AustLII Home | Databases | WorldLII | Search | Feedback

Monash University Law Review

Monash Law School
You are here:  AustLII >> Databases >> Monash University Law Review >> 2010 >> [2010] MonashULawRw 26

Database Search | Name Search | Recent Articles | Noteup | LawCite | Author Info | Download | Help

Mares, Radu --- "A Gap in the Corporate Responsibility to Respect Human Rights" [2010] MonashULawRw 26; (2010) 36(3) Monash University Law Review 33


A GAP IN THE CORPORATE RESPONSIBILITY TO RESPECT HUMAN RIGHTS

RADU MARES*

In 2008, Professor John Ruggie, the UN Special Representative for business and human rights, presented his ‘Protect, Respect, Remedy’ Framework to the Human Rights Council. It was well received and his mandate was renewed for another three years. Corporate responsibility is one of the Framework’s three pillars. When applied to a multinational enterprise, questions inevitably arise about what responsibilities the core company has when its affiliates infringe human rights. The principle of ‘do no harm’ on which the responsibility to respect is based adequately covers situations where the core company’s own decisions create negative ripple impacts throughout affiliate operations. However, the same cannot be said about instances in which affiliates infringe rights in the absence of a core company’s own harmful decision. Then the core company is merely associated with abusive affiliates. Even in such a situation, Ruggie rightly proposes that the core company should act with due diligence to prevent and remedy abuses. But what is the foundation upon which the core company’s responsibility to act is based? A close reading of Ruggie’s argument shows that this responsibility rests problematically on an emerging social norm, soft law and notions of non-legal complicity. This article explains why this foundation is inadequate. As a result, the very existence of the responsibility becomes questionable, its legitimacy debatable and the sound due diligence steps that Ruggie proposes less consequential in practice. To address this weakness, this article appeals to negligence jurisprudence to establish a more carefully grounded responsibility to act applicable to core companies. The analysis concludes with implications for the legal institutionalisation of the responsibility to respect in the wake of the SRSG mandate.

I INTRODUCTION

The creation of the mandate of the UN Secretary General’s Special Representative (‘SRSG’) on human rights and business in 2005, and the appointment of Professor John Ruggie to the job, have made a contribution to the corporate social responsibility (‘CSR’) debate that cannot be underestimated.[1] Ruggie’s ‘Protect, Respect, Remedy’ Framework, presented in 2008, was well received by the UN Human Rights Council and his mandate was renewed for another three years.[2] During this period, the three pillars of the Framework are to be further operationalised. Since 2005, Ruggie and his small team have taken on the very demanding task of clarifying a multitude of outstanding issues surrounding corporate responsibilities.[3]

Ruggie deems it crucial to articulate a minimum, clear baseline applicable to all types of companies ranging from large multinationals to small enterprises. The corporate responsibility to respect (‘RtR’) should be self-standing and independent from government’s own failures. RtR is a reflection of the ‘do no harm’ principle, which boils down to a responsibility to take due diligence steps to prevent or remedy harm.[4] At the same time, the special situation of corporate groups and networks, including large multinational enterprises (‘MNEs’) cannot be overlooked. The RtR, as applied to such large enterprises, has to answer questions about a core company’s[5] responsibility for the conduct of its own partners.[6] This raises practical questions that require serious thought.

It is surprising, if not paradoxical, that the RtR is presented uniformly, without further differentiating any types of businesses, be they small companies or controlling companies at the top of multinational groups or networks. It is also surprising that the promoters of this responsibility do not show more sensitivity to the difficulties raised by dealing with the culpable omissions of parent companies. The emblematic situations that put human rights on the international business agenda in the mid 1990s were all about companies being held responsible for the wrongful conduct of their affiliates: Nike was a buyer company that was blamed for the sweatshop practices employed by its suppliers overseas,[7] and Shell was a large subsidiary blamed for not intervening with the Nigerian government to stop executions following an unfair trial.[8] Ruggie himself identified Shell and Nike as ‘early cases [that] have acquired iconic status’.[9]

This article is concerned with the foundation on which the RtR rests in the specific case of large business enterprises, and particularly where core firms wield significant influence over their business group or network. If MNEs have to respect human rights, as Ruggie proposes, what special responsibilities does that entail for the core company, as distinguished from its affiliates (which of course are themselves expected to respect human rights)? What acts and omissions of the core company might be deemed culpable and why? In a legal context, Richard Meeran, the solicitor for the plaintiffs in groundbreaking transnational litigation that took place in the UK,[10] wrote about the special place and responsibility of the core company: ‘The key issue raised is whether an MNC parent company owes a legal duty of care to those injured by its overseas operations’.[11]

Ruggie states that companies must take due diligence steps ‘to become aware of, prevent and address adverse human rights impacts’.[12] There are two key stages of responsibility contained in this formulation: first, a responsibility to become aware and knowledgeable about a business partner’s (mis)conduct, through for example undertaking human rights impact assessments (‘HRIAs’) or monitoring and auditing; and second, a responsibility to act on such information and do something in terms of prevention and remediation. This article maintains that Ruggie allocates due weight to these twin responsibilities and attempts to advance a policy-making process that accounts for the roles of leading companies, various market actors, civil society groups, and governments.[13] As a contribution to this policy-making process, the analysis herein focuses on the SRSG mandate and Ruggie’s work so far. The aim is to critically assess the coverage of the RtR, the foundational principles on which the legitimacy of RtR depends, its content (due diligence steps a company should take), and the necessary limits placed on corporate due diligence efforts. The article presents the way in which Ruggie has reasoned about these four dimensions of the RtR, identifies strengths and weaknesses, and suggests a way forward for addressing weaknesses.

Part II below gives an overview of the SRSG reports to present his conceptualisation of the RtR. Part III identifies weaknesses in Ruggie’s argument. Part IV seeks solutions in jurisprudence, particularly in the law of negligence, a body of law that the SRSG team has only taken cursory note of and which could be of great help to Ruggie’s efforts to account for third party’s abuses of human rights.

II RUGGIE’S CORPORATE RESPONSIBILITY TO RESPECT

‘To respect rights essentially means not to infringe on the rights of others — put simply, to do no harm’, Ruggie wrote.[14] The challenge for core companies then ‘is to ensure they are not complicit in violations’[15] committed by their affiliates. In an attempt to understand properly and reflect faithfully the thinking of the SRSG team, I have split my analysis into a few steps: What are the sources — the foundations — of corporate responsibilities? Who is the subject of responsibility? What corporate conduct does Ruggie’s corporate responsibility cover? What are the due diligence steps a company should take to fulfil the RtR? What is the scheme of attribution, the concept used to attribute responsibility to the core company for abuses in affiliates’ activities? The search for answers in the SRSG reports reveals at times silences and ambiguities on some key aspects. My contention is that these aspects are neither irrelevant nor should they be overlooked given the danger that part of the RtR will atrophy and lose significance in practice.

A Foundations of Responsibility — Sources of Legitimacy

Ruggie considers not infringing human rights as a minimum responsibility, ‘the baseline expectation for all companies in all situations’.[16] For support, he draws on soft law instruments and identifies an emerging social norm that has been reinforced from various quarters:

Social norms may vary by region and industry. But one of them has acquired near-universal recognition by all stakeholders, namely the corporate responsibility to respect human rights, or, put simply, to not infringe on the rights of others. By near-universal is meant two things. First, the corporate responsibility to respect is acknowledged by virtually every company and industry CSR initiative, endorsed by the world’s largest business associations, affirmed in the Global Compact and its worldwide national networks, and enshrined in such soft law instruments as the ILO Tripartite Declaration and the OECD Guidelines. Second, violations of this social norm are routinely brought to public attention globally through mobilized local communities, networks of civil society, the media including blogs, complaints procedures such as the OECD NCPs, and if they involve alleged violations of the law, then possibly through the courts. This transnational normative regime reaches not only Western multinationals, which have long experienced its effects, but also emerging economy companies operating abroad, and even large national firms.[17]

In addition to soft law and social norms, Ruggie uses the concept of complicity (discussed in Parts II(E) and III(C) below), which plays a role in CSR frameworks to establish the core company’s responsibility for its affiliates’ activities. Part III below raises concerns about the solidity of each of these three building blocks: soft law, societal norm and complicity.

B Subject of RtR

Although Ruggie addresses the RtR to all companies that his mandate covers — ‘transnational corporations and other business enterprises’[18] — he takes note that MNEs are ‘the most visible manifestation of globalization today’ and there are ‘some 70 000 transnational firms, together with roughly 700 000 subsidiaries and millions of suppliers spanning every corner of the globe’.[19] Ruggie refers to business groups as ‘extended enterprises’ and his definition of transnational business is broad enough to cover both equity-based as well as network-based corporate groups.[20] He recognises business networks as a distinct, important business entity and appears critical of how liability is (not) legally shared upwards in ‘extended enterprises’: it is ‘exceedingly difficult to hold the extended enterprise accountable for human rights harm’.[21] So the problem that business groups and networks raise is clearly acknowledged. Furthermore, Ruggie commissioned studies showing that a significant number of CSR cases reveal ‘indirect forms of company involvement’; that is, they involve some responsibility of the core company for abuses committed by third parties.[22]

Talking of ‘corporate’ responsibilities or about ‘businesses’ having to inflict no harm does not make it immediately obvious who is the exact subject of responsibility. Does the RtR apply to business groups as a whole or merely to each and every entity therein considered separately? In his reports, Ruggie has not felt the need so far to clarify this for RtR purposes. From personal communication, it becomes clear that Ruggie addresses the RtR to the entire business group or network deemed a single economic entity.[23] Furthermore, a recent discussion paper clarifies that a core company has a responsibility to exercise its leverage over its affiliates (see Part II(C) below).

The RtR, uncontroversial as it appears at first sight, can be a sensitive topic when it is applied to large, complex business groups.[24] Could it be that addressing the RtR to a large business group or network inherently entails a special responsibility to protect falling upon the core company placed at the top of that group or network? If so, such a responsibility requires a careful analysis. But first, one has to examine how Ruggie distinguished between different types of corporate conduct.

