LJIL Symposium: Names, Labels, and Roses

[Jens David Ohlin is Associate Professor of Law at Cornell Law School; he blogs at LieberCode]

In his excellent essay, James Stewart advocates for a unitary model of perpetration. To the extent that this means the end of modes of liability, so be it says Stewart. We don’t need them. They codify distinctions that we don’t need, promote confusion over coherence, and so we should instead streamline the centrifugal doctrines into a single account of causal contribution. On the elegance scale, Stewart’s proposal should score a 10 from most judges.

Stewart pitches his account as revisionary, an attempt to right the ship after years of confusing scholarly and judicial debate about modes of liability and the difference between principals and accessories (or other categories that occupy similar conceptual space). But I think that it is the wrong light in which to see the argument.

I see Stewart’s proposal as urging return to a substantially similar state of affairs under the original Joint Criminal Enterprise scheme proposed by the Tadic Appeals Chamber during the early days of the ICTY. Cassese was the prime mover behind the JCE doctrine, and it covered all members of the collective endeavor, regardless of their level of contribution. Eventually, the doctrine was modified to require a heightened contribution requirement, and eventually the leadership level defendants were “de-linked” from the foot soldiers and placed in separate JCEs. But the important point is that the original JCE doctrine included everyone from an architect of the crime (mastermind or hintermann) as well as the foot soldiers or what the later ICTY cases often referred to as the Relevant Physical Perpetrators, or RPP.

So under the original JCE doctrine, each member of the group was prosecuted for participating in the JCE. That was, in essence, a unitary model of perpetration. True, as a formal matter, aiding and abetting and accomplice liability survived the creation of JCE, but their relevance and practical import was greatly reduced. Most defendants at the ICTY were prosecuted under a JCE theory and it seemed to me that in most cases JCE could have replaced the other modes of liability given the collective nature of international crimes.

The great objection to this early notion of JCE was that it ignored differentiation between different levels of culpability, especially as between principal perpetrators and foot soldiers best described as accomplices. All of them could be described as participants in a JCE. I pressed this point several times with Cassese, and his answer was always the same: let the Trial Chamber sort out these distinctions at sentencing. I take Stewart to be saying roughly the same thing. To the extent that his unitary model fails to make these distinctions, let the sentencing reflect them.

So why doesn’t this solve the problem, either for Cassese or Stewart? First of all, it proves too much. There’s something significant about the criminal law’s need to classify perpetrators and codify their relative culpability with different labels. Moreover, this function needs to take place on the side of the substantive doctrine, not the side of sentencing. Otherwise, as I have said before, we could eliminate the crimes themselves and replace them with a global crime called “Felony” (or in ICL “Atrocity”) and let all other matters relevant to culpability get decided in sentencing.

Of course, this would violate the principle of legality. The sentencing phase is characterized by a notoriously high degree of discretion. As I have complained before in my writings, it’s not clear to me that the results in sentencing are always coherent and well-reasoned, and some of that result might be implicit in the very process of trying to apply a numerical term (years in prison) to an inherently qualitative function: culpability or guilt. It’s not an easy task.

So as a practical matter, legality is strengthened when objective doctrines, with particular elements, are used to make these determinations regarding relative culpability. The results are likely to be more governed by reasons and less governed by ad hockery.

Second, there is a more a priori argument that concludes that the criminal law ought to codify these distinctions in its substantive doctrine. When the doctrine collapses the distinction between principals and accessories, or between masterminds and instruments, it falsely labels both as being roughly equivalent. In so doing it sends the wrong message to the world. Scholars often refer to this normative requirement as “Fair labeling” – a term of art that has gained increased prominence in this debate. So Stewart’s model, and its single label, has the effect of inflating the culpability of the accessory and deflating the culpability of the principal, since both are described under the same name. For criminal lawyers, labels matter, and it is no consolation that the sweetness of the rose will be correctly evaluated during sentencing.

Stewart has a few answers to this objection. First, he points out that “dispositions” in international judgments rarely mention the mode of liability. If by disposition he means the last paragraph of the judgment where the result is stated, this seems unnecessarily formalistic, since the mode of liability clearly looms large in the document as a whole. Second, he claims that these labels are inscrutable for lay audiences, though that strikes me as demolishing a straw man – no one claims that fair labels must be understood, in all their complexity, by the general population. In any event, I think it is uncontroversial that the general public understands that an “accomplice” is something less than a mastermind.

Of course, there is always the problem that international law is attempting to develop a universal Dogmatik even though different legal cultures use radically different concepts and terminology. So in that sense the problem is even worse than Stewart describes. It’s not just the laity that won’t understand what an “indirect co-perpetrator” is. Even a criminal lawyer practicing law in the United States wouldn’t understand what this phrase means, unless they happened to take criminal law with me, George Fletcher, Mark Osiel, Markus Dubber, and a handful of other criminal law scholars who bother to explain the continental terminology to first-year students. Or they happened to take an elective class in ICL. But to most American lawyers, it’s all Greek to them because these terms hail from a different legal culture.

But does that really matter? Again, the argument proves way too much. The laity doesn’t understand most legal categories that are used in any field of law, whether property, anti-trust, or intellectual property. The guy on the street understands the Rule Against Perpetuity? I don’t think so. She understands the concept of collateral estoppel? Hardly. The law expresses all sorts of things, but that isn’t to suggest that the message will come in the form of an easily digestible Tweet.

Of course, one could revise the law to downplay as many of these terms of art as possible. We could replace most distinctions in tort law and have the court decide if the defendant is a “Tortfeasor” with a capital T and decide how much he should pay. Such a move would bring us much closer to the equity courts of yesteryear, and turn back the clock on the best developments that the path of the law has bequeathed to us.

All of this is to say that I obviously don’t share the underlying impulse behind Stewart’s proposal. I’m opposed to it at an atomic level, if that isn’t already clear by now. However, I do have to say that Stewart’s contribution is that he has cast the best possible light on this position and defended it with great rigor, sophistication, originality, and clarity. I wish Cassese had lived long enough to see it in print. Although Stewart is advocating for a replacement of JCE – something that Cassese would no doubt object to – I do believe that the “unitary” aspect of Stewart’s model is actually very close to the animating spirit behind Cassese’s original doctrine when he drafted the Tadic decision.