The SCSL’s Incoherent — and Selective — Analysis of Custom

by Kevin Jon Heller

As readers know, the Special Court for Sierra Leone’s Appeals Chamber upheld Charles Taylor’s conviction and 50-year sentence yesterday. It’s been interesting to watch human-rights groups and advocates claim, predictably, that the judgment is a milestone in the fight against impunity, a position that seems to wilfully ignore the significant failure of the prosecution in the case. After all, both the Trial Chamber and the Appeals Chamber have now categorically rejected the prosecution’s claim that Taylor was the puppet-master of the atrocities in Sierra Leone, each concluding that he was merely an accessory to the RUF and AFRC’s crimes. And, of course, both the Trial Chamber and Appeals Chamber imposed a sentence 30 years shorter than the one the prosecution demanded. (Taylor’s sentence remains grossly disproportionate, as I explain in my article on the sentence, which was recently published. I’ll have more to say how the AC dishonestly evaded that conclusion in a later post.)

In this post I want to discuss the aspect of the judgment that has already received the most attention: the Taylor AC’s rejection of the specific-direction requirement, which the ICTY (re-) adopted the Perisic AC judgment. According to the Taylor AC, the Perisic AC failed to consider whether the specific-direction requirement was consistent with customary international law:

476. The Perišić Appeals Chamber did not assert that “specific direction” is an element under customary international law. Its analysis was limited to its prior holdings and the holdings of the ICTR Appeals Chamber, which is the same body. Rather than determining whether specific direction is an element under customary international law, the Perišić Appeals Chamber specifically and only inquired whether the ICTY Appeals Chamber had previously departed from its prior holding that “specific direction” is an element of the actus reus of aiding and abetting liability. In the absence of any discussion of customary international law, it is presumed that the ICTY Appeals Chamber in Perišić was only identifying and applying internally binding precedent.

Marko Milanovic notes today at EJIL: Talk! that this is — to say the least — an unconvincing interpretation of what the Perisic AC thought it was doing. As is widely known, the ICTY’s mandate is to apply only those legal principles that are “beyond doubt part of customary international law.” The Taylor AC is thus claiming that the Perisic AC deliberately ignored its own mandate — a troubling claim, especially given that Justices Fisher and Winter insist in their concurring opinion (para. 717) that their colleagues at the ICTY “are sworn to act independently… honestly, faithfully, impartially and conscientiously.”

Underlying the Taylor AC’s accusation of bad faith on the part of the Perisic AC is, of course, the Taylor AC’s conclusion that there is no customary foundation for the specific-direction requirement. Here is what it says:

482. Having considered the Statute and customary international law, the Appeals Chamber finds that the actus reus of aiding and abetting liability is established by assistance that has a substantial effect on the crime, not by the particular manner in which such assistance is provided. The Appeals Chamber rejects the Defence submission that the Trial Chamber was required to find that Taylor provided assistance to the specific physical actor who committed the actus reus of each underlying crime. The Appeals Chamber accordingly affirms its prior holding that the actus reus of aiding and abetting liability under Article 6(1) of the Statute and customary international law is that an accused‘s acts and conduct of assistance, encouragement and/or moral support had a substantial effect on the commission of the crimes charged for which he is to be held responsible.

To begin with, the Taylor AC never explains why the specific-direction requirement has to have a customary foundation. Ad hoc tribunals are limited to applying customary international law because of the nullem crimen sine lege principle: relying on non-customary principles to convict a defendant would convict a defendant of acts that were not criminal at the time they were committed. The specific-direction requirement, however does not expand criminal liability beyond custom; it narrows it. There is thus no reason why the requirement has to have a customary foundation. Unfortunately, like critics of the requirement, the Taylor AC fails to understand the relationship between custom and principles that narrow instead of expand criminal responsibility.

To be sure, the irrelevance of custom does not mean that tribunals should adopt the specific-direction requirement. That is a question for criminal-law theory — what elemental structure does the normative foundations of aiding and abetting require? But at least the Perisic AC attempted to answer that question. The Taylor AC simply ignored it in favour of an irrelevant analysis of customary international law.