C Types of Corporate Conduct:
Indirect Impacts and Leverage

Whether in the form of commissions or omissions, corporate conduct might conceivably entail responsibility due to the influence that core companies have over third parties. But more clarity is needed. In his writings on the concept of ‘sphere of influence’, Ruggie distinguished between influence as ‘impact’ resulting from a core company’s own decisions and influence as ‘leverage’ that a parent might have over third parties.

On influence as ‘impact’, Ruggie is mindful that a core company’s decisions may have ripple effects throughout the business group or network. He clarifies that not only direct impacts are relevant but a duly diligent core company should account even for its more remote, indirect impacts. Where the company’s activities or relationships contribute to harm, ‘impact falls squarely within the responsibility to respect’.[25] That would cover the buyer company who sets its contractual terms in a way that suppliers are bound to take corner-cutting measures, including labour rights violations, to comply. Ruggie referred to these as ‘brand-induced problems’, such as flexible production, fast turnaround, surge orders, changed orders and so on.[26] Ruggie’s RtR accounts for actual and potential impacts, close or remote impacts, and thus makes a significant extension to a core company’s legal liability. As he wrote:

If the scope of due diligence were defined by control and causation this could imply, for example, that companies were not required to consider the human rights impacts of suppliers they do not legally control, or situations where their own actions might not directly cause harm but indirectly contribute to abuse.[27]

When it comes to influence as ‘leverage’, in situations where companies have influence but omit to act to prevent or redress human rights abuses committed by third parties, responsibility does not follow:

companies cannot be held responsible for the human rights impacts of every entity over which they may have some influence … Asking companies to support human rights voluntarily where they have influence is one thing; but attributing responsibility to them on that basis alone is quite another.[28]

In philosophical terms, this would mean ‘can implies ought’, which Ruggie argues is not acceptable. It would appear that he is not prepared to accept such a corporate responsibility to use leverage and core companies could rest content if they, through their own decisions, did not cause even indirectly harm. That would be mistaken.

A discussion paper Ruggie released in 2010 clearly states a responsibility to exercise leverage:

Where human rights abuses in the supply chain are identified, the enterprise should assess:
(a) whether the enterprise is implicated in the abuse solely by the link to the goods or services it procures (e.g., without contribution from the enterprise, the product is produced by bonded or child labor; or where an enterprise’s external security provider commits human rights violations in protecting company facilities); [and]
(b) whether the enterprise is also contributing to the abuse by its own actions and omissions (e.g., where the buyer demands significant last-minute changes in product specifications without adjusting price or delivery dates, leading to labor standard violations by a supplier in a low-margin business).[29]

Is there an inconsistency, or maybe a change of position from previous reports? It should be first observed that a responsibility to exercise leverage can arise in two very different situations: abuses committed in connection with affiliate business operations and abuses lacking any such connection whatsoever. Simply put, abuses arising in the operation of the business enterprise as a whole are to be differentiated from abuses happening totally unrelated to such business activities. When Ruggie wrote that the basic social expectation is for business to do no harm, he eliminated the latter responsibility to exercise leverage but meant to keep within the RtR the former responsibility because it concerns abuses linked to business activities. Thus it is a responsibility to ‘respect’ rights being infringed by the business groups as whole. In short, affiliates linked to a core company’s enterprise would be covered but not state actors or other private actors infringing human rights in other parts of the country with no connection whatsoever to the enterprise’s operations. This is in line with Ruggie’s addressing the RtR to the business enterprise as a whole, as described in Part II(B) above.

To my knowledge, Ruggie had not written explicitly to allow this distinction until 2010, however he maintains that this has been his understanding all along.[30] Clearly suppliers’ operations have been covered, but that could have referred only to the direct and indirect impacts of buyers’ decisions, not to buyers’ potential responsibility to exercise leverage. In fairness, all examples he provided in the 2008 Report to explain his rejection of a responsibility to exercise leverage refer to non-business related abuses only. For instance, Ruggie wrote it would be unreasonable to hold a company ‘responsible for the human rights impacts of every entity over which they may have some influence’ and even not ‘desirable to have companies act whenever they have influence, particularly over governments’.[31] Elsewhere, Ruggie commented on corporate silence, that is, the omission to take a stance against abuses. He clarifies that a rather often discussed situation — business operations in a country with systematic, large-scale human rights violations — is likely to fall outside the responsibility to respect:

In the business and human rights context the question often arises whether a company’s mere presence in a country where human rights violations are occurring can amount to complicity. Legal liability for complicity when a company is merely present is unlikely. Analogizing from international criminal law cases, it would have to be shown that the company’s silence amounted to a substantial contribution to the crime, such as legitimizing or encouraging the crime, and that the company provided such encouragement knowingly.[32]

At a more general level, Ruggie questioned the wisdom of imposing overly extensive positive responsibilities such as that to exercise leverage:

Yes, corporations are organs of society, but they are specialized organs, not microcosms of the social whole. Therefore, apart from acts that constitute international crimes or complicity in crimes — and these matters themselves are not fully settled — the character and limits of corporate obligations ought to reflect their social role, especially when it comes to positive obligations.[33]

To conclude, the SRSG writings on leverage should not be interpreted as denying the responsibility of core companies to exercise leverage over their business partners. Furthermore, his strong rejection of ‘leverage’ needs to placed in the context of his mandate, which was created in the wake of the demise of the UN Norms. From the earliest moments, Ruggie perceived the need to break with the legacy of the Norms and to state unequivocally that corporate responsibilities have to do with the harm that business activity itself has inflicted.[34] Indeed, the Norms formulated extensive corporate responsibilities which could arguably have covered third party abuse irrespective of whether such parties were related to businesses operations or not.[35]

D Due Diligence

A company, according to Ruggie, has ‘to become aware of, prevent and address adverse human rights impacts’[36] through four key steps: adopt a human rights policy; conduct human rights impact assessments; integrate human rights policies throughout the company; and track performance.[37] Elsewhere Ruggie wrote that due diligence (‘DD’) is a ‘comprehensive, proactive attempt to uncover human rights risks, actual and potential, over the entire life cycle of a project or business activity, with the aim of avoiding and mitigating those risks’.[38] At the core of the RtR are thus two discrete responsibilities. One has to do with acquiring knowledge about human rights impacts by collecting information through, for example, human rights impact assessments, monitoring and auditing.[39] The other refers to actually doing something about (the risk of) human rights abuses thus uncovered by setting up appropriate corporate policies, procedures and structures to prevent/redress third party abuse.[40] In a nutshell, the RtR has two sub-responsibilities: to collect information and form knowledge; and to act on that knowledge.

How does this play in the situation of core companies? As outlined in the previous section, there are two scenarios where the RtR is applicable to the core company. On the one hand, where the company’s own business decisions directly or indirectly cause abuses, the diligent company should first acquire knowledge about how its own decisions pose risks to human rights through rippling effects and about what abuses take place in affiliates’ operations, and second, change its decisions to prevent such rippling effects. On the other hand, when abuses occur independently of the core company’s own decisions, the diligent company should first form knowledge about abuses taking place in its affiliate operations, and second, exercise leverage over the affiliate. The details of these sub-responsibilities, including foundations and limiting concepts, are spelled out in detail in Part IV(B) below.

It is clear now that in Ruggie’s thinking, the core company is linked to abuse through its own action (business decision with rippling effects) and even inaction (mere relation with affiliate). Two sub-responsibilities attach. The question is whether the SRSG Reports employ concepts that can really support the weight of these two sub-responsibilities. The SRSG reports, together with much of the CSR literature,[41] uses complicity as the suitable notion to capture responsibility for third party abuse. Can this concept really deliver?

E Complicity

The SRSG reports employ complicity as a useful notion to cover abuses occurring in partners’ operations: ‘Complicity refers to indirect involvement by companies in human rights abuses — where the actual harm is committed by another party, including governments and non-State actors’.[42] After examining how complicity is treated under international criminal law and the US Aliens Tort Claims Act,[43] Ruggie concludes that, in law, an accomplice (aider and abettor) is one who knowingly provides practical assistance or encouragement that has a substantial effect on the commission of a crime.[44] He chose to begin his incursion into complicity reasoning at the hard legal core provided by international criminal law. However he went further to highlight complicity’s relevance in the CSR context: ‘The concept has legal and non-legal pedigrees, and the implications of both are important for companies’.[45] This is similar to the UN Global Compact, which uses complicity in both its legal and non-legal meanings.[46] The International Commission of Jurists (‘ICJ’) reflected on the legal and policy meanings of complicity:

For a number of years now the word ‘complicity’ has been used on a daily basis in policy documents, newspaper articles and campaigning slogans, to describe the different ways in which one actor becomes involved in an undesirable manner in something that someone else is doing. Frequently, the term is not used in the legal sense denoting the position of the criminal accomplice, but rather in a rich and multi-layered colloquial manner to convey the connotation that someone has become caught up and implicated in something that is negative and unacceptable. Such use of the term has become commonplace in the context of work on business and human rights, and it has provided a tool to capture and explain in simple terms the fact that companies can become involved in human rights abuses in a manner that incurs responsibility and blame.[47]

One of the leading pieces on complicity in the CSR area is Clapham and Jerbi’s three-part categorisation of complicity: direct, beneficial and silent.[48] Their contribution proved influential as it tried to look beyond the strictly legal concept applied in courts of law to layman uses of complicity employed in the ‘court of public opinion’. The UN Global Compact[49] and the International Organization for Standardization (‘ISO’) 26000 Guidance on social responsibility[50] refer to it, and even Ruggie mentioned it too.[51] In a nod to beneficial complicity notions, in his 2008 Report, Ruggie touches upon ‘benefiting’ from harm created by the ‘extended enterprise’. He writes that ‘deriving a benefit from a human rights abuse is not likely on its own to bring legal liability. Nevertheless, benefiting from abuses may carry negative implications for companies in the public perception’.[52] This cautionary tone is also evident in other reports: ‘benefiting is a relevant consideration in non-legal contexts’.[53]

What is the precise role played by ‘complicity’ in Ruggie’s Framework? The question is not easy to answer. On the one hand, Ruggie uses it to account for a parent’s ‘indirect involvement’ in its affiliate’s human rights abuses: ‘avoiding complicity is viewed as an essential ingredient in the due diligence carried out to respect rights because it describes a subset of the indirect ways in which companies can have an adverse effect on rights through their relationships’.[54] On the other hand, complicity is amenable to many uses. It is clearly a label, a convenient shorthand, to describe and cover diverse situations of a company being associated with harm occurring in affiliate operations. Complicity is also a legal concept that could provide practical remedies in courts of law, as explained in Ruggie’s Pillar Three. More than this, it is a charged term that can affix to companies lasting stigma in courts of public opinion, which may provide non-legal remedies. What becomes clearer,[55] is that Ruggie does not rely on complicity to elaborate a scheme of attributing responsibility to core companies, a scheme that would aim to establish the two sub-responsibilities to gather information and to act on it. A scheme of attribution would state and explain the foundation and the limiting concepts on a core company’s responsibility for affiliate wrongdoings.

