UNIVERSITEIT VAN AMSTERDAM COLUMBIA LAW SCHOOL
AIDING AND ABETTING INTERNATIONAL CRIMES
IN THE SOCIAL MEDIA LANDSCAPE:
An inquiry into the individual criminal responsibility of
corporate social media actors
Master’s Thesis
LL.M International Criminal Law – Joint Program Amsterdam Law School and Columbia Law School Kaisa de Bel
kaisa_de_bel@hotmail.com Student number: 12534617
Supervisor: Prof. dr. mr. G.K. Sluiter Date: 30 June 2020
1
ABSTRACT
During the Rohingya crisis in Myanmar, the military allegedly made extensive use of Facebook to spread hate and incite violence, while, according to the United Nations, the same military actors committed acts that may amount to international crimes. In the context of volatile societies such as Myanmar, should the corporate officials of social media companies be deemed complicit in the commission of international crimes for their failure to effectively govern their platform? While the modes of liability of international criminal law are tailored to attributing responsibility to such remote actors, the aiding and abetting principle still requires some ‘causation’ link between the act of assistance and the perpetration of a crime. This thesis therefore examines whether the corporate executives of social media companies should meet the requisite degree of ‘causation’ under the aiding and abetting principle of international criminal law. Drawing inspiration from theories of attribution and leadership modes of liability under international and domestic law, the thesis argues that we should look at whether a corporate executive had control over the acts within the corporation, and whether by failing to intervene he or she was taking a socially unacceptable risk in violation of a corporate duty of care. As such, even though the mismanagement of a social media platform might facilitate the commission of international crimes, this should not be attributed to the corporate leaders of social media companies both because of their limited control over the governance of the platform, and more importantly, because of the lack of (international) consensus on the corporate duty of care in social media platform moderation. This thesis therefore calls for the development of a clear international legal framework for platform regulation, before corporate executives of any social media company should be held liable under international criminal law.
2
TABLE OF CONTENTS
I. Introduction ... 3
II. The potentially facilitating role of social media platforms in the commission of international crimes ... 6
2.1 Case example: The role of Facebook in the Myanmar Rohingya crisis ... 6
2.2 The ‘weaponization’ of social media platforms by nefarious users ... 10
2.3 The potential act of assistance of social media companies ... 10
III. The causation standard of the aiding and abetting principle ... 12
3.1 The principle of aiding and abetting ... 12
3.2 Causation standard at the ad hoc tribunals ... 13
3.3 Causation standard at the ICC ... 16
3.4 Interim conclusion: Causation as a case-by-case assessment ... 18
IV. Attributing the ‘neutral’ business activity to the remote corporate executive ... 19
4.1 Obstacles in assessing individual criminal liability of business actors ... 19
4.2 Theories of attribution ... 22
4.3 Leadership modes of liability ... 24
4.4 Having the ability and duty to intervene in corporate misconduct ... 26
4.5 Interim conclusion: the substantial contribution by a corporate actor ... 28
V. Corporate executives of social media companies: a substantial contribution? ... 29
5.1 Social media and issues of causal remoteness ... 29
5.2 The substantial act of social media platform mismanagement ... 30
5.3 Attributing the social media platform mismanagement to corporate executives ... 32
VI. Conclusion ... 35
3
I. Introduction
In 2018, the United Nations’ Fact-Finding Mission (FFM) on Myanmar reported that Myanmar’s military made extensive use of Facebook to incite violence against the country’s Rohingya Muslim minority, resulting in real-world discrimination and violence.1 Not only did
the FFM find reasonable grounds to conclude that war crimes, crimes against humanity and genocide have been committed, but also that the ‘role of social media is significant.’2 Facebook
specifically ‘has been a useful instrument for those seeking to spread hate, in a context where, for most users, Facebook is the Internet.’3 The detailed report demonstrates how Facebook and
other social media companies have the potential to facilitate the commission of international crimes, when they operate in the context of volatile societies.
With the opening of an official investigation into the situation in Myanmar by the International Criminal Court (ICC),4 the question arises as to whether social media companies
such as Facebook, or their executives, could be held accountable under the legal framework of international criminal law. As this legal framework is designed in such a way to respond to the complex nature of international crimes, perpetrators from afar – who plan, order and facilitate crimes – can be held criminally liable for the atrocities committed on the ground. Under this extended notion of criminal liability, it might be feasible to hold social media operators accountable for the mismanagement of their platforms.
There is a growing body of scholarly literature on the roles of business actors assisting in the commission of international crimes, and the criminal liability of corporate actors.5
Moreover, case law from both domestic courts and international tribunals show that it is possible to hold businessmen and other remote facilitators accountable for their facilitating role in international crimes.6 This includes cases involving media entities and their role in war
propaganda. In 1946, the International Military Tribunal (IMT) at Nuremberg found the
1 UN Human Rights Council (UNHRC), ‘Report of the Detailed Findings of the Independent International
Fact-Finding Mission on Myanmar’ (17 September 2018) A/HRC/39/CRP.2, paras 1342-1356.
2 UNHRC, ‘Report of the independent international fact-finding mission on Myanmar’ (12 September 2018)
A/HRC/39/64, para 74.
3 Ibid.
4 Situation in the People’s Republic of Bangladesh/Republic of the Union of Myanmar (Decision on the
Authorisation of an Investigation into the Situation in the People’s Republic of Bangladesh/Republic of the Union of Myanmar), ICC-01/19-27 (14 November 2019).
5 See, for example, the special issue of the JICJ, Florian Jessberger and Julia Geneuss (eds), ‘Transnational
Business and International Criminal Law’ (2010) 8 Journal of International Criminal Justice.
6 The most cited international precedents include the Nuremberg IMT convictions of businessmen, such as Trials
of War Criminals Before the Nuernberg Military Tribunals Under Control Council Law No. 10, Vols VI (Flick),
VII and VIII (IG Farben), XIV (Ministries), Nuremberg October 1946–April 1949 (Washington, DC: United States Government Printing Office, 1953). Domestic examples include the Dutch conviction of businessmen Van
Kouwenhoven: Court of Appeal of the Hague, 21 April 2017, ECLI:NL:GHSHE:2017:1760; and Van Anraat: Court of Appeal of The Hague, 9 May 2007, ECLI:NL:GHSGR:2007LBA4676.
4
publisher of an antisemitic newspaper, Julius Streicher, guilty of crimes against humanity in connection with his incitement towards persecution of the Jewish population in Europe.7 In
2003, the International Criminal Tribunal of Rwanda (ICTR) convicted the founders of the extremist radio station RTLM for incitement to genocide, because the radio station provided the platform for a propaganda campaign encouraging the Hutu population to kill Tutsis.8
Parallels can be drawn between the role of the media entities in these cases and Facebook in Myanmar. Like RTLM in Rwanda, Facebook was used as a platform for the spread of hateful messages, inciting people to commit real-world violence, and resulting in mass atrocities. However, as has been highlighted in the literature,9 the legal framework of incitement appears
to be a bad fit for the accountability of social media operators, because of a crucial difference with traditional media entities: social media companies do not generate hateful or inciteful content themselves. Instead, they allow their users to post such content on their platform.
