art. 89 and Art. 95. In addition – and so far not mentioned in this discussion – the same suspense effect is stipulated (as a mandatory!) in situations of competing requests. See Art. 90 3.: “[the state] shall not extradite the person until the Court has determined that the case is inadmissible“. 2. Which is the applicable provision that allows Libya to suspend extradition ? a) While agreeing with Kevin’s conclusion that Art. 19 (8) is not the applicable provision, i would not make my argument on the base of the finding that the provision doesn’t deal with situations where the suspect is already in custody.  I would rather argue that you can not draw ANY conclusions from Art. 19 with regard to state cooperation, as Art. 19 deals exclusively with the scope of investigative instruments of the prosecution during a pending ruling on admissibility.  b) Art. 89… Read more »


[…] any admissibility challenges. For a really good exchange of views, see posts by Kevin Jon Heller at Opinio Juris and by the equally excellent Jens David Ohlin (who teaches at Cornell Law School) at his new blog […]


[…] any admissibility challenges. For a really good exchange of views, see posts by Kevin Jon Heller at Opinio Juris and by the equally excellent Jens David Ohlin (who teaches at Cornell Law School) at […]


[…] Three – Saif al-Islam Gaddafi and Abdullah al-Senussi – is currently raging (see here, here and here). It’s an important debate, and, as readers will know, I have chirped in as […]