So Ruggie does not aim to propose such a scheme of attribution and has not elaborated on the foundational issues of the RtR for core companies more than cursorily drawing on soft law and social norms as the basis of RtR for all types of companies. Therefore he has not used complicity for such purposes. This however does not prevent investigation on a ‘what if’ basis: could the concept of complicity be useful to create a legitimate basis for the two sub-responsibilities? The question is whether complicity could play a more foundational role than Ruggie assigns to it and that numerous CSR writers seem to believe: when the legal definition is relaxed through the legal-non-legal ‘manoeuvre’, can the internal logic of complicity still work? Part III(C) below is dedicated to this discussion.

III SOME CONCERNS ABOUT THE FOUNDATIONS OF RTR

Although Ruggie accounts for different types of business in his examples, he uniformly talks of ‘companies’.[56] Ruggie does not deem it necessary to differentiate between companies. This might be a valid choice for the purposes of defining the content of the RtR, that is, what DD steps should be taken. But the choice backfires and the pursuit of uniformity is harder to defend when the purpose is to investigate the principled foundation of the RtR. Indeed the core company is different from other companies because it is placed at the centre of business groups and networks over which it wields influence. From this realisation come two responsibilities very different in nature: one is to account for a company’s own business decisions that have rippling effects through affiliate operations;[57] the second responsibility potentially appears in the situation when no such own business decision exists but the core company is still associated with affiliates who infringe rights.[58] This is the fault line that Ruggie himself drew very clearly by distinguishing influence as impact from influence as leverage.[59]

But somehow Ruggie does not take this realisation to its logical conclusion in the case of core companies. It becomes immediately clear that the first responsibility — to account for own impacts — is an expression of the ‘do no harm’ principle, which Ruggie uses to explain the RtR.[60] Similarly clear is that, in the absence of its own decisions or other type of fault, the core company is asked to apply the ‘reach out and help’ principle. These two principles are, needless to explain, totally different in nature. A responsibility based on ‘reach out and help’ needs to be justified properly and circumscribed carefully to guard against charges of illegitimacy and open-endedness. This principled discussion is absent for the time being from the SRSG reports. It renders questionable his DD prescriptions in as much as they are targeted at core companies and refer to situations where none of the core company’s own business decisions have direct or indirect impacts.

While Ruggie did not discuss specifically the small part of the RtR analytically isolated above, he did nevertheless build a foundation for his RtR and DD proposals on soft law, societal norm and possibly complicity. The three subsections below examine whether each of these three building blocks lay a solid foundation under the questionable part of the RtR identified above.

A Soft Law

Ruggie relies on soft law to amass legitimacy for the RtR, including the responsibilities of core companies in MNEs.[61] Indeed such instruments explicitly address MNEs and ask them to respect human rights. Thus the OECD Guidelines ask companies clearly to ‘respect the human rights of those affected by their activities’.[62] Then the ILO Tripartite Declaration also states that companies ‘should respect the Universal Declaration of Human Rights’ and other international instruments mentioned in the Declaration.[63] In a nutshell, Ruggie implies that the RtR has been with us long enough to be accepted as legitimate. But there is more to these instruments than meets the eye. A closer look at how soft law deals with the responsibilities of a particular actor in the MNE — the core company — is warranted. The analysis below questions whether the RtR of core companies is so clearly and legitimately settled as Ruggie assumes.

Firstly, the ILO Tripartite Declaration and the OECD Guidelines — two important soft law instruments — as well as the defunct UN Code for TNCs[64] from the Cold War era and the 2004 UN Norms on business and human rights[65] have all adopted a definition of MNEs that allows these instruments to address responsibilities simultaneously, and without further distinctions, for each and every individual company as well as to the enterprise as a whole.[66] This is a deliberate choice of the OECD and ILO which is intrinsically linked to the non-regulatory purposes of the instruments (with the exception of the Norms which had misplaced regulatory ambitions).[67] Secondly, while these instruments note the influence or even control that the core company might have over its affiliates, it is only the OECD Guidelines that engage in detail with a core company’s responsibility. Additionally, what the OECD has to say makes the silence of the other instruments quite meaningful and problematic for RtR purposes, as will be discussed below. Thirdly, responsibility for whom or to do what? The OECD Guidelines state that ‘the different entities are expected to co-operate and to assist one another to facilitate observance of the [OECD Guidelines]’.[68] Almost identically, the ILO Tripartite Declaration expects of various entities in the MNE that ‘they will cooperate and provide assistance to one another as necessary to facilitate observance of the principles laid down in the Declaration’.[69] From these broad formulations it hardly appears that the responsibility on the core company is very demanding; to ‘assist’ and ‘cooperate’ seems miles away from a core company taking ‘responsibility’ for affiliates’ abusive operations. What soft law actually expects from core companies might be surprisingly little.

The OECD and ILO instruments could without difficulty be interpreted or modified to state that core companies are expected to take reasonable steps towards preventing and mitigating harms throughout affiliate operations. That would be fully compatible with Ruggie’s DD prescriptions. The pertinent question still remains: how do these instruments reason about the foundations of the core company’s responsibility? A majority of instruments do not address this question at all, conveying the impression that the instruments are self-legitimising; that is, signatory governments provide support on public policy grounds and thus ensure legitimacy for the RtR. This would be too simplistic a treatment. Luckily, the OECD enters a principled discussion and reveals how contestable the foundations are and how thin the actual legitimacy for Ruggie’s RtR might be.

The OECD qualifies a core company’s responsibility in numerous ways. Interestingly, the OECD carries this very important, principled discussion not in the text of the OECD Guidelines themselves, nor in the ‘Commentary’, but in the ‘Clarifications’ section. The heading is suggestively to the point: ‘Responsibilities of the various entities of a multinational enterprise’.[70] The first and key qualification comes as the OECD takes note of the legal separation of entities: ‘The question whether parent companies should assume responsibility for certain financial obligations of subsidiaries as part of good management practice raises complex problems in view of the limited liability principle embodied in adhering countries’ national laws’.[71] The OECD Guidelines cannot supersede or substitute national laws governing corporate liability. ‘They do not therefore imply an unqualified principle of parent company responsibility’.[72]

What are the implications for Ruggie’s RtR? I submit that this simple qualification extinguishes altogether a core company’s responsibility for affiliate misconduct unless: 1) the fault of the core company is established, which amounts to a classical, direct responsibility for own faulty conduct; 2) public policy expressly overrides limited liability, which is tantamount to strict liability; 3) the company taking DD steps can justify the expenses solely as good risk-management practice, which is the common insight of the ‘business case’ of CSR arguing for voluntary approach based on enlightened self-interest; or 4) the taking of DD steps has immaterial costs for the company, which is just old-style charity thinking with expenses limited to the philanthropy budget.

There can be no doubt that Ruggie or anyone else in the CSR movement would loath to see DD efforts dependent on charitable inclinations and confined to philanthropy budgets; DD proposals are not calls to be more philanthropic, but to do business responsibly and assume the expenses necessary to ensure that. Neither could it be inferred that Ruggie or other CSR proponents build exclusively on the ‘business case’ of CSR; important as it is, the RtR relying solely on the business case rationale would be relevant only to a very small percentage of companies in a few exposed industries. For that reason, Ruggie does count on regulatory, market and social pressures to bring the less ‘enlightened’ companies into compliance. With the third and fourth bases for dealing with the legal separation principle off the table, the question is what does soft law, particularly the frank discussion within the OECD Guidelines, have to say about public policy leading to strict liability and the own fault of the core company leading to direct liability/responsibility?[73] Solely at these two junctures can soft pronouncement add important legitimacy to the RtR. Is there evidence of how the OECD reasons about public policy and a core company’s fault that might supply legitimacy to the RtR? Yes, and here one becomes mindful of some principled limitations that are explicitly laid down.

As shown below, the OECD singles out two areas where the core company has to cover the activities of the entire business group. Though it is far from clear that these are examples provided for illustrative purposes, they represent exceptional instances when a core company’s responsibility is accepted. This interpretation is further strengthened by an explicit refusal to get too specific on a model of interactions between core company and affiliate for the purposes of a RtR. Taken together with the initial refusal to recognise an unqualified principle of a core company’s responsibility in light of the limited liability principle, the OECD stance offers well circumscribed legitimacy to Ruggie’s RtR. The implication is that if this is what public policy is prepared to do to overrun the separation of entities principle, it is only reliance on the core company’s fault that can legitimise the RtR.

A clarification that a core company’s responsibilities extend to cover the entire enterprise comes from the special treatment given to reporting. The core company has a responsibility to gather, prepare and disclose information about ‘the enterprise as a whole’:

Enterprises should ensure that timely, regular, reliable and relevant information is disclosed regarding their activities, structure, financial situation and performance. This information should be disclosed for the enterprise as a whole and, where appropriate, along business lines or geographic areas.[74]

The information needs of employee representatives and tax authorities are singled out.[75] At the same time, a responsibility to report is very different from a responsibility to exercise leverage over affiliates and take some responsibility for the entire enterprise. It is one thing for policymakers to ask for reporting and another thing to ask the core company to change its own harmful decision rippling through the network or to exercise leverage over affiliates. For reasons of public policy, the OECD seems ready to support only this particular type of responsibility to report being imposed on the core company.