At the same time, however, scholars are underlining that social media operators are not entirely passive actors either.10 Platforms such as Facebook utilize algorithms to personalize
newsfeeds, thereby accelerating the effect of some content. As such, they do not only allow harmful content on their platform, but also potentially provide greater exposure to and credibility of misinformation and propaganda advocating violence.11 It has been argued that social media operators should therefore not be held liable as a media entity under the standards of incitement, but rather as analogous to business actors facilitating international crimes by selling weapons to the principal perpetrator, and being held accountable under the aiding and abetting mode of liability.12
Grappling with questions of complicity in the context of technology is increasingly relevant, given its present-day dominance in our global society. A development coined by the World Economic Forum as the ‘Fourth Industrial Revolution’, new digital and data-driven technologies have altered the way we live our economic and social lives.13 The popularity of
7 Judgment of the Nuremberg International Military Tribunal (1947) 41 AJIL 172, 293–296 (Streicher case).
8 Prosecutor v Ferdinand Nahimana et al (Judgement and Sentence) ICTR-99-52-T (3 December 2003) (‘ICTR
Media case’).
9 Shannon Raj Singh, ‘Move Fast and Break Societies: The Weaponisation of Social Media and Options for
Accountability under International Criminal Law’ (2019) 8 Cambridge International Law Journal 331.
10 Tarleton Gillespie, ‘Platforms Are Not Intermediaries’ (2018) 2 Georgetown Law Technology Review 198. 11 Ingrid Burrington, ‘Could Facebook Be Tried for Human-Rights Abuses?’ (The Atlantic, 20 December 2017)
<https://www.theatlantic.com/technology/archive/2017/12/could-facebook-be-tried-for-war-crimes/548639/?utm_source=twb> accessed 19 June 2020.
12 Raj Singh (n 9).
13 Klaus Schwab, ‘The Fourth Industrial Revolution: What It Means and How to Respond’ (World Economic Forum, 14 January 2016) <https://www.weforum.org/agenda/2016/01/the-fourth-industrial-revolution-what-it-means-and-how-to-respond/> accessed 19 June 2020.
5
social media is part of this development, as it provides for greater connectivity and rapid exchange of information. It thereby plays a significant role in constituting the public debate. This also means that with one click of the mouse, remote actors can have a major influence on events on the other side of the world, increasing the risk that secondary actors play a role in atrocities committed in distant locations. Considering this context, it is important to find answers to questions as to how norms of international criminal law should respond to this development. How far should the net of criminal liability be thrown? When can we say that the link between the acts of distant actors and atrocities on the ground is sufficient for criminal liability? Having clear standards about such ‘causation’ specifically, and secondary liability of corporate actors generally, is important for two reasons. First, it enhances the effectiveness of criminal prosecutions, thereby improving the protection mechanisms for victims of mass atrocities. Second, it allows for corporate actors to adjust their conduct accordingly, contributing to the deterrent effect of international criminal law while also preventing an unnecessary chilling effect on legitimate actions in international commerce and cooperation.
For these reasons, this thesis will assess whether the corporate executives of social media companies should meet the degree of causation required under the aiding and abetting mode of liability of international criminal law. For an accomplice to be held liable as an aider and abettor, the requisite mental element (mens rea, or ‘guilty mind’) and material element (actus reus, or ‘criminal conduct’) need to be satisfied. This thesis will focus on the actus reus element, requiring the aider and abettor to make a ‘causal’ contribution to the commission of the crime. Moreover, since the statutes of the modern international criminal tribunals, including the ICC, do not constitute jurisdiction over abstract legal entities such as (social media) companies, but only provide for the liability of individuals, the focus will be on the ‘corporate’ criminal liability of individual actors operating in their corporate capacities for social media companies. In the context of social media platforms, it is especially interesting to consider to what extent the mismanagement of the platform should be attributed to individuals within the corporation.
The structure of the thesis will be as follows. Chapter II will identify how social media companies play a potentially facilitating role in the commission of international crimes, using Facebook’s role in Myanmar as a case study. In chapter III, the actus reus causation standard of the aiding and abetting principle will be discussed, by scrutinizing case law of both the ad hoc criminal tribunals and the ICC. Given the lack of case law on business actors, chapter IV will consider how this aiding and abetting liability should be attributed to business leaders acting in their corporate capacities. It will specifically focus on theories of attribution and
6
modes of liability tailored to leadership positions. It will be argued that, on the basis of these sources, it is important to look at whether a corporate executive had control over the criminal acts within the corporation, and whether by failing to intervene, he or she was taking a socially unacceptable risk in violation of a corporate duty of care. Chapter V will answer the research question, assessing whether the role of social media platforms should be considered significant enough to be able to say there was a sufficiently causal link with the commission of atrocities for aiding and abetting criminal liability, and should be attributed to the corporate executive of a social media company. The thesis will close with concluding observations, including some reflections on the desirability of holding social media companies accountable.
II. The potentially facilitating role of social media platforms in the commission of international crimes
2.1 Case example: The role of Facebook in the Myanmar Rohingya crisis
Myanmar has a long history of ethnic tension and conflict between its Buddhist and Muslim communities. On numerous occasions since the early 1990s, the UN has reported on widespread human rights violations linked to the Myanmar political regime and military.14 The Rohingya
Muslim minority is subject to systematic human rights abuse in particular. For decades, the discrimination and violence targeted against them have forced large numbers of Rohingya people to migrate to neighboring state Bangladesh.15 In 2012, tensions started to escalate again
with outbreaks of anti-Muslim violence in Rakhine state, where most Rohingya’s live. In August 2017, a humanitarian disaster enfolded when the Myanmar military, known as the Tatmadaw, conducted a number of ‘clearance operations’ aimed at the Rohingya community in response to attacks on a military base by the Arakan Rohingya Salvation Army.16 According to
reports by human rights organizations, Rohingya villages were destroyed with fire, civilians were indiscriminately killed and Rohingya women and girls raped and abused.17 The targeted
attacks resulted in a mass exodus of the Rohingya people, with more than 700,000 of them fleeing to Bangladesh.18
14 UNHRC (n 1), para 97.
15 OCHA, ‘Rohingya Refugee Crisis’ (OCHA, 21 September 2017)
<https://www.unocha.org/rohingya-refugee-crisis> accessed 19 June 2020.
16 UNHRC (n 1), paras 749-751.
17 See, for example, Amnesty International, ‘Myanmar: Crimes against Humanity Terrorize and Drive Rohingya
Out’ (Amnesty International, 18 October 2017) <https://www.amnesty.org/en/latest/news/2017/10/myanmar-new-evidence-of-systematic-campaign-to-terrorize-and-drive-rohingya-out/> accessed 19 June 2020.
