A Comparative Perspective on the “Purpose” Requirement

Dr Cassandra Steer – Cassandra is the author of a forthcoming book Translating Guilt: Identifying Leadership Liability for Mass Atrocity (T.M.C Asser Press, 2015). She is a Lecturer and Researcher at the Amsterdam Centre for International Law, but about to join McGill Law School.

As with so much terminology in the Rome Statute, the debates on the content and meaning of Article 25 are fueled by problems of translating domestic criminal law notions to the international context. It is tempting to copy-paste the MPC approach because the wording looks similar, however the Rome Statute is a patchwork of legal traditions, and no single domestic interpretation will provide the clarity that is sought. Concepts and words in one system do not necessarily have the same connotations as they do in others. The concern that the Rome Statute departs from customary law is in my view not the central issue, since one could argue that the case law of the ad hoc tribunals does not itself amount to custom; all that has emerged is differing trends in different international and hybrid tribunals, each of which draw on various domestic criminal law models. Instead I argue a broader theory of liability should be applied to solve questions of interpretation; namely a theory that differentiates between principal and assistant liability.

Aiding and abetting comes from old English common law terminology, when there used to be a legal distinction between principal in the first degree (the physical perpetrator), principal in the second degree (anyone who was physically present and aiding the crime), and an accessory who was, according to the Blackstone commentaries; “not the chief actor, nor present at its performance, but is some way concerned therein, either before or after the fact”. The distinction was made because the death penalty applied to every felony for principals, but not for accessories. However over time this legal distinction was eradicated in the common law tradition. Although the terms “principal” and “accessory” remain, every actor is considered equally liable for the full commission of the crime, regardless of his or her actual contribution. This amounts to a functionally unitary system.

In the US, the MPC attempted to make a clear distinction between the fault elements ranging from intent, purpose, knowledge to recklessness, as applicable to different crimes, however it is unclear whether there should be differing requirements for different participants in a crime, given that there is no legal distinction between them. James Stewart’s argument for a dynamic system of volitional requirements would make sense in this context; whatever is required for the crime should apply to all participants.

The different interpretations of “purpose” that appear in US case law with respect to aiders and abettor agree that in any case it should not be seen as equivalent to motive. As one judge put it in a 1940 case, even the person who sells a gun to another, knowing it will be used for murder, cannot escape liability by saying the gun was sold merely for profit and not for the purpose of the crime:

“Guilt as an accessory depends, not on ‘having a stake’ in the outcome of the crime [. . . ] but on aiding and assisting the perpetrators; and those who make a profit by furnishing to criminals, whether by sale or otherwise, the means to carry on their nefarious undertakings, aid them just as truly as if they were actual partners with them” (Backun v United States (1940) 112 F2d 635 (Fourth Circuit Appeals Court) p 637.)

This somewhat loose standard was refined by Judge Learned Hand in Peoni: the accused must “in some way associate himself with the venture, that he participates in it as something he wishes to bring about, that he seeks by his action to make it a success. All the words used [. . . ] carry an implication of purposive attitude towards it.” (United States v Peoni (1938) 100 F2d 401 (Second Circuit Court of Appeals) p 402)

The original MPC draft had included “knowingly” as a loose standard for accomplice liability, however the final draft follows Judge Learned Hand’s formulation, and requires
“purpose to facilitate the crime”. Most federal and state courts follow this standard, however as the other two excellent blogs in this discussion point out, interpretation of the standard can sometimes differ.

In Canada there is a definitional difference between an aider and an abettor; s. 21 of the Criminal Code requires for the aider that she “does or omits to do anything for the purpose of aiding any person to commit it” (emphasis added), whereas no such requirement exist for the abettor. Despite this, the same functionally unitary system of liability applies where there is no legal distinction between a principal and an assistant. As was stated in the leading Thatcher case: “[this provision] is designed to make the difference between aiding and abetting and personally committing an offence legally irrelevant. It provides that either mode of committing an offence is equally culpable, and, indeed, that whether a person personally commits, or only aids or abets, he is guilty of that offence.” (R v Thatcher S.C.J. No. 22, 1 SCR 652, p 690.)

Generally in Canadian case law there are high mens rea levels required for parties to a crime who do not actually commit the crime, such as knowledge or purpose, regardless of the mens rea requirement for the crime committed, as a way of placing some limits on the deliberate policy attempt to broaden the net of liability. The Supreme Court has stated that “the more peripheral the accused’s involvement to the completed crime, the more sense it makes to require a higher form of subjective mens rea.” (R v Roach (2010) 2 SCR 98 (Supreme Court of Canada) para 36.)

There must be a double intent; an intent to assist the physical perpetrator, as well as knowledge of at least the type of crime that is to be committed, though not necessarily the exact crime. The purpose requirement for the aider is interpreted with flexibility in the case law, since it “would be fulfilled if he had either intent or knowledge of both the crime, and of the intent of the perpetrators to commit the crime.” (R v Briscoe (2010) 1 SCR 411 (Supreme Court of Canada) para 16.) At the same time, willful blindness would also suffice in the place of actual knowledge.