But the OECD actually hints at one situation where the core company would be responsible for ensuring that abuses do not occur in affiliate operations. It is the ‘actual control’ situation. The OECD Clarifications state that ‘[t]o the extent that parent companies actually exercise control over the activities of their subsidiaries, they have a responsibility for observance of the [OECD] Guidelines by those subsidiaries’.[76] Actual operational control over a subsidiary, not to mention other affiliates, is rather hard to find in the operations of MNEs and holding the parent responsible would raise minimum difficulties. Indeed, even legal liability could be established, either under the ‘lifting the veil’[77] doctrine or via direct responsibility for own fault in exercising control.[78] It is not clear whether by ‘actually exercise control’ the OECD means ‘operational’ or ‘strategic’ control. To the extent that strategic control is exercised, maybe the OECD Guidelines would legitimise a call for responsibility beyond what the law currently covers. This might be possible, as the OECD cautiously speaks of ‘a responsibility’. Most likely, this amounts to a strong responsibility to exercise leverage and not a responsibility for result; that is, fault-based rather than strict liability independent of fault. Be that as it may, the vast majority of CSR cases involve more or less autonomous affiliates over which core companies do not have control, but rather influence. In situations where only influence rather than control exists, whether the OECD considers that the core company has a responsibility to exercise leverage cannot be assumed. Ruggie breaks new ground here with his RtR, as compared to the OECD system.

What the general responsibility ‘to co-operate and to assist one another to facilitate observance of the [OECD Guidelines]’[79] implies for the core company is not immediately clear, as the OECD Clarifications note: ‘As long as enterprises can ensure this co-operation and assistance, it would be up to the various entities to decide the division of responsibilities between parent companies and local entities’.[80] Indeed, ‘it would be reasonable to expect that a “prudent enterprise” would set up whatever internal procedures would be necessary to ensure that the [OECD] Guidelines are known and applied by its various entities’.[81] This is a nod to Ruggie’s DD orientation, but the text continues in a cautionary tone with ‘[t]he [OECD] Guidelines, do not, however, imply a model of corporate decision making nor do they interfere with the way parent companies communicate with their affiliated entities’.[82] As the OECD Guidelines are currently being revised, this stance could easily be brought into line with Ruggie’s principles of DD. Until now, however, the OECD Guidelines deliberately avoided being too specific on what the core company should do in its ‘prudential efforts’; the drafters seemed reluctant to create benchmarks allowing one to judge the adequacy of DD efforts and find the core company wanting. This might be a way to impress upon readers a conscious public policy decision aimed at de-emphasising a core company’s responsibility. Ruggie’s specification and insistence on more precise DD steps is a welcome change of approach, although it seems to stretch beyond what the OECD has so far aimed to legitimately cover with its OECD Guidelines.

Ruggie rightly covers supply chains in his Framework and expects buyer companies to exercise DD in their relationships with suppliers. The OECD clearly draws the line much tighter: it excludes supply chain responsibility altogether by saying that ‘enterprises should … [e]ncourage, where practicable, business partners, including suppliers and sub-contractors, to apply principles of corporate conduct compatible with the [OECD Guidelines]’.[83] Instead, the OECD Guidelines mainly cover equity-based relations:[84] ‘the [OECD] Guidelines have been developed in the specific context of international investment by multinational enterprises and their application rests on the presence of an investment nexus’.[85] This limitation of the core company’s responsibility can nevertheless be discarded by Ruggie as an institutional, rather arbitrary limit the OECD set for itself, with no bearing on his RtR.[86]

Overall, soft law instruments have not addressed foundational issues regarding the attribution of responsibility in parent-affiliate relations, with the exception of the OECD, which dealt with the subject in a rather obscure ‘Clarifications’ section. The OECD, attuned to legal realities, proposes a fault-based responsibility for core companies, in tune with the limited liability principle. Fault is nowhere discussed explicitly, neither are schemes of attribution of responsibility, such as the doctrines of complicity or negligence. For the OECD this is not a problem because the OECD Guidelines either propose responsibilities for core companies in keeping with current legal systems or speak of a ‘role’, an ‘opportunity’, an ‘encouragement’ for the core company to use its influence. The rest of the soft law documents are silent about their relation with current laws and schemes of attributing responsibility to the core company for affiliates’ operations. For example, regarding complicity, only the UN Norms mention it once in the ‘Commentary’, not even the main text. This lack of concern in soft law with finding and presenting the various grounds for holding the parent company accountable has worked to confuse CSR writers. The ICJ rightly noted a common ‘failure to distinguish correctly between situations in which a parent is allegedly liable on the basis of its own faulty conduct, and situations in which a court is asked to “pierce the corporate veil” and hold a parent company vicariously liable for the acts of its subsidiary’.[87]

Finally, the OECD Guidelines and other soft law instruments are indeed not hard law instruments. They do nevertheless represent ‘firm expectations’[88] of governments and they are statements of public policy. However their nature must somehow be given weight in the legitimacy discussion, potentially decreasing further the legitimacy provided to Ruggie’s RtR. The OECD’s carefully reasoned and cautiously delineated responsibilities for core companies come from its concern not to create conflict with important legal principles prevalent in member states. From this point of view, the OECD offers a principled discussion creating a useful reference point, maybe too narrow, but from which careful extensions can be pursued.

The bottom-line is that the soft law on which Ruggie relies leaves the core company’s responsibility either poorly grounded or qualifies it significantly. In other words, uncertainty and tensions remain that render the legitimacy provided by soft law dangerously thin. It would be preferable for the SRSG Framework to acknowledge this and undertake a serious conceptual effort to address these tensions in a genuine effort to develop potential ways to strike a balance. When all is said and done, soft law is fully compatible with Ruggie’s DD orientation. However, while support for a core company’s responsibility for third party misconduct exists, it is more qualified and obscured than Ruggie makes it appear.

The OECD is currently revising the Guidelines, including the rather general ‘respect human rights’ provision. There is no doubt that the OECD will closely examine the work of Professor Ruggie[89] and most likely draw on it. It remains to be seen whether the OECD will add further conceptual clarity regarding the responsibilities of core companies. Irrespective of whether this will come in the text of the Guidelines, the Commentary or even the Clarifications section, it is hoped that the OECD will continue and expand the principled discussion of a responsibility beset by ‘complex problems’ which prevented the OECD from laying down ‘an unqualified principle of parent company responsibility’.[90]

B Societal Norm

The corporate RtR is the expression of a powerful social norm at the heart of a transnational normative regime, Ruggie writes.[91] The baseline expectation is for companies, and business activities, to not infringe on the rights of others. Ruggie identifies correctly a norm based on widespread outrage at the practice of large companies that relinquished responsibility as they outsourced operations in search for new markets and opportunities. Indeed the whole CSR movement that gathered steam from the mid-1990s is based on this outrage that questioned the fairness of economic globalisation characterised by intensified foreign investment and international trade. It is undebatable that Ruggie has captured this unrest simply and effectively with his RtR; the DD steps he proposes are fully aligned with the reasonable, popular expectations for companies not to remain ignorant and passive bystanders as affiliates infringe rights.

One cannot overlook however that such a social norm, in the situation of core companies, is on a collision course with another important norm: that of legal separation of entities (herein ‘separation of entities norm’ or ‘limited liability norm’). Much less glamorous, with its often-questioned corollary of limited liability, the separation of entities norm represents more than a legal principle common to all advanced legal systems, as noted by the OECD Guidelines above.[92] It is an economic norm allowing investors and companies to spread and manage their risks in large, complex and efficient economic undertakings that have delivered enormous social gains.[93] Indeed, the law simply rubber stamped economic imperatives that delivered economic growth and societal benefits.[94] Importantly, when it comes to RtR as applied to core companies in large business groups, Blumberg noted that the efficiency of limited liability increases with the size of the entity.[95]

The existence of this norm did not elude Ruggie who wrote in his 2008 Report:

the legal framework regulating transnational corporations operates much as it did long before the recent wave of globalization. A parent company and its subsidiaries continue to be construed as distinct legal entities. Therefore, the parent company is generally not liable for wrongs committed by a subsidiary, even where it is the sole shareholder, unless the subsidiary is under such close operational control by the parent that it can be seen as its mere agent. Furthermore, despite the transformative changes in the global economic landscape generated by offshore sourcing, purchasing goods and services even from sole suppliers remains an unrelated party transaction. Factors such as these make it exceedingly difficult to hold the extended enterprise accountable for human rights harm.[96]

It appears that Ruggie properly identifies an emerging societal norm in favour of the RtR and correctly accounts for the limited liability norm. However, he appears to relegate the latter norm to the status of a legal technicality to be contested in courts of law alone. Indeed Ruggie treats this in his Pillar Three discussions regarding remedies and judicial mechanisms. There he takes note of ‘challenges stemming from the complexity of modern corporate structures’[97] and particularly the legal challenge regarding ‘the attribution of responsibility among members of a corporate group’.[98] He writes that ‘applying those provisions [civil or criminal law] to corporate groups can prove extremely complex, even in purely domestic cases’.[99] Then he acknowledges various legal grounds based in negligence, complicity and agency before concluding: ‘In short, far greater clarity is needed regarding the responsibility of corporate parents and groups for the purposes of remedy’.[100] Thus Ruggie takes stock of the separation of entities norm in a descriptive way and arguably misplaces it in his conceptual Framework: it is not a matter of remedy alone, but a substantive issue at the heart of the RtR in Pillar Two that affects the legitimate existence of the RtR and its applicability in practice.

This cursory treatment of the separation of entities norm is facilitated by the fact that Ruggie counts not only on legal remedies but also on non-legal remedies. Indeed his Framework incorporates social and economic incentives that can be remarkably consequential and effective, independent of legal incentives. While Ruggie is obviously right in throwing the net wide, there is no discussion about how the clash of norms will unfold in the regulatory arena of Pillar One or in the adjudicatory arena of Pillar Three. Neither can we discern Ruggie’s views about the odds that the CSR norm has to trump or displace the separation of entities norm. This silence sits most uncomfortably in any serious conceptual, principled treatment of RtR that aims to be consequential in real life. Should the separation of entities norm, with its strong support in public policy, prevail in the regulatory and adjudicatory arena, the RtR will depend entirely on voluntary uptake by core companies and on non-legal remedies and social pressure. This result can hardly help us advance towards a systemic solution in CSR. Further, it is neither desirable nor unavoidable as the RtR could be grounded in a more principled legal foundation, as argued in Part IV below. Hopefully a more careful conceptual treatment would not only enlarge the role of lawmakers and courts in legally institutionalising the RtR but also strengthen the hand of pro-CSR stakeholders as they confront and bargain with the CSR-sceptical establishment.