7
In 2018, the UN independent Fact-Finding Mission (FFM) on Myanmar published a report on the escalating tensions in Myanmar since 2012, and the 2017 state-led attacks. It concluded that there are reasonable grounds to believe that the acts by the Tatmadaw amount to international crimes, including genocide, crimes against humanity and war crimes.19 The
report highlighted that the exodus of and violence targeted against Rohingya people happened in a context where anti-Rohingya hate speech was rife. It noted that a ‘carefully crafted hate campaign’ by key government players, portraying the Rohingya and other Muslims as an existential threat to Myanmar and Buddhism, had created a ‘conducive environment’ for hardening of repressive measures against the Rohingya and the subsequent state-led violence in Rakhine state.20 Social media, and Facebook in particular, was found to have been of
significant importance in this hate campaign.
There are elaborate journalistic reports of the Myanmar military utilizing Facebook as a powerful tool to spread hate and misinformation.21 The military allegedly put approximately
700 people on the operation of a sophisticated strategy: Military personnel created what appeared to be news or celebrity pages on Facebook, managed these fake pages to attract large numbers of followers, and then used these pages to spread misinformation and offensive speech, timed for peak viewership.22 They also created fake ‘trolling’ accounts which were used to help
disseminate dangerous content, incite arguments between commenters and harass critics.23 The
hateful posts were often aimed against Rohingya Muslims – calling them ‘dogs, maggots and rapists, suggesting they be fed to pigs, and urging they be shot or exterminated.’24
Although anti-Muslim campaigns are not new in Myanmar, and hateful rhetoric is widely prevalent on all sorts of media platforms (including print media, broadcasts and pamphlets), the FFM highlighted that social media has a distinctive role in enabling the dissemination of divisive rhetoric in Myanmar, because of the population’s relatively new exposure to the internet and the emergence of Facebook as their main source of information.25
When Myanmar was still under military rule prior to 2011, access to the internet was heavily
19 ibid para 1385 – 1517. 20 ibid 696.
21 Paul Mozur, ‘A Genocide Incited on Facebook, With Posts From Myanmar’s Military’ (The New York Times,
15 October 2018) <https://www.nytimes.com/2018/10/15/technology/myanmar-facebook-genocide.html> accessed 19 June 2020; Steve Stecklow, ‘Why Facebook Is Losing the War on Hate Speech in Myanmar’ (Reuters, 15 August 2018) <https://www.reuters.com/investigates/special-report/myanmar-facebook-hate/> accessed 19 June 2020.
22 Mozur (n 21). 23 ibid.
24 Stecklow (n 21).
8
restricted and only a small few could afford it.26 This changed when in 2013 the government
started to deregulate tele-communications, allowing for foreign commercial phone operators to enter the market and compete with the state-owned phone company. This led to an enormous decrease in SIM-card prices, resulting in a rapid increase in the population’s internet access. In 2012, only 1.1% of the population had access to the internet. By 2016, half of the population had phone subscriptions.27 While Facebook was already becoming increasingly popular in this
context, the company launched a ‘light’ text-only version of their app in the Myanmar market at the same time. This ‘Facebook Flex’ version enabled users to access Facebook without incurring any data charges.28 Due to the combination of the population’s relative unfamiliarity
with the internet and the cheap access to Facebook, it quickly became the main mode of communication and source of online news for the public, as well as a tool for the Myanmar government to make public statements. As such, Facebook can be considered ‘the internet’ itself in Myanmar.29
This dominance of Facebook started at around the same time when tensions were rising again in Myanmar. There are many allegations that Facebook contributed to the increased discrimination and violence, and as such facilitated the forced migration of the Rohingya people.30 The Report found indications of ‘peaks of online hate speech’ emerging around
outbreaks of violence.31 In 2014, for example, false rumors of a Buddhist woman having been
raped by two Muslim men spread on Facebook, which is believed to be the cause of riots in Mandalay, resulting in the killing of both a Buddhist and a Muslim man.32 In 2017, the military
spread rumors on Facebook messenger through their fake news and celebrity pages to both Muslim and Buddhist groups, claiming that an attack from the other side was imminent.33 The
prevalence of such misinformation may have contributed to a feeling of vulnerability and fear in society, legitimizing increased (violent) interference by the military.34
Facebook has been blamed for not effectively responding to the misuse of their platform by the Myanmar military. Multiple civil society activists claim they repeatedly warned Facebook that their platform was being abused to promote hatred and violence in Myanmar,
26 ibid 1343. 27 Stecklow (n 21). 28 UNHRC (n 1) para 1344. 29 ibid 1345. 30 ibid 1347. 31 ibid 1326.
32 Stecklow (n 21); UNHRC (n 1) para 1325. 33 Mozur (n 21).
9
and that their content regulation system was ineffective in addressing the problem.35 Many
discriminatory and derogatory posts remained online for months, despite them being in violation of Facebook’s community standards.36 Like most social media platforms, users must
agree to these community standards when they sign up, which also address hate speech and the company’s approach to fake accounts and misinformation. Content is to be removed by the platform’s moderators when it violates these policies. To that effect, harmful or unauthentic content can be either reported by the platform’s users, after which moderators review it for removal, or it can be flagged through technology automatically identifying certain types of hateful content.37
However, civil society activists have highlighted a number of problems with Facebook’s regulation system.38 First, the automated system that is supposed to flag problematic content
turned out to be incompatible with the language fonts of Burmese script, making it difficult to interpret and translate posts and thus undermining its ability to identify hate speech.39 As
Facebook’s own system did not effectively detect harmful content, the regulation system relied too much on third parties, such as users and volunteers from civil society, to report dangerous posts.40 However, most users in Myanmar were unable to report problematic posts, as the
population mostly does not speak English and the Facebook interface, including the option to report problematic posts, remained English for a long time.41 Moreover, once content was
flagged, it was often not effectively or swiftly removed by the moderators. Some posts flagged as dangerous would remain on the platform for months and would not be removed until they were reported and followed up upon numerous times.42 This problem was exacerbated by the
lack of Burmese-speaking content reviewers employed by Facebook, who understand the nuances of the language and the context of the comments. By 2015, there were 7.3 million active Facebook users in Myanmar, but only 4 Burmese speakers reviewing the content.43
Finally in 2018, Facebook acknowledged that it had been too slow in its response to the misuse of its platform in Myanmar.44 Since then, it has undertaken several efforts to remove
35 ibid; Stecklow (n 21). 36 Stecklow (n 21). 37 UNHRC (n 1) para 1349. 38 Stecklow (n 21). 39 ibid. 40 ibid. 41 ibid. 42 UNHRC (n 1) para 1351. 43 Stecklow (n 21).
44 Facebook, ‘Removing Myanmar Military Officials From Facebook’ (Facebook, 28 August 2018)
10
fake accounts and pages linked to military leaders, hire more Burmese speakers as content reviewers, introduce a Burmese language interface, and hiring a consultancy company to make a human rights impact assessment of Facebook in Myanmar.45 However, by the time these more
elaborate measures were taken, more than 700,000 Rohingya had already fled the country.