Since abetting is generally understood to entail encouraging, instigating or promoting, it implies some intention to see the crime committed, and is interpreted more strictly.

It is difficult to compare the civil law tradition to these definitions, since the English language terms “aiding and abetting” can only be translated by approximation, and there are many different forms of assistant participation. While many jurisdictions that follow the civil law tradition have modes of liability that amount to instigation, soliciting or assisting, the definitions and mental fault elements differ. Suffice to say there is a double intent requirement in Argentina, Belgium, Germany, and the Netherlands, such that an actor must both intend her own participation and also intend the commission of the predicate crime (or a lesser crime. She is only liable for a further crime if it was foreseeable in the course of the crime she intended to participate in.) Whether this intent is the same as or stricter than “purpose” is uncertain for reasons of translation, but it would appear in case law trends that something similar to “knowledge” would usually suffice.

James Stewart has pointed out that in many of these jurisdictions dolus eventualis suffices whenever intent is required. Many common law lawyers are uncomfortable with this notion, but the best way to understand this is a lower limit of intent, the same as willful blindness. There must be evidence of the actor’s acceptance of the risk of a criminal outcome; this is what separates it from recklessness. Where the MPC has attempted to make clear distinctions between intent, purpose and knowledge, in most civil law jurisdictions there is only a clear distinction between intent and recklessness; the difference is that under intent there are many judicial interpretations which offer a sliding scale, the lower limit of which is dolus eventualis. Thus a term such as “purpose” may fit into this sliding scale without having a specific statutory definition. It would be something more than dolus eventualis and something less than pure intent, but as long as there is something willful (and therefore blameworthy) about the assistant’s actions, it would fit the generic test.

The point of the double intent requirement in these jurisdictions is the same as the reasoning that emerges in Canadian jurisprudence, namely that when extending liability to those who have not physically committed the crime, it is necessary to compensate the lesser physical contribution with a greater requirement of mental fault. However this has an even more important role in these systems where a normative legal differentiation exists between principals and assistants. A principal is considered to be more morally blameworthy, and therefore receives a greater punishment and the full weight of the conviction for having committed the crime. An assistant (including but not limited to aiders and abettors) has contributed less and is therefore less morally blameworthy, and receives a lesser legal qualification and usually a lesser punishment.

This distinction is embedded in the notion of moral agency; there is a difference between what I should do ‘simpliciter’, or the morality of principalship, and what I should do by way of contribution to what you do, or the morality of complicity. The latter is still wrong if you commit a crime, but it is a secondary wrong. For this reason a principal is convicted for the commission of the crime proper, whereas a secondary participant is convicted for her role in assisting the principal, but legally speaking she is not convicted for having committed the crime herself, since she did not fulfil the elements of the crime. She is not a genocidaire, but an assistant to the genocidaire, and is legally qualified as such.

In interpreting the wording of Article 25 of the Rome Statute, the terminology in paragraph (3)(b) seem to be drawn from civil law models (and then translated into English terminology, already risking translation problems!) and those in paragraph (c) are familiar to the common law model. Schabas suggests that since in practice the two paragraphs overlap very considerably, they should not be viewed as two different or distinct bases of liability, but rather as an effort to codify exhaustively.

This is why I believe the inclusion of “purpose” for the aider and abettor is justifiable; because he has contributed less, there must be some compensation for this in the form of requiring purposeful facilitation, in order to protect against guilt by association. Similarly, paragraph (3)(d) requires intentional contribution for an assistant to a group with a common purpose.

This aligns with what James Stewart states, that the “purpose” requirement goes to contribution, but the mens rea fault element to be determined still depends on the predicate crime. However I disagree with James that this necessarily amounts to a unitary theory of liability, for the reasons argued here. In fact it matches a differentiated theory particularly well. If the ICC continues its interpretation of Article 25 as differentiated, therefore requiring less physical contribution for the morally less blameworthy modes of liability, then it needs compensate this with high standards of volition with respect to the participation. This still leaves room for a dynamic theory for the mens rea with respect to the predicate crime. There must be safeguards against guilt by association for the assistant modes of liability, and the “purpose” requirement under (3)(c) as well as the “intentional” requirement under (3)(d) fulfill this need.

A comparative perspective on James’ final question “what does purpose mean” may still mean that there is some flexibility in interpreting the precise meaning of these safeguards. As long as guilt by association or recklessness are excluded, it may be possible to include knowledge, willful blindness or dolus evenutalis, especially since in civil law jurisdictions these all amount to gradations of intent. So as long as there is evidence of wilful participation, regardless of the motive, it would be possible to include business men and women who act for profit, but in full knowledge of the crimes they are facilitating.