My contention is that relegating the separation norm to the status of legal technicality is a mistake. Rationalising away the aforementioned conflict of norms clouding over the RtR of core companies is short-sighted. Notwithstanding the power of the CSR norm, the separation of entities norm significantly weakens the legitimacy of the foundation of RtR as applied to core companies. The separation of entities norm is right at the centre of discussions of a core company’s responsibility and represents a legal principle, an economic norm and a very important societal norm on which modern societies depend for their welfare. Ruggie might be too optimistic when he writes that this is ‘a well established and institutionalized social norm’.[101] It would be preferable for him to soberly account for this clash of norms around core companies and examine ways forward. The reference[102] made in the OECD Guidelines in the section ‘Concepts and Principles’ is there for a reason: it belongs with any principled discussion about RtR applied to business groups that aims for persuasiveness and legitimacy.

C Complicity — Notion and Jurisprudence

The third possible building block at the foundation of the RtR is the idea of complicity. Ruggie covers the operations of affiliates with the neutral term of ‘relationships’ and also uses complicity as a notion able to cover responsibility for the conduct of third parties. Complicity, of course, is a legal concept providing a conceptual scheme of attributing liability/responsibility to an actor when a third party commits abuses. The key factors therein and the applicable thresholds are known from jurisprudence. By accepting the non-legal meaning of complicity, Ruggie might indicate that he is ready to lower the thresholds and allow the complicity idea inform discussions where courts of law would be unable to deliver a finding of liability. However, varying the thresholds does not mean that Ruggie accepts the scheme of attribution provided by complicity jurisprudence for the purposes of RtR. Neither does he deem it necessary to provide a scheme of attribution for the purposes of his mandate, such as that of complicity, not to mention committing himself to such a scheme. Therefore, he is not bound by jurisprudence and the complicity scheme adopted therein in his treatment of the RtR. In short, complicity appears for Ruggie to be a useful notion in Pillar Two, able to impress that responsibility attaches even in situations where third parties’ operations infringe rights. Complicity jurisprudence and the scheme of attribution it provides is an important legal remedy accounted for in Pillar Three.

Ruggie’s formulations on complicity are consistent with his broad use of the term in both its legal and non-legal meanings; they also result from his tendency to not elaborate in detail on the legitimate foundations of RtR and therefore the lack of need to draw on jurisprudence. So Ruggie is consistent in his use of ‘complicity’ but at a price of creating ambiguities for his readers who are tempted to draw on jurisprudence to strengthen and legitimise the CSR case. The multitude of meanings of complicity transpires from Ruggie’s formulation of DD that refers to being merely ‘associated’ or ‘indirectly involved’ with wrongdoers. A company should ensure that ‘it is not implicated in third party harm’ and should assess whether it might ‘contribute to or be associated with harm’ caused by partners.[103] Contribution and association are very different things with the second surely not belonging to complicity. Furthermore, complicity refers ‘to indirect involvement by companies in human rights abuses — where the actual harm is committed by another party’,[104] Ruggie wrote. Of course, ‘involvement’ is so imprecise a term that it can equally mean contribution/assistance and association.

The question raised by this article is whether Ruggie’s treatment of jurisprudence, as exemplified by the complicity discussion, is most fruitful for the purposes of RtR, especially for laying the foundations for the two key sub-responsibilities of gathering information and acting upon it. These sub-responsibilities, it will be remembered, encompass the key DD steps that Ruggie recommends to all companies in all situations as a universal baseline. So far this article has separated the special situation of core companies and found weaknesses in Ruggie’s reliance on soft law and the social norm. I now turn to complicity and its jurisprudence as a third possible foundational block. For those CSR observers tempted to rely on jurisprudence and complicity as a scheme of attributing responsibility to core companies, the question is how much variation in the thresholds of these schemes is possible before parting with the internal logic of complicity.

Complicity, often referred to as ‘aiding and abetting’, contains three key elements: assistance, knowledge and substantive effect.[105] Based on this scheme of attribution one can legitimately say that a company, such as the core company herein, shall not act (actus reus) with knowledge (mens rea) in a way that assists the perpetrator, the affiliate, in infringing human rights. Otherwise one is an accessory that aids the principal offender.[106] It remains to be seen how helpful this scheme is where the company did not know its conduct was somehow assisting the perpetrator and/or did not act to assist the perpetrator. The basic challenge raised by complicity is: could the actus reus and the mens rea in the complicity scheme be ‘relaxed’ to cover ignorance and/or conduct in the form of inaction by simply adopting a broader, non-legal definition of complicity, without going outside the logic of the scheme? An answer is needed because the SRSG reports and CSR writers tend to mention in one breath that complicity can result through either one’s actions or inactions.[107] Furthermore, these same writings often declare that a company will be deemed complicit when it knew or should have known that its conduct was assisting the perpetrator.[108]

But how does jurisprudence really handle cases of conduct by omission and the absence of knowledge? The answers are consequential because if omissions do not lead to complicity, there is no responsibility to act in the first place and therefore no sub-responsibilities to collect information and to act upon it. Where the core company did actually act and somehow assisted the perpetrator, but did not have the knowledge (actual or imputed from the circumstances)[109] that it was doing so and therefore is not complicit, there is no sub-responsibility to collect information in the first place, and therefore no sub-responsibility to act upon it to prevent or remedy the harm. In a nutshell, if complicity jurisprudence refuses to attach liability to conduct by omission or conduct by commission while ignorant, then jurisprudence cannot be a building block under the RtR and offers no legitimacy to calls that the two sub-responsibilities should be undertaken. Reliance on other foundational blocks must be sought.

Regarding cases of a core company’s passivity when affiliates infringe rights, the question is about potentially culpable conduct by omission. The analysis has to look into the actus reus of complicity jurisprudence. Too often one can find documents referring to corporate conduct by omission as complicity.[110] Indeed, not only does Ruggie use complicity to refer to a core company’s decisions which have rippling impacts, but also to capture situations where the core company merely stood by without any of its decisions contributing to harm caused by affiliates’ operations; Ruggie’s actus reus of complicity appears to include commission and omission. Complicity by omission in Clapham and Jerbi’s terminology could be beneficial or silent complicity.[111] But can the actus reus element of complicity really accommodate a shift from action to inaction, with both forms of conduct being treated as contributions aiding the perpetrator? Doubtfully.[112] Omissions could surely be seen as ‘culpable’, but it is not a culpability based on aiding/contributing to the violation. Rather, it is a culpability based on association with the wrongdoer, for benefiting from the latter’s operations and so on. Blameworthiness due to association strains the logic of complicity jurisprudence beyond breaking point.[113] The result is that complicity jurisprudence fails to provide Ruggie’s followers with the much needed support for a responsibility to act; when no such responsibility exists, the two sub-responsibilities disappear as well. The core company does not have to gather information and act upon it.

Regarding situations where the core company did act somehow to assist the perpetrator, but did not have the information and knowledge it was doing so, the question is about potentially culpable ignorance. For RtR proponents the challenge is to ground Ruggie’s request for HRIAs and monitoring in complicity jurisprudence. Can the mens rea element cope with this expectation? The legal concept of complicity, varying with national jurisdictions, covers contributions undertaken with knowledge. In some jurisdictions, mere awareness will suffice; in others, stricter requirements of acting with purpose — or intention — apply. Whatever the higher threshold of knowledge is, the problem is at the lower threshold: an ignorant accomplice is no accomplice at all. There is no ‘should have known’ standard in complicity jurisprudence. Ruggie inadvertently increased confusion when he wrote on the ‘should have known’ standard:

Legal interpretations of ‘having knowledge’ vary. When applied to companies, it might require that there be actual knowledge, or that the company ‘should have known’, that its actions or omissions would contribute to a human rights abuse. Knowledge may be inferred from both direct and circumstantial facts. The ‘should have known’ standard is what a company could reasonably be expected to know under the circumstances.[114]

Ruggie confuses imputed knowledge with the knowledge a reasonable person has. He refers to both as ‘should have known’ but the difference is clear: in the former case, the actor had the knowledge but denies it; in the latter case, it is agreed the actor had no knowledge, but such information should have been gathered and knowledge should have been formed, and here lies the culpability. This standard is well established in negligence jurisprudence, as explained in Part IV below, but it is alien to complicity jurisprudence. Based on complicity jurisprudence, one can hardly argue legitimately that a core company’s ignorance can lead to complicity. Ruggie’s sub-responsibility for a core company to actively seek information about potentially contributing to the main perpetrator’s (the affiliate) misconduct vanishes and so does the sub-responsibility to act to prevent or remedy abuse.

The risk now becomes clear: the demands on the internal logic of the complicity scheme are too much. Its elements can be interpreted differently than in courts of law and the thresholds in each element necessary to find liability can be relaxed. Variations of degree may lower the thresholds: less knowledge of circumstances of the abuse, of perpetrators’ conduct and of a company’s own assistance might suffice for the mens rea and lesser acts of assistance might be enough for the actus reus to attract complicity charges in CSR. Yet some seem to mistakenly use this scheme to arrive not at new, lower thresholds but at different elements altogether. Basically, complicity jurisprudence cannot accommodate these shifts from ‘actual knowledge’ to a ‘should have known’ standard and from commission to pure omission; these are not variations of degree but different concepts altogether.