2.2 The ‘weaponization’ of social media platforms by nefarious users
The case study of Facebook in Myanmar shows how a poorly moderated social media platform can have a significant effect on human rights violations in conflict-ridden parts of the world, especially when it is the public’s main source of information. It appears that the Myanmar military used three tactics for their hate campaign: the creation of fake celebrity and news pages that attract a large number of followers, the creation of trolling accounts that help spread harmful content and provoke arguments, and the flooding of these pages and accounts with false and derogatory content.46
Similar disinformation tactics were used by Russia during the 2016 United States presidential election, and by President Duterte in the Philippines to silence political opponents and promote political support.47 And it does not stop there: Sri Lanka, India, Indonesia and
Ethiopia are all examples of volatile countries where Facebook and other social media platforms were misused, thereby fueling preexistent tensions.48 As such, it offers autocratic regimes a
powerful tool for spreading propaganda, hate and misinformation. News reporters and scholars characterize this abuse of the platforms as the ‘weaponization’ of social media.49 In Myanmar,
this practice created an environment conducive to offline and state-led violence and discrimination – the consequences of which turned out to be devastating.
2.3 The potential act of assistance of social media companies
As mentioned in the introduction, it is not the first time that media entities, such as newspapers and radio stations, have been linked to the commission of international crimes. However, a core
45 UNHRC (n 1) para 1352. 46 See section 2.1.
47 Natashya Gutierrez, ‘Is the Philippines in Step with Russian Online Propaganda Warfare?’ (Rappler, 10 March
2018) <http://www.rappler.com/newsbreak/in-depth/197558-philippines-russia-online-propaganda-warfare-techniques> accessed 19 June 2020.
48 Paul Barret, ‘Who Moderates the Social Media Giants? A Call to End Outsourcing’ (NYU Stern Center for Business and Human Rights, June 2020) <https://bhr.stern.nyu.edu/tech-content-moderation-june-2020> accessed 19 June 2020.
49 Lauren Etter, ‘How Duterte Turned Facebook Into a Weapon—With Help From Facebook’ (Bloomberg, 7
December 2017) <https://www.bloomberg.com/news/features/2017-12-07/how-rodrigo-duterte-turned-facebook-into-a-weapon-with-a-little-help-from-facebook> accessed 19 June 2020; Raj Singh (n 9).
11
difference between those traditional media entities and social media is that the former are the authors and editors of their own content, whereas the content on social media platforms is user-generated. For that reason, most social media platforms tend to present themselves as ‘neutral’ tech companies providing a service, rather than media entities who aim to provide news reflecting the ‘truth’.50
Yet, scholars are starting to point out that social media sites do not merely provide a neutral platform.51 They also play an active role in the moderation of content and curation of a
user’s newsfeed. Through their content regulation system, moderators make active decisions on how to interpret the community guidelines, thereby choosing to remove some posts and ban some users, but not others.52 In addition, social media companies design and apply algorithms
to provide a personalized newsfeed for each user, both with regard to content and advertisement.53 In fact, it is the very business model of social media companies to design a
newsfeed that promotes user engagement, thereby increasing revenue from advertisement and facilitating data collection.54 With the combination of an algorithm deciding what content
deserves priority by promoting it to the top of someone’s newsfeed, and a regulation system moderating some information away, social media sites play a big role in shaping the public debate. Unfortunately, the information that is often the most engaging, is sensational, aggressive or simply fake. As such, they potentially facilitate, amplify or exacerbate misinformation and derogatory content.55
In many ways, social media can be used to improve democracy and human rights, as it offers a unique platform for marginalized groups in society to connect, unify and voice their views and concerns to the whole world.56 However, when the service is weaponized by a
nefarious regime as described in the previous section, it runs the risk of fueling atrocity crimes. In the latter case, the potentially facilitating act of social media companies is threefold. First, they provide a platform that can be used as a propaganda tool to disseminate dangerous or false content. Second, they govern their platform by deciding what content is to remain online and
50 Etter (n 49). 51 Gillespie (n 10). 52 Burrington (n 11).
53 World Wide Web Foundation, ‘The Invisible Curation of Content. Facebook’s News Feed and Our
Information Diets’ (World Wide Web Foundation, April 2018) <https://webfoundation.org/research/the-invisible-curation-of-content-facebooks-news-feed-and-our-information-diets/> accessed 19 June 2020.
54 Gillespie (n 10). 55 ibid.
56 Jose Antonio Vargas, ‘Spring Awakening’ (The New York Times, 17 February 2012)
<https://www.nytimes.com/2012/02/19/books/review/how-an-egyptian-revolution-began-on-facebook.html> accessed 19 June 2020.
12
what is to be taken down. When the platform regulation is slow or ineffective, this could result in allowing dangerous content to be published. Third, the curation of content through algorithms means that there is an automated process deciding what people see in their newsfeed, with what frequency and with how much impact, thereby shaping the public debate in important ways. When speaking in terms of the ‘weaponization’ of social media, these algorithms can in a way be considered as turning the platform into an ‘automated’ weapon, selecting and targeting content without human intervention. Before being able to assess whether this should amount to an act of assistance that aids and abets the commission of atrocity crimes, the next chapter will first discuss the actus reus of this mode of liability, in particular its causation standard.
III. The causation standard of the aiding and abetting principle 3.1 The principle of aiding and abetting
International crimes paradigmatically concern acts committed in a collective and systematic manner. They typically involve multiple actors with varying degrees of criminal responsibility, depending on their individual level of participation. International criminal law sets out various modes of participation to capture the range of ways in which an individual may participate in the crime, and makes a distinction between modes of principal perpetration, and accomplice liability.57 Unlike the principal perpetrator, who is the mastermind behind the crime or controls
its commission, an accomplice can be held liable under the aiding and abetting mode of liability if he assists, encourages or morally supports the commission of an international crime.58
The aider and abettor can thus be set apart from the principal perpetrator by both the mens rea and actus reus requirement. With regard to the former, the aider and abettor does not need to share the perpetrator’s criminal intent. As set out by the Appeals Chamber of the International Criminal Tribunal for the former Yugoslavia (ICTY) in Tadić, mere knowledge that his or her acts will assist the commission of a crime by the principal is sufficient.59 With
regard to the actus reus, the principal can be distinguished from the aider and abettor by the level of ‘control’ over the commission of the crime. A principal perpetrator can frustrate the commission of the crime by not taking his part. For aiding and abetting, an act or omission60
57 Miles Jackson, ‘The Attribution of Responsibility and Modes of Liability in International Criminal Law’
(2016) 29 Leiden Journal of International Law 879.
58 See, for example, Prosecutor v Furundžija (Trial Judgement) ICTY-IT-95-17/1-T (10 December 1998)
(‘Furundžija Judgement’), para 235.
59 Prosecutor v Tadić (Appeals Judgement) ICTY-IT-94-1-A (15 July 1999) (‘Tadić Appeals Judgement’), para
229.