And what an analysis of custom it is! To describe the Taylor AC’s analysis of the customary elements of aiding and abetting as shoddy risks flattering the judges. To begin with, consider the AC’s summary of why custom does not require the specific-direction requirement:

474. The Appeals Chamber has independently reviewed the post-Second World War jurisprudence, and is satisfied that those cases did not require an actus reus element of specific direction in addition to proof that the accused‘s acts and conduct had a substantial effect on the commission of the crimes. Similarly, the Appeals Chamber has examined the ILC Draft Code of Crimes and state practice, and is satisfied that they do not require such an element.

The Taylor AC is on firm ground with regard to the post-WW II jurisprudence. Nothing in that jurisprudence supports the specific-direction requirement. But here is Article 2(3)(d) of the 1996 Draft Code, which the AC quotes in full (emphasis added):

3. An individual shall be responsible for a crime set out in article 17, 18, 19 or 20 if that individual…(d) knowingly aids, abets or otherwise assists, directly and substantially, in the commission of such a crime, including providing the means for its commission.

The Taylor AC simply wishes the word “directly” out of existence, reading “directly and substantially” as “substantially.” And it does so (n. 1343) by quoting an ILC comment on a different mode of participation and by citing an ICTY Trial Chamber judgment that concluded the ILC’s use of “direct” could not be interpreted to require causation between the abettor’s act and the abetted crime (see Furundzija, para. 232) — a completely unrelated issue.

Now, I’m not particularly bothered by the Taylor AC’s seemingly-wilful misreading of the 1996 Draft Code. The Code was never adopted by the General Assembly, so it’s not particularly strong evidence of custom anyway. But I find the Taylor AC’s analysis of state practice to be deeply troubling. It is black-letter law that national legislation and jurisprudence provide evidence of state practice and opinio juris. And, of course, the ad hoc tribunals have consistently looked to both when determining the content of customary international law, including the elements of aiding and abetting. But the Taylor AC? Not so much:

429. Domestic law, even if consistent and continuous in all States, is not necessarily indicative of customary international law. This is particularly true in defining legal elements and determining forms of criminal participation in domestic jurisdictions, which may base their concepts of criminality on differing values and principles. Therefore, the reliance by the Defence on examples of domestic jurisdictions requiring or applying a purpose standard to an accused‘s mental state regarding the consequence of his acts or conduct is misplaced.

So: regardless of the “values and principles” that underlie the specific-direction requirement, the Taylor AC can apply the requirement only if it has a customary foundation. But in determining whether the specific-direction requirement has a customary foundation, domestic law is irrelevant, because that law may rest on “differing values and principles.” That’s quite literally nonsense. And even worse, it’s nonsense designed to make defendants as easy to convict as possible.

To be clear, I am not arguing that aiding and abetting under customary international law requires specific direction. I strongly doubt that it does. Still, it’s important to note, as the defence did, that one of the primary reasons you don’t find specific direction in national law is that many states require proof that the defendant intended to facilitate the principal offender’s actions — a requirement that accomplishes the same thing, as the Perisic AC acknowledged. The Taylor AC simply dismisses the importance of those states (para. 430) on the ground that it can equally identify “a number of States that explicitly provide that an accused‘s knowledge of the consequence of his acts or conduct is culpable mens rea for aiding and abetting liability.” In making that argument, however, it simply compounds its mistaken dismissal of domestic law. Now, instead of claiming that domestic law is irrelevant, the Taylor AC contends that a split in domestic law concerning the mens rea of aiding and abetting (intent vs. knowledge) permits it to adopt the broader mens rea — the one that makes it easier to convict. That is precisely the same mistake the Special Tribunal for Lebanon made when it dismissed the requirement, found in the domestic law of many states, that acts of terrorism must be committed for a political or ideological reason.

Ultimately, then, the Taylor AC’s dismissal of the specific-direction requirement rests solely on the post-WW II jurisprudence. The 1996 Draft Code directly (pardon the pun) contradicts the AC’s conclusion, and a significant number of states adopt definitions of aiding and abetting that are functionally equivalent to the one adopted by the Perisic AC.