My thesis is that complicity as a concept, and all the jurisprudential wisdom accumulated there so far, is unhelpful in legitimising, on the one hand, the existence of a RtR as a core company’s responsibility to act when affiliates infringe rights where a core company’s own decisions did not contribute to harm in any way and, on the other hand, the sub-responsibility to acquire information when the core company did somehow assist the affiliate/perpetrator but without having the information and knowledge that it was doing so. Both these situations are decisive blows to anyone attempting to use the authoritativeness of complicity jurisprudence to strengthen the legitimacy and persuasiveness of Ruggie’s DD steps. Because complicity jurisprudence provides no legitimacy in such situations, Ruggie’s followers can only default on the raw force of the social norm and on non-legal meanings of ‘complicity’. To strengthen the foundation of the RtR, Ruggie could take a hard look at jurisprudence as a source of legitimacy for his DD recommendations and employ a suitable concept to cover core companies that did not contribute to harm though own decisions and/or are ignorant. As argued in Part IV below, negligence jurisprudence is more able to establish a duty to act and cover ignorance, and thus provide crucial legitimacy to the RtR. The legal thresholds can still be lowered, which undoubtedly will be necessary for RtR purposes, but Ruggie and his followers will still argue in the shadow of law.

D Non-Legal Meaning of ‘Complicity’

The non-legal meaning of complicity is no doubt useful for advocacy and rhetorical purposes. There can be no objections here: this is the social norm in action. But for more rigorous, conceptual treatments of corporate responsibility, the above distinctions can be consequential. If the last decade carries any lesson, CSR writers are prone to fall into the trap of believing that responsibility for third party abuse can always be treated as complicity and thus legitimately advanced, no matter whether or not the core company acted to aid the perpetrator or had knowledge. Aware of the narrow legal definition, such writers contentedly invoke Clapham and Jerbi’s three-part complicity scheme: direct, beneficial and silent complicity. The latter has been referenced by important CSR initiatives like the Global Compact, the ISO, and even by the SRSG.[115] By repeated invocation it became unquestioned. Initially it must have been a way of drawing attention to responsibility for third party abuses, a formulation meant to prop up advocacy with a criminal law term, and to make callous companies be aware that social and market sanctions might follow even when the law falls short. But now this scheme might have only reinforced something akin to a conceptual stranglehold. Clapham and Jerbi’s scheme was not even a conceptualisation that backfired, but a mere ‘categorisation’, as the authors themselves state in the title, which placed three adjectives in front of ‘complicity’.

In fairness, Ruggie does not use complicity to carry a principled discussion and to clarify how responsibility is allocated to the core company. Ruggie uses complicity to point to the raw force of the social norm (society has framed its concerns against core companies in terms of complicity) that covers the ‘relationships’ a company has with third parties and/or a way to account for legal and non-legal ‘remedies’. Courts of law will hear charges of criminal complicity as the courts of public opinion will rule on charges of culpable association with, or involvement in, affiliate misconduct (non-legal complicity). Ruggie takes note of these various possible foundations but does not aim to commit himself to one of them — complicity or negligence — as the ‘right’ way to attribute liability or responsibility to core companies, nor does he analyse their comparative strengths. The expectation is that any such doctrines can be employed depending on circumstances and the legal or non-legal arena where the charges are made. Be this as it may, the justification of what really happens to the core company remains somehow in suspension between Ruggie’s Pillar Two that discusses the DD principles of what a company should do and his Pillar Three that addresses what remedies will be brought to bear on companies that do not follow the DD steps.

My concern is that this important principled discussion around the RtR of corporate groups hinges uncertainly, for the time being, between the silence in Pillar Two and the pragmatism in Pillar Three. The absence of this principled persuasive argumentation on RtR leaves the working of remedies in Pillar Three and the practical relevance on the RtR as a whole at the mercy of powerful attacks from the separation of entities norm. Furthermore, diligent readers of the SRSG reports looking in vain for a principled discussion and a scheme of attribution endorsed by Ruggie will likely default on the complicity idea, which has had a much higher profile in RtR discussions[116] than other legal concepts mentioned only cursorily and descriptively in Pillar Three. This over-reliance on complicity is only augmented by the fact that most serious cases of corporate abuses, though not the most numerous, can in fact be dealt with properly as complicity. However, this only makes it more difficult to realise that the rest of the cases do require a different scheme of attribution and appeal to a different kind of jurisprudence. This difficulty is made obvious by countless writings of legal scholars that succumb to the temptation to argue improperly in terms of non-legal complicity thus obscuring further the weak conceptual treatment and illusory legitimacy of the foundation of RtR as applied to core companies.

The overuse of complicity has eventually obscured the need to search deeper for other more able concepts. This treatment does not reflect the internal logic of complicity and therefore does not facilitate the underpinning of CSR with legal foundations. Indeed, complicity jurisprudence has nothing to say on the ‘should have known’ standard (the knowledge element) and precious little on omission (the conduct element). Furthermore, this treatment does not add any new measure of legitimacy to the RtR because non-legal complicity is a mere reaffirmation of the social norm on which Ruggie already relies directly. By hindering a search for other more able legal concepts to conceptualise the RtR, the misuse of complicity has worked to prevent Ruggie and CSR writers from accessing new reservoirs of knowledge and legitimacy that would have come from arguing in the shadow of law and of proper legal concepts.

IV A CORE COMPANY’S RTR: A WAY FORWARD

In his 2009 Report, Ruggie said that he ‘will continue to explore how human rights due diligence might legitimately vary across businesses of different roles and sizes, as he provides a principled elaboration of human rights due diligence applicable to all businesses’.[117] The previous section found it problematic that a core company’s situation has not been analytically separated for the purposes of RtR. It also found that the three foundational blocks offer the RtR an insufficient foundation in the case of core companies and in those situations where its own decisions did not have direct or indirect impacts throughout affiliates’ operations. As a result, this small but important part of the RtR remains seriously contestable. This section seeks a principled treatment of the problem and draws attention to the promise of negligence jurisprudence. The aim is to offer support to Ruggie’s two sub-responsibilities to form knowledge and to act to prevent or mitigate abuses taking place in affiliates’ activities.

This article has already highlighted the need to reinforce the RtR in a critical place; to reiterate, the social norm is strong and vital for making the RtR case, but it is on a collision course with the separation of entities norm. Soft law generally aligns with the RtR but soft law instruments are either silent on the core companies’ responsibilities or refrain from an unqualified endorsement by drawing attention to complexity in light of the limited liability principle. Finally, complicity works well for conduct by commission with knowledge, but complicity jurisprudence offers RtR neither insight nor legitimacy in cases of conduct by omission[118] and/or absence of knowledge.

The unsettling conclusion is that Ruggie’s proposals for a core company’s responsibility to gather information through HRIAs and monitoring, and the responsibility to act on it, rest on a weak foundation when a core company did not do anything to contribute to the harm; in other words, its own business decisions had no direct or indirect impacts on affiliates. Furthermore, there is no principled discussion of these responsibilities in the special case of core companies because the SRSG Reports treat all companies the same for RtR purposes. This in effect asks sympathetic readers to make a leap of faith in embracing the RtR and offer less likeminded observers a golden opportunity to discredit this important part of the RtR.

A The Promise of Negligence Jurisprudence

Negligence jurisprudence can fill the gaps in RtR argumentation left open by an uncritical reliance on complicity. In short, there are criteria for liability for omissions to act on which Ruggie’s RtR can naturally build, and there is a ‘should have known’ standard on which the sub-responsibility to gather information can easily draw. In the case of omissions, first it should be established that the core company has a responsibility to act (the duty of care to exist). If this is the case, then both the sub-responsibility to have the knowledge of a ‘reasonable person’ and the sub-responsibility to act on that knowledge follow. Certain conditions must be met and certain thresholds must be reached to deliver a finding of liability or, in CSR, responsibility for third party abuses. Negligence jurisprudence can clarify both the legitimate foundation of the duty of care and the scope[119] of the two sub-responsibilities.

Negligence law lays down the basic rule that one is not responsible for third party misconduct. US negligence law, for example, has clear principles such as: ‘there is no duty so to control the conduct of a third person so as to prevent him from causing physical harm to another’;[120] and ‘the employer of an independent contractor is not liable for physical harm caused to another by an act or omission of the contractor’.[121] Importantly though, there is a long list of exceptions to such principles.

The first principle continues by saying that no duty to act exists unless:

(a) a special relation exists between the actor and the third person [tortfeasor] which imposes a duty upon the actor to control the third person’s conduct; or

(b) a special relation exists between the actor and the other [injured party] which gives to the other a right to protection.[122]

Such special relations expressly exist in situations such as where the actor created a situation of peril through his own conduct resulting in a duty to act to prevent harm,[123] or where the actor has control over third parties committing the abuse resulting in a duty to exercise such control.[124]

Regarding the second principle on companies outsourcing to independent contractors, the American Law Institute’s (‘ALI’) Restatement on Torts lays down a long list of exceptions that fall into three very broad categories: negligence of the employer in selecting, instructing, or supervising the contractor; non-delegable duties of the employer; and work which is ‘specially, peculiarly, or “inherently” dangerous’.[125]

A core company’s conduct is mediated by its affiliate misconduct and causality becomes complicated. A finding of insufficient causality can preclude liability altogether. The fact that third party misconduct was foreseeable may relieve problems of causality. According to ALI, ‘[a] negligent act or omission may be one which involves an unreasonable risk of harm to another through ... the foreseeable action of the other, a third person, an animal, or a force of nature’.[126] Foreseeability thus goes some way in countering defences that intervening forces[127] break causality and relieve the core company of responsibility. To further highlight the potential that negligence jurisprudence has to enrich the SRSG mandate, one needs to look no further than ALI’s nuanced explanation that contains elements strikingly familiar to CSR discussions:

It is not possible to state definite rules as to when the actor is required to take precautions against intentional or criminal misconduct. As in other cases of negligence … it is a matter of balancing the magnitude of the risk against the utility of the actor’s conduct. Factors to be considered are the known character, past conduct, and tendencies of the person whose intentional conduct causes the harm, the temptation or opportunity which the situation may afford him for such misconduct, the gravity of the harm which may result, and the possibility that some other person will assume the responsibility for preventing the conduct or the harm, together with the burden of the precautions which the actor would be required to take.[128]

What is important to grasp from this brief, merely suggestive list of exceptions are the norms advanced by negligence law. Apparently simple and non-controversial principles of non-responsibility when third parties’ operations produce harm had to be qualified in law. The exceptions have incrementally piled up, as suggested by ALI:

the law has progressed by the recognition of a large number of ‘exceptions’ to the ‘general rule’. These exceptions are stated in §§ 410–429. They are so numerous, and they have so far eroded the ‘general rule’, that it can now be said to be ‘general’ only in the sense that it is applied where no good reason is found for departing from it … Indeed it would be proper to say that the rule is now primarily important as a preamble to the catalog of its exceptions.[129]

The implication for the RtR is straightforward: Ruggie could draw on these exceptions in law. Of course, their mere invocation will not by itself make the RtR legitimate. Exceptions have to be interpreted narrowly and analogies should be sound. But Ruggie could find a way to capture the reasoning behind these exceptions and the spectacular evolution of law. In this area, negligence jurisprudence dovetails with Ruggie’s societal norm behind the RtR. So far Ruggie has failed to capitalise conceptually on this crucial potential alignment of authoritative jurisprudence with the newer societal norm advanced by the CSR movement. Should Ruggie choose to address explicitly and seriously the foundational issues of RtR and draw on favourable jurisprudence, the result would be a strengthened foundation for the RtR that comes from careful reasoning in the shadow of the law. This would not only compound the legitimacy that the social norm grants to the RtR, but also assist in advancing a principled discussion on why the core company has a responsibility to act when affiliates’ operations infringe human rights, even in situations where a core company’s own decisions have not contributed to harm through direct or indirect impacts.