13
that assists, encourages or lends moral support to the crime is required, thereby making a causal contribution to the principal’s wrongdoing. As such, the accomplice need not have control over the crime, but his conduct must still have a sufficient link with the actual perpetration for criminal liability to arise.61
Aiding and abetting first appeared as an explicit mode of liability in Control Council Law No. 10 of the Nuremberg trials, and has since been applied by all international criminal tribunals to a wide variety of factual circumstances in which someone participated in the principal perpetrator’s criminal activity.62 Evidenced by its codification in the statutes of all
modern ad hoc criminal tribunal and the ICC,63 the doctrine of aiding and abetting is now firmly
established in international criminal law. Nevertheless, some controversies remain with regard to the more specific parameters of its elements. When it comes to the actus reus, the question of whether the act of assistance needs to have been ‘specifically directed’ at the commission of the crime was a contentious issue for many years.64 Second, the law remains somewhat unsettled
on the exact degree of participation that is required to say that it had a sufficient link with the commission of the crime. This latter requirement is relevant for the standard of causation, and will thus be subject to scrutiny in the rest of this chapter.
3.2 Causation standard at the ad hoc tribunals
The statutes of the ICTY, ICTR and the Special Court for Sierra Leone (SCSL) provide that ‘[a] person who […] aided and abetted in the planning, preparation or execution of a crime […] shall be individually responsible for the crime.65 This general provision has been specified by
the jurisprudence of these ad hoc tribunals. They have consistently held that an act of assistance needs to reach a certain causation threshold in order to meet the actus reus standard of aiding and abetting liability. More specifically, it must have a ‘substantial effect on the commission of the crime.’66
61 Furundžija Judgement, para 235.
62 See an elaborate historical overview in Manuel J Ventura, ‘Aiding and Abetting’ in Jérôme de Hemptinne,
Robert Roth, Elies van Sliedregt, Marjolein Cupido, Manuel J Ventura and Lachezar Yanev (eds), Modes of Liability in International Criminal Law (Cambridge University Press 2019).
63 Art. 7(1) ICTY Statute; Art. 6(1) ICTR Statute; Art. 25(3)(c) ICC Statute; Art. 6(1) SCSL Statute.
64 Although this has been a contentious issue for years, it seems to have been settled in favor of removing the
‘specifically directed’ requirement. See Ventura (n 62).
65 Art. 7(1) ICTY Statute; art. 6(1) ICTR Statute; art. 6(1) SCSL Statute.
66 See, for example, Furundžija Judgement para 235; Prosecutor v Blagojević and Jokić (Appeals Judgement)
ICTY-IT-02-60-A (9 May 2007) (‘Blagojevic Appeals Judgement’), para 127; Prosecutor v. Ntakirutimana and Ntakirutimana, (Appeals Judgement) ICTR-96-10-A and ICTR-96-17-A (13 December 2004) (‘Ntakirutimana Appeals Judgement’), para. 530.
14
This threshold was first posed by the ICTY Trial Chamber in Tadić, which rejected the prosecutor’s argument that even ‘the most marginal acts of assistance’ could qualify as aiding and abetting.67 Although the Tadić Trial Chamber initially held that the act of assistance must
have a substantial and direct effect on the commission of the crime,68 the ‘direct effect’
requirement was quickly dropped again by the Furundžija Trial Chamber because it was found to be too restrictive. The Furundžija Trial Chamber held that this requirement would be misleading because ‘it may imply that assistance needs to be tangible, or to have causal effect on the crime.’69 Instead, it found that customary international law merely requires a ‘substantial
effect’, which allows for non-tangible moral support and encouragement to qualify as a criminal act in addition to practical assistance.70 This interpretation was subsequently followed by the
Tadić Appeals Chamber, and has since been consistently applied by the ICTY, the ICTR and SCSL.71 The same holds for aiding and abetting by omission, for which it is required that ‘it
would have been substantially less likely’ that the crime be committed had the accused acted.72
Overall, the ad hoc tribunal’s jurisprudence shows a general consensus that the causation standard is one of ‘substantial effect’, sometimes used interchangeably with the term ‘substantial contribution’.73 Nevertheless, the tribunals have not adopted one common
definition of what conduct exactly rises to this level. The Tadić Trial Chamber acknowledged that there is no definition of ‘substantially’, but found on the basis of Nuremberg case law that this requirement calls for a contribution that ‘in fact has an effect on the commission of the crime’.74 Likewise, the Furundžija Appeals Chamber required that the contribution makes a
‘significant difference to the commission of the criminal act’.75 This means that even when the
act of assistance was ‘significantly removed’ from the ultimate criminal activity, it can still have a substantial effect if ‘the criminal act most probably would not have occurred in the same way had not someone acted in the role that the accused in fact assumed’.76 It follows that it is
not required that the crime would not have occurred at all without the act of assistance. As the
67 Prosecutor v Tadić (Opinion and Judgement), ICTY-IT-94-1-T (7 May 1997) (‘Tadić Judgement’), para 670. 68 Tadić Judgement, para 688.
69 Furundžija Judgement, para 232. 70 Ibid, para 234.
71 Tadić Appeals Judgement para 229; Ntakirutimana Appeal Judgement para 530; Prosecutor v Taylor (Appeals
Judgement) SCSL-03-01-A (26 September 2013) (‘Taylor Appeals Judgement’), para 368.
72 Prosecutor v Mrkšić & Šljivančanin (Appeal Judgement) ICTY-IT-95-13/1-A (5 May 2009) (‘Mrksic &
Sljivancanin Appeal Judgement’), paras 97 and 100. Note, however, that omissions may only lead to aiding and abetting liability when the accused had a legal duty and an ability to act.
73 Ibid, para 81.
74 Tadić Judgement para 688. 75 Furundžija Judgement para 233. 76 Tadić Judgement para 688.
15
term ‘substantial’ indicates – and in line with the ‘control’ distinction between principal and accomplice liability – the case law confirms that the threshold does not require the act of assistance to be a precondition, or a conditio sine qua non, to the commission of the crime.77
On a related matter, it also does not matter whether there were other persons also assisting in the commission of the crime, as long as the acts by the alleged aider and abettor were substantial enough by itself. In the case of aiding and abetting by omission, for example, the fact that there were other officers better placed to intervene does not negate the failure to intervene of the accused itself.78 In addition, the finding of a substantial contribution can be
based on a cumulative assessment of multiple acts by the accused.79 That is to say, it is not
required that each isolated act amounts to a substantial contribution to the commission of the crime. Moreover, whereas merely having a role in the system without influence is insufficient,80
it is not required to possess a certain level of authority or power to have the ability to make a substantial contribution. Nevertheless, someone’s level authority is a factor to take into account in the assessment.81
The causation standard of aiding and abetting was extensively discussed by the SCSL Appeals Chamber in Taylor. After reviewing various cases in which the assistance was found to have a substantial effect, it rejected the position of the Defense that in order for the assistance to have a substantial effect, it need to have been provided to the physical perpetrator who then uses it for the commission of a specific crime. Instead, the Taylor Appeals Chamber found that the ‘sustained operational support and military personnel Taylor provided to the RUF/AFRC’ had a substantial effect on their capacity to commit crimes,82 indicating that assistance provided
to a criminal organization or group can amount to a substantial contribution. Likewise, the ICTY Appeals Chamber clarified that the actus reus can be satisfied both when assisting a crime committed by an individual, or by a plurality of persons.83 Nevertheless, the mere
provision of means to a group or organization would not be sufficient in itself, and factors to consider in that regard are the ‘fungibility of the means’ and whether the group or organization committed isolated crimes or rather committed those crimes in furtherance of a widespread or systematic attack.84
77 Blagojević & Jokic Appeal Judgment para 134; Prosecutor v Kalimanzira (Appeal Judgement) ICTR-05-88-A
(20 October 2010) para 86.