I don’t have time in this post to discuss whether it is appropriate for an international tribunal to base its customary analysis solely on the judgments of national military tribunals. And in any case, that ship sailed many years ago. (Tadic and JCE III?) But it is important to recognize that the Taylor AC’s reliance on post-WW II jurisprudence to justify rejecting the specific-direction requirement is profoundly hypocritical, because the AC has no problem completely ignoring that jurisprudence when it feels like it. Indeed, that is precisely what it does when it determines that the mens rea of aiding and abetting is recklessness, not knowledge (emphasis added):

411. In addition to these submissions, the Defence argues that the Trial Chamber erred in law by adopting an awareness of the substantial likelihood standard for an accused‘s mental state regarding the consequence of his acts or conduct. In support, it contends that the ICTY‘s jurisprudence provides that an accused must have actual knowledge regarding the consequence of his acts or conduct.


414. The Appeals Chamber does not accept that the mens rea of aiding and abetting liability is a matter of first impression for this Court. The Appeals Chamber, guided by the caselaw of the ICTY and ICTR Appeals Chambers, has consistently held that for aiding and abetting liability under Article 6(1) of the Statute and customary international law, the requisite standard for an accused‘s mental state regarding the consequence of his acts or conduct is as follows:

the accused knew that his acts would assist the commission of the crime by the perpetrator or that he was aware of the substantial likelihood that his acts would assist the commission of a crime by the perpetrator.

This is just stunningly dishonest. Every post-WW II case cited by the Taylor AC against the specific-direction requirement adopts knowledge, not recklessness, as the mens rea of aiding and abetting. Indeed, the Taylor AC does not cite anything other than its own jurisprudence in defence of recklessness — all of the ICTY and ICTR cases it mentions in two footnotes (nn. 1282 & 1283) adopt knowledge. So why, despite the jurisprudence of the post-WW II tribunals, the ICTY, and the ICTR, does the Taylor AC adopt a mens rea of recklessness?

Because, in its view (n. 1289), recklessness is “appropriate for aiding and abetting liability.”

So, to recap. Perisic AC’s belief that specific direction is the appropriate actus reus of aiding and abetting must be rejected, because the specific-direction requirement does not have a foundation in customary international law. But Taylor AC’s belief that recklessness is the appropriate mens rea of aiding and abetting is acceptable, even though it does not have a foundation in customary international law. Or, more concisely: only an aiding-and-abetting requirement that makes it more difficult to convict a suspect requires a customary foundation. An aiding-and-abetting requirement that makes it easier to convict a suspect does not.

Such is the illogic and hypocrisy of the “landmark” Taylor AC judgment.

11 Responses

  1. “Specific direction” is something which Perisic traces back to the relevant paragraph of Tadic (indeed all the supposed case law on the concept pre-Perisic just appears to be boilerplate from Tadic). And that paragraph in Tadic discusses the difference between aiding and abetting and JCE responsibility, in the context of an overall discussion of JCE doctrine generally. That discussion is explicitly based upon customary law. So, it seems to me that customary law is relevant. Highly relevant, in fact.

  2. Response…
    Yes Charles Taylor is guilty of aiding and abetting many of the crimes committed in Sierra Leone. Very doubtful that he planned the actual invasion of Freetown. But the point I want to make is that the Appeals Chamber of the SCSL has never been anything but a rubber stamping machine, confirming some of the worst practices of the court including extremely vague indictments. I don’t have the time to do an article myself, but I would encourage others out there to analyse SCSL Appeals Chamber jurisprudence to see whether it has ever reversed a particular TC conviction. The AC judgement with respect to Gbao is particularly galling but more generally I suspect that it has confirmed convictions with respect to each and every allegations against each and every accused.