The list of the exceptions is long; however, having the status of ‘exceptions’, the legal thresholds to bring these exceptions to work are high. Ruggie is not bound by these thresholds but he would be well advised to tread carefully or lose the legitimacy payoff. Furthermore, in negligence jurisprudence these exceptions are sometimes based on fault, whereas other times they are independent of fault and based on reasons of public policy. However, such public policy considerations might be very different in national and international contexts. It is important not to lose the complex institutional context in which these exceptions are invoked. In short, sound analogies are not easy to make but this should not detract from acknowledging the normative orientation behind these exceptions and using it to the advantage of the RtR.

Leaving legitimacy aside, that negligence jurisprudence can be a source of valuable insights is visible also from its explicit treatments of the ‘reasonable person’, an evaluative cornerstone in negligence law. ALI describes the qualities of the ‘reasonable man’ as ‘attention, knowledge, intelligence, and judgment which society requires of its members for the protection of their own interests and the interests of others’.[130] What the law expects in terms of knowledge from a reasonable person — what the diligent actor should know — is explained in ALI’s Restatement on Torts:

For the purpose of determining whether the actor should recognize that his conduct involves a risk, he is required to know (a) the qualities and habits of human beings and animals and the qualities, characteristics, and capacities of things and forces in so far as they are matters of common knowledge at the time and in the community; and (b) the common law, legislative enactments, and general customs in so far as they are likely to affect the conduct of the other or third persons.[131]

ALI states that the actor is, under certain circumstances, ‘bound to anticipate and provide against the negligent or intentional misconduct of … a third person’.[132] ‘[T]he actor, as a reasonable man, must therefore take life as it is and not as it should be, and must realize the likelihood that third persons may act in a variety of ways, all of which are not only morally but legally wrongful’.[133] So

if the known or knowable peculiarities of even a small percentage of human beings, or of a particular individual or class of individuals, are such as to lead the actor to realize the chance of eccentric and improper action, he is required to take this chance into account if serious harm to a legally important interest is likely to result from such eccentric action and his own conduct has not such pre-eminent social utility as to justify the serious character of the risk involved therein.[134]

Taking note that the ‘reasonable person’ standard is a key concept in negligence jurisprudence, the ICJ explains the legal threshold of knowledge in negligence law:

In the law of civil remedies, liability can arise even where a company has no knowledge as to the risk of harm, because the law may hold that it should have known, as the risk was reasonably foreseeable ... [I]n determining whether a company is liable, courts in both common law and civil law countries will ask whether a reasonable person in the company’s shoes, with the information reasonably available at that time, would have known that there was a risk that its actions could harm a person. This means that the court will look at both what the company itself knew, and what a reasonable company in its shoes would have known about the risk that harm would occur. The civil law term ‘reasonable person’ does not mean an average person, but a responsible, careful member of society. In this way, the fact that a company did not know there was a risk of harm will be irrelevant under the law of civil remedies, as the law will regularly hold that in fact it should have known.[135]

It appears that negligence jurisprudence is packed with such suggestive provisions and formulations.[136] In fairness, the SRSG team has explored some of the terrain of negligence jurisprudence[137] and noted case law where plaintiffs ‘claim that corporations should be held accountable for their indirect participation in human rights abuses’.[138] The SRSG team’s special paper on complicity takes note of other legal doctrines, including negligence law, under the heading ‘Guidance from law’.[139] Such guidance should be encouraged, but so far the SRSG reports do not seem to integrate fully those insights, particularly when it comes to omissions of core companies. His team has confined itself to noting the existence of this legal basis — the tort law of negligence instead of the criminal law of complicity — and that the hurdles are more relaxed in civil law cases, as is the case, for example, with the knowledge requirement.[140] That these discussions have found their place in Pillar Three dedicated to remedies is only symptomatic that Ruggie has chosen to acknowledge the importance of jurisprudence pragmatically as a matter of remedies and not conceptually as a source of legitimacy and insight for the RtR. This, I would argue, is a missed opportunity.

Furthermore, from the material available in the public domain, it seems that Ruggie has received limited prompting and guidance from the legal community on why and how to offer a principled treatment of RtR by making more use of jurisprudence. For example, the ICJ undertook its large-scale comparative study on corporate responsibility for third party abuse,[141] but for various reasons relied on complicity as the legal and non-legal concept covering this important area of CSR. It is unfortunate that the ICJ ably presented principles of civil law (mainly negligence), but fell short in explaining why the negligence scheme has more potential than the complicity scheme to strengthen Ruggie’s RtR argumentation. Complicity jurisprudence, as explained, does not operate on a ‘should have known’ standard but on the knowledge the actor actually had. Nor is conduct by omission a significant part of complicity jurisprudence, which forecloses the chance of insightful treatments on how to reason about the responsibility to act. Overall this makes it impossible to reason in the shadow of law and to derive from jurisprudence some legitimacy for the RtR. As a result, despite the excellent and vast comparative study that the ICJ contributed, it does not seem to have had an effect on Ruggie’s conceptualisation of the RtR. I maintain the ICJ did all a disservice by affixing the complicity label on its important analysis and forcing it under such an overstretched concept poorly suited for the task of legitimising and clarifying the corporate responsibility for third party abuse. Furthermore, the ICJ might have failed to make clear enough the need for a principled argumentation on the RtR as applied to core companies in corporate groups and networks; indeed the ICJ treated all companies uniformly for purposes of RtR just as Ruggie has been doing.[142]

The ICJ report rightly observed that vast experience has already accumulated in the law of negligence: ‘[t]he laws of tort and non-contractual obligations are hundreds of years old, and in all jurisdictions have regulated the interactions of different actors, including businesses, in society, long before international human rights standards were developed’.[143] One can only take heart from the ICJ’s observation that ‘the basic principles of criminal and civil legal responsibility are clear’.[144] However one cannot underestimate the flexibility and tensions within the negligence scheme of attribution. There is a need for adaptation to transnational CSR and for profound discussions around each of the key concepts to arrive at narrower or broader interpretations.[145] Clearly the legal thresholds of negligence will need to be discussed and carefully adapted. This task is one that the SRSG mandate could significantly advance.

The influence Ruggie has already had is notable in different areas: from the recent excellent report of the UN Special Rapporteur on Indigenous Peoples, elaborating Ruggie’s DD in that specific human rights area;[146] to the impressive study the ICJ conducted in 2008 in an attempt to bring some jurisprudential insights into CSR discussions. Regarding the ICJ’s adaptation of legal concepts to the CSR area, see for example the matter of ‘proximity’. The closeness or remoteness between the parent’s decision and the rippling impacts overseas will be a key issue in a negligence analysis: ‘the closeness of a company to the principal perpetrator, to the victims, or to the harm inflicted on the victims, is highly relevant in determining legal responsibility’.[147] The ICJ further explained the multifaceted concept of proximity:

1. geographical proximity: ‘[a] company may have more knowledge and more opportunity to influence events if the human rights abuses are occurring in same place, or nearby, the company’s operations’;[148]

2. economic and political relationships: ‘[i]n practice, the more a company economically dominates a marketplace, the more it has access to the corridors of power, access to inside information and the opportunity to influence the actions of third parties who depend on the business relationship’;[149]

3. intensity, duration and texture of relationships: ‘[t]he quality of the relationship, the openness, closeness, frequency and duration of informal or personal contacts and discussion will also be evidence towards the degree of proximity between a company and perpetrators or victims’;[150] and

4. legal relationships: ‘[a] company may have considerable control, influence and knowledge because of the legal nature of the business relationship it has with a third party that violates rights. A joint venture or other long-term strategic partnership may lead to shared decision-making and close coordination between the parties. Despite the fiction that every legal entity is completely separate, the relationship of parent-subsidiary or cross-membership of boards between different companies in long-term business arrangements will sometimes lead to a proximity that increases the shared knowledge and influence’.[151]

To conclude this analysis, Ruggie has not carried out a principled discussion explaining why the core company has a responsibility to act when none of its own decisions has direct or indirect impacts on affiliates’ misconduct, nor has he perceived the need for such a discussion and for a scheme for attributing responsibility. He pragmatically leaves both the freedom and the responsibility to remedy providers in Pillar Three and lawmakers in Pillar One to institutionalise the RtR in any way they can, and deem best, to fit to their particular institutional contexts, which of course vary greatly from country to country and cover diverse legal and non-legal arenas. With that freedom Ruggie has also passed over the burden to pro-CSR lawmakers and remedy-providers to battle with the separation norm. Ruggie’s silence on a key conceptual aspect of the RtR appears at odds with his early stated ambition to develop a ‘principles-based conceptual and policy framework … a foundation on which thinking and action can build’[152] and his expectation that the United Nations, taking into account the results of his mandate, ‘can and must lead intellectually’[153] in the business and human rights area.