78 Mrkšić & Šljivančanin Appeal Judgement para 200. 79 Blagojević & Jokić Appeal Judgment para 284. 80 Furundžija Judgement para 233.
81 Blagojević & Jokić Appeal Judgement, para 195. 82 Taylors Appeal Judgement para 368 and 521.
83 Prosecutor v Kvocka et al. (Appeals Judgement) ICTY-IT-98-30/1-A (28 February 2005), para 90. 84 Taylor Appeal Judgement para 391.
16
Although these factors provide some guidance for the ‘substantial effect’ assessment, the Taylor Appeals Chamber stressed that this causation standard recognizes the ‘variety of fact patterns which confront tiers of fact’, indicating that the determination of a ‘substantial effect’ requires a factual assessment on a case-by-case basis.85 It is thus not possible to provide an
all-encompassing definition on the substantial effect threshold. Instead, a case-by-case approach allows for flexibility and judicial discretion, which functions to ensure that ‘the culpable are properly held responsible for their acts’ while ‘the innocent are not unjustly held liable for the acts of others.’86 This is an important consideration in the context of business actors facilitating
international crimes, as their acts of assistance will usually be provided in a commercial setting with the potential of contributing to both lawful and unlawful activities.
3.3 Causation standard at the ICC
As opposed to the ad hoc tribunals, who have been consistent in requiring a certain causation threshold in order to meet the actus reus standard of aiding and abetting liability, ICC case law lacks such consistency. Art. 25(3)(c) of the Rome Statute provides that a person can be held criminally liable if that person ‘[f]or the purpose of facilitating the commission of […] a crime, aids, abets or otherwise assists in its commission or its attempted commission, including providing the means for its commission.’ This provision does not mention a causation threshold. At the same time, it introduces the new element of requiring the assistance to be ‘for the purpose of’ facilitating the commission of a crime. Therefore, it has been suggested in the literature that the Rome Statute requires a lower actus reus for aiding and abetting compared to the requirements under the Statutes and case law of the ad hoc tribunals, in exchange for a very high mens rea threshold.87
The ICC Bemba et al. Trial Chamber confirmed this approach by acknowledging that although a general causal link was necessary, the wording of the statute ‘does not suggest the existence of a minimum threshold.’88 So rather than maintaining the substantial effect threshold,
it was considered to be sufficient for an accomplice to merely contribute to the crime by furthering, advancing or facilitating its commission.89 The higher mens rea standard would, in
their view, justify the application of a lower actus reus standard than required by the ad hoc
85 Ibid para 368. 86 Ibid para 391.
87 Kai Ambos, ‘Article 25: Individual Criminal Responsibility’ in O Triffterer (ed), Commentary on the Rome Statute of the International Criminal Court (Nomos 1999) 384; William A Schabas, ‘Enforcing International Humanitarian Law: Catching the Accomplices’ (2001) 83 International Review of the Red Cross 439.
88 Prosecutor v Bemba et al. (Judgement) ICC-01/05-01/13 (19 October 2016), paras 90 – 93. 89 Ibid para 94.
17
tribunals.90 However, ICC case law is not consistent in this regard.91 While the Lubanga Trial
Chamber chose to follow the ‘substantial effect’ threshold as posed by the ad hoc tribunals,92
the need for a causation standard was rejected in its entirety by the Pre-Trial Chamber’s Ongwen Confirmation of Charges Decision.93 The latter approach implies that the assistance need not
have had any effect on the commission of the crime, not even a marginal one, as long as the assistance was for the purpose of facilitating the crime.
In the context of this ambiguity, the Bemba et al. Appeals Chamber had the opportunity to provide more clarity on the matter, with one of the issues on appeal being the degree to which the assistance should have an effect on the commission of the crimes.94 However, while the
Appeals Chamber did go into more detail on the mens rea element, it remained vague with regards to the causation standard. It merely noted that the ICC was not bound by the ad hoc tribunal’s jurisprudence, and that Article 25(3)(c) did not indicate whether the assistance need to have had an effect on the commission of the crime.95 Instead of elaborating on this issue to
clarify the law, it held that ‘whether a certain conduct amounts to ‘assistance in the commission of the crime’ […] even without the showing of such an effect can only be determined in light of the facts of each case.’96 It thereby held open the possibility of both casual and causeless
complicity.97
However, the majority of scholarly opinion rejects the complete dismissal of any causal link in the context of the ICC. First, scholars favor an interpretation that is consistent with the substantiality threshold, since there is no need to depart from this well-established standard at the ad hoc tribunals.98 According to Ambos, ‘[t]he lack of an explicit reference to [the
substantial effect or specific direction] requirements in the text of the provision cannot be interpreted as a conscious departure from this standard for Articles 7(1) ICTY and 6(1) ICTR Statute do not contain this reference either.’99 Second, Ventura argues that the dismissal of a
causation standard would be inconsistent with Article 17(1)(d) of the Rome Statute, which
90 Ibid para 95.
91 Goran Sluiter and Sean Shun Ming Yau, ‘Aiding and Abetting and Causation in the Commission of International
Crimes–The Cases of Dutch Businessmen Van Anraat and Kouwenhoven’ (2019) in Nina HB Jorgensen (ed) International Criminal Responsibility of War’s Funders and Profiteers (Cambridge University Press, Forthcoming).
92 Prosecutor v Lubanga (Judgment) ICC-01/04-01/06 (14 March 2012), para 997.
93 Prosecutor v Ongwen (Decision on the confirmation of charges) ICC-02/04-01/15 (23 March 2016) para. 43. 94 Prosecutor v Bemba et al. (Appeal Judgement) ICC-01/05-01/08-A (8 March 2018) (‘Bemba Appeal Judgment). 95 Ibid, paras 1326-1327.
96 Ibid, para 1327. 97 Sluiter and Yau (n 91).
98 Ambos (n 87) p. 1008; Schabas (n 87). 99 Ambos (n 87) p. 1008.
18
requires the Court to find a case inadmissible if it is not of ‘sufficient gravity’. This standard ‘would exclude aiding and abetting cases where the alleged contribution is miniscule or de minimis’.100 Third, since the substantiality threshold is subject to wide judicial discretion in its
application to the facts, there should be no concern that the substantial effect threshold is too high.101
3.4 Interim conclusion: Causation as a case-by-case assessment
After reviewing the relevant case law, it can generally be concluded that the ad hoc tribunals require a ‘substantial effect’ assessment for the aiding and abetting actus reus, while the ICC seems to adopt a somewhat more lax approach. Considering the Bemba et al. case law in light of the broader jurisprudence and scholarly literature, the appropriate approach for the ICC is probably to still require some causal link, but with a lower threshold than the one applied by the ad hoc tribunals. The emerging trend is that this is to be compensated for by a higher mens rea standard, but it remains unclear how these parameters will develop exactly in the future.