  3. Good point, but does the process by which the ICTY identified the customary-international-law elements of aiding and abetting fair much better?  I don’t think it’s a matter of bad faith, but I would say (and have argued) that the real problem is that customary international law simply doesn’t provide answers to many of the detailed individual liability issues these tribunals have had to address.  So what we have instead is various forms of gap-filling by judges who unconvincingly claim to be identifying custom. 
    That said, I do think a specific-direction requirement (or something like it) is appropriate for aiding and abetting cases, even if some ICTY judgments (especially Stanisic & Simatovic) may have gone too far in applying it.  But the argument is more complicated than saying that specific direction is a pre-established principle of customary international law that reflects consistent state practice and opinio juris.

  4. Sasha,

    I completely agree with both points. I am in no way defending the ICTY’s customary analysis, which is also notoriously bad. But I do think we have to distinguish between rules that create or expand criminal responsibility and rules that limit it. The former, it seems to me, must have both a foundation in custom and a theoretical justification, while the latter only need a theoretical justification. The problem — as you rightly point out — is that a strict emphasis on custom for rules that create criminal responsibility would lead to a very desiccated ICL…

  5. an older post at by James Stewart, “‘Specific Direction’ is Unprecedented: Results from Two Empirical Studies” (Sept. 4, 2013) is very useful and demonstrates that claims that “specific direction” is a required element in customary international law are bunk.
    It should also be noted that an alleged requirement of “substantial effect” is bunk.
    Here it should also be noted that “national legislation and jurisprudence provide evidence of state practice and opinion juris” only if there is an expectation that the practice is required by or consistent with international law.  Relevant opinion juris would consist in generally shared patterns of expectation that something was legally appropriate or required as a matter of international law.  A comparative approach is far different, e.g., comparing how domestic legal processes have responded to something using domestic law as such.

  6. Hey Kevin,
    I’m hoping we won’t still be debating this from our respective old people’s homes :)
    I’m travelling at the moment, so just a few quick questions:
    1.     Are national law approaches to complicity really just divided between “intent vs knowledge?” What about systems that require two mental elements, unitary theory systems and those that use “purpose”?
    2.     You say that intention as a mental element “accomplishes the same thing as “specific direction”. Do you stand by this if intention includes dolus eventualis?
    3.     You speak of “a significant number of states” that adopt “functional equivalents” of “specific direction.” Which ones, and how?
    4.     Do you agree that all jurisdictions (international and national) that adopt knowledge as a basis for complicity dilute it to recklessness, too?
    5.     What do you make of the arguments by leading criminal theorists like Kadish, Moore, Alexander and Ferzan (just to name the Americans), who think reckless is acceptable for complicity when it suffices for perpetration?
    6.     You mention that “specific direction” must only have a “theoretical justification”. But does it? If so, please elaborate.
    I don’t agree with everything in Taylor, but without addressing these questions, can you really slam it this hard?

  7. James,

    1. Numerous systems — more common law than civil law — require the accessory to intend to facilitate the principal offender’s conduct. That is all that is relevant to my argument.

    2. No. I am talking about intent as in purpose. But see 1.

    3. All states that require 1. are functionally equivalent to specific direction.

    4. No. Many domestic systems do not. No international system does, with the exception of the SCSL and probably the ICC. No post-WW II system did.

    5. No idea. But not relevant.

    6. I have discussed that previously.

    Implicit 7: regardless of what one thinks about specific direction, the SCSL’s analysis of custom is a joke. So yes, I can slam it this hard.


  8. Hi Kevin,
    I’m afraid I don’t agree with any of your answers to 1 through 7, but I sense we may have done our dash on this issue. Thanks for a fun interchange – I hope the discussion was of some utility to others and look forward to future debates.

  9. Intentionally engaging in conduct while knowing or being aware that one’s conduct can or will facilitate can be different than “intend to facilitate.”

Trackbacks and Pingbacks

  1. […] the Appeals Chamber’s discussion on the specific direction matter here, as does Kevin Jon Heller here. Dov Jacobs’ first thoughts on the judgments can be found […]

  2. […] vigorous rejection of Perisic and the ‘specific direction’ requirement (see Kevin Heller on the SCSL’s incoherent and selective analysis of custom re: specific direction and Marko Milanovic’s post […]