What the two lines of jurisprudence have in common, is that neither require a conditio sine qua non standard, and that they both indicate that the actus reus standard requires a case-by-case assessment, considering the factual situation as a whole. This latter consideration provides a welcome flexibility when it comes to applying the aiding and abetting principle to the conduct of business actors, as their assistance will often have a ‘dual-purpose’ quality, and thus have the potential of being used either lawfully or unlawfully.102 At the same time,
however, the case-by-case assessment brings along issues of legality, as the application of the standard in a given case remains uncertain.103 Such limited foreseeability and legal uncertainty
is not only problematic from the perspective of the defendant’s rights, but also undesirable if it has a chilling effect on socially desirable or legitimate commercial acts. The next chapter will therefore scrutinize how this should be given more substance when attributing aiding and abetting liability to business actors acting within the framework of a corporation.
100 Manuel J Ventura, 'The International Criminal Court’s Bemba et al. Case: The ICC Trial and Appeals Chamber
Consider Aiding and Abetting under Article 25(3)(c) of the Rome Statute' (1998) for the First Time' in Evelyn A. Ankumah (ed.), Essays in Honour of Judge Sanji Mmasenono Monageng (Forthcoming), pp. 9–11.
101 Ventura (n 100) p.14.
102 Sabine Michalowski, ‘Doing Business with a Bad Actor: How to Draw the Line Between Legitimate
Commercial Activities and Those That Trigger Corporate Complicity Liability’ (2015) 50 Texas International Law Journal 403.
19
IV. Attributing the ‘neutral’ business activity to the remote corporate executive The purpose of this thesis is to assess what the requisite degree of causation should be for the corporate executives of social media companies, under the aiding and abetting mode of liability of international criminal law. This question requires both an understanding of how to apply the standard of causation to business activity, but also how to attribute that to officials operating within the structure of a corporation. The previous chapter examined the standard of causation as applied by the ad hoc criminal tribunals and the ICC. This chapter will further examine how this can be attributed to corporate officials. To that effect, it will first elaborate on two obstacles in attributing individual criminal liability to corporate officials: the prima facie neutral nature of business activity, and the hierarchical remoteness of a corporate official to the commission of the crime (4.1). Second, it will examine theories of attribution and leadership modes of liability to provide a framework to address these issues (4.2 – 4.3). Finally, it will be argued that the responsibility for the substantial contribution made by a corporation as a whole, may be attributed to a corporate leader when he or she has both the ability to intervene in the corporate conduct by virtue of his or her leadership position, and has a legal duty to do so (4.4).
4.1 Obstacles in assessing individual criminal liability of business actors
In the past decade, there has been an increasing awareness of business involvement in gross human rights violations around the globe. The International Commission of Jurists has outlined the various ways in which corporate actors can become complicit in mass atrocities, including the provision of goods, services and information in the commission of crimes, the provision of personnel or logistical assistance to commit crimes, and the provision of banking facilities so that proceeds of the crime can be deposited.104 Most scholars identify aiding and abetting as the
most appropriate mode of liability in international criminal law to hold business actors liable.105
However, there are two main obstacles in assessing whether the commercial activity of a corporate actor rises to the level of a substantial contribution, as required for individual aiding and abetting criminal liability.
First, in the context of business involvement, the act of assistance will very often concern ordinary professional commercial activities, making only an indirect contribution to
104 International Commission of Jurists, 'Corporate Complicity & Legal Accountability, volume 2: Criminal law
and International Crimes' (ICJ, 2008) <https://www.icj.org/report-of-the-icj-expert-legal-panel-on-corporate-complicity-in-international-crimes/> accessed on 19 June 2020.
105 See, for example, Linde Bryk and Miriam Saage-Maaß, ‘Individual Criminal Liability for Arms Exports
under the ICC Statute, A Case Study of Arms Exports from Europe to Saudi-Led Coalition Members Used in the War in Yemen’ (2019) 17 Journal of International Criminal Justice 1117.
20
the crime.106 This difficulty relates to the provision of ‘neutral’ goods and services in particular,
such as the provision of financial resources. But even ‘grey zone’ activities that are more clearly dangerous, such as the provision of weapons, are not unlawful per se.107 Both the provision of
loans and arms trade can be perfectly legitimate, but also have the potential of becoming criminal when doing business with a ‘bad actor’, such as a criminal political regime.108 The
criminality of such acts can be limited either under subjective or objective theories, respectively raising the mens rea or actus reus threshold of aiding and abetting.109 In the Mbarushimana
Decision on the confirmation of the charges, the ICC Pre-Trial Chamber stressed the importance of an actus reus threshold in this regard, arguing that ‘many members of a community may provide a contribution to a criminal organization in knowledge of the groups criminality, especially where such criminality is public knowledge’.110 Without a threshold for the level of
assistance, it continued, ‘every landlord, every grocer, every utility provider, every secretary, every janitor or even every taxpayer who does anything’ could be considered as providing assistance to the crime, thereby risking that such acts give rise to individual criminal liability where it was clearly not intended by the drafters of the Statute.111 Thus, the more indirect or
remote the assistance gets, the more difficult it becomes to draw the line between morally condemnable but lawful business activity, and a criminally relevant contribution to someone else’s crime.
Second, the organizational complexity of a business actor can provide an obstacle to attributing individual criminal liability.112 The organizational structure of a business entity
ranges from an individual businessman acting by him- or herself, to the complex organizational composition of a transnational corporation. In the former case, it is clear who is to be held responsible. For example, in the Dutch Van Anraat case, businessman Van Anraat was found guilty of aiding and abetting war crimes committed by Saddam Hussein’s Iraqi regime, because
106 Christoph Burchard, ‘Ancillary and Neutral Business Contributions to ‘Corporate–Political Core Crime’,
Initial Enquiries Concerning the Rome Statute’ (2010) 8 Journal of International Criminal Justice 919; Carsten Stahn, ‘Liberals vs Romantics: Challenges of an Emerging Corporate International Criminal Law’ (2018) 50 Case Western Reserve Journal of International Law 91; Michalowski (n 102).
107 Bryk and Saage-Maaß (n 105). 108 Michalowski (n 102).
109 Hans Vest, ‘Business Leaders and the Modes of Individual Criminal Responsibility under International Law’
(2010) 8 Journal of International Criminal Justice 851.
110 Prosecutor v Mbarushimana (Decision on the Confirmation of the Charges) ICC-01/04-01/10 (16 December
2011) para 277.
111 Ibid. This argument was made in the context of art. 25(3)(d) RS liability on providing assistance to a criminal
organization – however, it is also relevant for aiding and abetting liability since the latter can also be applied to contributions made to the activities of a criminal organization or plurality of persons (see para 3.2; SCSL Taylor Appeal Judgement).
21
for some time he was the sole supplier of the thiodiglycol (TDG) chemical required for the production of mustard gas, which was used in attacks against the Kurdish population in Iraq.113
Van Anraat was an individual trader, and did not direct a multinational company with complex organizational structures.114 As such, the supply of TDG was directly traceable to him
personally. As the organizational structure of a corporation becomes more complex, however, it also becomes more difficult to identify the person within that corporate structure to attribute criminal responsibility to. It may very well be that the aggregate sum of acts by the ownership, top executives, lower management and employees of a corporation amount to a substantial contribution to a crime, but as international criminal law does not provide for criminal liability of corporations, the question is who should incur individual liability. This is not only complicated because of complex internal corporate structures and webs of relations, but also because those responsible for a transnational company’s involvement in a crime, are often located far away from the actual place of commission.115
The hardest cases will be those where the extremes of these two obstacles are combined: factual remoteness in the sense of the neutral and indirect character of business conduct, and organizational remoteness due to division of labor in a transnationally operating corporate entity.116 Unfortunately, there is very little case law that can guide us in this situation. Although
the ICTR has dealt with business leaders acting through corporate structures on multiple occasions, these cases did not concern typical business activities. In the Ntakirutimana case, the medical director of a hospital was prosecuted for aiding and abetting to the crime of extermination by providing a weapon to a principal, knowing that the weapon would be used in mass killings.117 In the Musema case, the director of a tea factory was prosecuted for
transporting and ordering employees of the factory to attack Tutsis, and for his own participation in the attacks. He was charged with aiding and abetting in genocide and extermination, and having superior responsibility in failing to prevent the crimes committed by his subordinate employees.118 Although he was held responsible for the acts of his employees
because of his strong position of control within the company, the acts he was prosecuted for had nothing to do with the business activity of the factory.119
113 Van Anraat case (n 6).
114 Marjolein Cupido, Mark Hornman and Wim Huisman, ‘Individual Liability for Business Involvement in
International Crimes’ (2017) 88 Revue Internationale de Droit Penal 223, p. 265.
115 Burchard (n 106). 116 ibid.
117 Ntakirutimana Appeal Judgement (n 66).
118 Prosecutor v Musema (Judgement) ICTR-96-13-A (16 November 2001). 119 Vest (n 109) p. 858.
22
Older case law by the Nuremberg IMT suggests that ordinary business activities cannot raise to the level of criminality. In the Ministries case, it was held that an ordinary businessman who does not go beyond the ordinary range of his duties, for example by only ‘financing’ core crimes, may be condemnable from a moral, but not from a legal point of view.120 However,
Burchard argues that this reasoning is outdated considering the evolution in international treaty regimes prescribing the criminalization of business activity that finances or furthers corruption, slavery and terrorism, for example.121 Likewise, Farrell contends that according to established
jurisprudence in international criminal law, a contribution ‘need not be criminal per se and could encompass what may otherwise be considered normal business transactions.’122
Moreover, acts of physically and hierarchically remote perpetrators can still be legally proximate enough to attribute criminal liability to them.123
Nevertheless, establishing that the business actor assisted in the crime of the principal perpetrator and that this conduct had a substantial effect on the commission of the crime, is difficult when it concerns corporate officials who operate from a physically and hierarchically remote position, and contribute only in an indirect or neutral way. In this context, it is challenging to find a balance between effective criminal liability for core crimes, and an excessive extension of criminal liability encroaching on legitimate business conduct. The remainder of this chapter will therefore further scrutinize if substantial acts of assistance should be attributed to high-ranking corporate officials, and what indicators to look out for.
4.2 Theories of attribution
From a theoretical perspective, the question of whether an act is substantial and should thus lead to criminal responsibility, is a normative one.124 Within the realm of international criminal
law, most scholars defend a risk-enhancing theory of attribution. According to Ambos, the aider and abettor must, with his or her contribution, create and increase the risk that the crime will be committed and thereby violate fundamental legal interests.125 The unique nature of international
crimes – namely causing extreme harm and often involving a high plurality of actors – argues in favor of such as risk-enhancing approach.126 The more serious the crime, the more it is
120 Ministries case (n 6), pp. 609-622. 121 Burchard (n 106) p. 931.
122 Norman Farrell, ‘Attributing Criminal Liability to Corporate ActorsSome Lessons from the International
Tribunals’ (2010) 8 Journal of International Criminal Justice 873, p. 893.
123 Farrell (n 122) p. 877. 124 Ambos (n 87) p. 1008. 125 ibid.
23
legitimate to penalize risk-based conduct, so the argument goes. Sluiter and Yau mention the crime of incitement to genocide as an example, which already criminalizes risk-based conduct as it does not require the incitement to actually result in genocide.127 Moreover, it is practically
impossible to prove factual causation for every accessory in the commission of international crimes, given their widespread and systematic nature.
At the same time, however, scholars agree that not every risk-enhancing act should be able to qualify as a criminal contribution. Sluiter and Yau caution that a risk-based causation standard should not make the aiding and abetting principle so extensive that it also covers the acts of those who are not in a position to make a difference or the risk is not very significant.128
In the same vein, legal scholars Vest and Ambos argue that only those risks that are realized through the commission of a crime and are disapproved of by a legal order, should qualify as such.129 Ambos stresses that this approach gains specific importance when identifying criminal
liability for neutral acts, such as business activity, so as not to criminalize legitimate or socially desirable conduct.130 Heyer provides the example of multinational corporations investing in conflict zones, who face the dilemma of either interacting with the conflicting parties, thereby risking the furthering of their potentially criminal activity, or abstaining from investing in such a zone, thereby missing out on profit while also depriving the local population of (re)gaining prosperity and thereby peace.131 She agrees that corporate agents should only incur criminal
liability when they take a socially unacceptable risk in that context, and supports that argument by showing that domestic criminal law systems are also based on this principle.
As such, only the acts of corporate actors that clearly increase the prohibited risk of the commission of a crime should be considered substantial enough to make it a criminal contribution. Yet, the notion of socially acceptable risk versus forbidden risk is still rather vague. What is a socially acceptable risk for a corporate actor to take, and when does it cross the border to criminal risk-taking? As both the ad hoc tribunals and the ICC phrase the aiding and abetting mode of liability in quite abstract terms, this leaves room to tailor it in a way unique for the position of corporate officials leading a large corporation.132 For that reason, the
following section will draw inspiration from the standards and requirements of leadership
127 ibid. 128 ibid.
129 Ambos (n 87); Vest (n 109). 130 Ambos (n 87) p. 1009.
131 Antje KD Heyer, ‘Corporate Complicity under International Criminal Law: A Case for Applying the Rome
Statute to Business Behaviour’ (2012) 6 Human Rights & International Legal Discourse 14, p. 39.