By not elaborating his reasoning on the core company’s activities, Ruggie has come to rely on the raw power of the social norm favouring the RtR. Soft law, silent on the issue or strongly qualifying the principle of the core company’s responsibility to act, adds precious little. The notion of complicity, a mere label to make the social norm more intelligible and striking for advocacy purposes, adds nothing conceptually. Neither could complicity jurisprudence, if Ruggie had decided to draw on it for increased authoritativeness, have added more. In the same time, Ruggie takes note of negligence law, but relegates it to one of the many remedies enumerated in Pillar Three. In other words, Ruggie acknowledged the law which might provide remedies, but failed to acknowledge jurisprudence that might offer a source of legitimacy in a principled, conceptual justification of the RtR. By doing that, he misses the chance to strengthen the foundation of the RtR and the opportunity to gather insights from this body of law about the scope of corporate DD efforts. Charges of illegitimacy and open-endedness can easily follow. The section below attempts to point to a way forward in reasoning about the RtR by placing the social norm and jurisprudence, and their interaction, at the foundation of a core company’s RtR.

B The Foundation Revisited

Ruggie can be applauded for defining his RtR to cover the activities of entire business groups and networks. Core companies in such extended enterprises remain responsible for human rights abuses that occur in their affiliates’ operations. His elaborations of the DD steps track good business practice, distil key steps of what a diligent company should do, and indentify numerous factors that can guide and limit such DD efforts. Furthermore, not only is the coverage of the RtR properly broad in light of social expectations and the content of DD conceptually streamlined, but also the institutionalisation of such DD is addressed in the form of legislation (Pillar One) and practical remedies (Pillar Three). In this commendable way, Ruggie keeps in the same picture regulatory, market and social pressures — their importance is properly acknowledged, and their complex interaction is not obscured in an effort to make the RtR relevant in practice.

These achievements of the SRSG mandate notwithstanding, they are likely to be insufficient to advance a policymaking process that is ‘coherent and cumulative’.[154] They enable policymaking, but the conceptual anchor is weak, leaving the entire structure floating precariously on strong undercurrents. Only one of these undercurrents is identified herein in the separation of entities norm. Not acknowledging the challenge can sink a part of the RtR, shortcut the policymaking process at every stage rendering it piecemeal and easy to derail, and eventually leave the RtR drifting far away from where Ruggie hoped it would sail. Reasoned discussion and principled argument is needed to lay down a conceptual foundation under a core company’s RtR and define the latter’s uneasy relationship with the separation of entities norm. Then the RtR can be presented more persuasively, which would facilitate work in numerous regulatory and remedy-providing arenas and thus move the policy-making process closer to being ‘coherent and cumulative’.

1 The RtR as Responsibility to Act

The challenge is how to justify a core company’s responsibility to act when its affiliates infringe rights, in the absence of any harmful conduct on the part of core company itself (for which, of course, it should be responsible). A responsibility to ‘act’, once established, entails two key sub-responsibilities: to gather information and to act on it to prevent and remedy harms. On what are these based so far? The raw force behind the CSR movement and the SRSG mandate is a social norm: outsourcing operations to subsidiaries and contractors does not entail outsourcing responsibility; some residual responsibility remains with the core company.

To ‘act’ means simply for the core company not to remain a passive bystander after it outsources operations to affiliates. The social norm states that a responsibility to act, to do something, does remain with the core company. It should be remembered that the conceptual difficulty does not come when the core company’s decisions have indirect impacts on affiliates’ operations thus contributing to abuses. This situation is well covered by Ruggie’s RtR, no matter whether the core company has control or influence over affiliates; we are still well within the territory of the ‘do no harm’ principle. In that case, the responsibility of the core company is to change its own decisions so as to not cause harm directly or indirectly.[155] The difficult situation is when the core company does not make any harmful decisions itself. The social norm states that something should be done because merely standing by is unacceptable.

The problem at this stage is not more precision about what should be done, but rather, to decide that anything should be done in the first place. For Ruggie’s two ‘sub-responsibilities’ to kick in, a core company’s responsibility to act must first be established. The social norm clashes at this level with the separation of entities norm. The separation of entities norm is so strong that it works to the effect that even if the core company has full ownership of the subsidiary but still gives it autonomy for its day-to-day operations, the law will shield the core company from liability if the affiliate misbehaves.[156] Should the link be weaker than 100 per cent equity or if it is a mere contractual link with a supplier, for example, the separation of entities norm appears even more compelling. This norm will be highly relevant in practice due to economic, social and public policy considerations supporting it. Given the unavoidable clash of norms, it must be better established that the core company should spend any of its resources (more than, say, its philanthropy budget) tracking and influencing its affiliates.[157] The risk is that, in the absence of a principled treatment, everything in Ruggie’s DD prescriptions that cannot be framed as grounded in the ‘do no harm’ principle will rest on thin air and be defeated in practice.

Following Ruggie, we take it as given that a core company should be responsible for the direct and indirect impacts of its own decisions. That requires knowledge: first and foremost, about where in its decision-making adverse impacts are born; how the effects of these decisions may ripple (get compounded or reduced) through the channels of the business group or network; and what exact violations of human rights take place in affiliate activities as a (partial) result of those decisions. However, as a matter of business practice, affiliates are more or less autonomous in their day-to-day operations. The more autonomy affiliates have, the more problematic the core company’s responsibility to act becomes. Why? When significant autonomy exists, Ruggie’s RtR logical chain begins to break because there is no rippling impact by definition, just the misconduct of an autonomous entity. No DD steps are called for anymore. Ruggie bases his RtR on the ‘do no harm’ principle and this is the taking of that principle to its logical conclusion when a core company’s own decisions do not cause harm.

So, in the very usual case where affiliates have significant autonomy to make business decisions that end up infringing rights, Ruggie’s logic is under significant strain. Because the core company has to be diligent only about the impacts of its own decisions, the only decision that links the core company to the abuse is the original choice of granting autonomy to a (potentially irresponsible) affiliate. If the core company should act with the DD that Ruggie recommends, this is based either on its original decision to grant autonomy or on something other than its own decisions having rippling harmful impacts.[158] Ruggie did, in some respects, point in the direction of the original decision when he defined complicity, which makes RtR applicable, as ‘contribution’ or ‘association’ with harm. But association is too broad and unrefined. The conscious decision to get associated with an affiliate and grant it autonomy seems the key. This decision is different from associating with an entity done for an illegal purpose (criminal enterprise) or vague association with the actual harm by simply being at that time in the same business enterprise. In sum, Ruggie could be clearer here either by pinpointing a prior decision of the core company to make the ‘do no harm’ principle work for him again or identify other principled, legitimate grounds outside the ‘do no harm’ principle. In other words, either go back in time to a key decision of the core company or identify other principles to support a responsibility to act. If not, the core company does not have to undertake any effort to even learn about abuses that take place in an affiliate’s business. It can remain conveniently — indeed wilfully — ignorant and likely legitimately so unless other grounds can be provided.

As shown in Part IV(A) above, negligence jurisprudence appears highly relevant for RtR purposes. It can reinforce the social norm at this stage in imposing responsibility. Although the principle of negligence of no ‘responsibility for third party misconduct’ is aligned with the separation of entities norm, jurisprudence admits numerous exceptions and could cover common CSR situations. By applying each negligence category, a responsibility to act would fall on a core company that:

• sets in motion a chain of events (prior conduct);

• non-diligently outsources to irresponsible contractors (negligent selection);

• is in a relationship with the irresponsible affiliate (special relationship); and

• undertakes conduct that creates a situation of peril (situation of peril) particularly if it involves an inherently dangerous activity (peculiar risk), with risks that were quite foreseeable (foreseeability) and affecting a vulnerable population (vulnerability).

Ruggie’s reasoning on the RtR could capitalise on this jurisprudence and seek to make analogies with the relationship of the core company and its affiliate in the CSR context. To take them one by one, rights holders in developing countries, such as local communities and workers, are often in a state of clear vulnerability given dysfunctional legal systems and power differentials. Business activities, when undertaken irresponsibly, carry well documented risks of abuse of the entire catalogue of human rights, as Ruggie correctly noted; this situation of peril is palpable for workers, trade unionists, local communities and so on. That such risks of abuse are becoming foreseeable is an understatement given the work done in CSR in the last 15 years with increased information of common types of abuses varying predictably for countries, industries, and stakeholders. It would be hard to persuade local stakeholders that the core company is not in a special relationship to the affiliate infringing their rights, or even directly to them, when local communities closely coexist with a mining company or when the workers affix the core company’s brand on each good produced. Similarly, when a buyer company outsources production to a supplier with the known propensity to infringe labour laws in a country well-known for weak protections of workers’ rights, it can look like negligent selection of an independent contractor. And finally, jurisprudence refers sometimes to such responsibility to act in terms of ‘non-delegable’ duties to impress the crucial point that outsourcing activities does not entail outsourcing responsibility as well.

They are all situations that illustrate different facets of the association between the core company and the ultimate harm that takes place due to affiliates’ activities. The jurisprudence is precise in identifying and explaining key elements such as fault, type of activity, the state of vulnerability, foreseeability of risk and so on that makes it problematic for the defendant to argue against a responsibility to act. The core company’s decision to grant and maintain autonomy of affiliates when the aforementioned elements exist becomes harder to defend and some responsibility might have to be accepted.

Not only does jurisprudence legitimise the responsibility to act but it also identifies key limiting concepts on the DD. First, the exceptions that jurisprudence recognises provide limiting concepts on a responsibility to act under the RtR. When these factors are not present, the core company does have responsibility and no further DD steps are called for. Second, there is the issue of thresholds on each factor that courts take into account. How vulnerable? How risky? How negligent in selection? How special a relationship should be? How foreseeable the risk? The exceptions in law allow courts to find liability if high thresholds are met. Ruggie could discuss these thresholds and lower them carefully.

The societal norm allows him to perform such lowering of thresholds. Such re-adaptation of thresholds in the special context of international CSR is desirable for considerations of fairness and possibly public policy. Lowering is further to be encouraged given that the RtR Ruggie advances is not only for adjudication in courts, but also more broadly for other non-legal arenas. In all arenas, however, thresholds should be varied attentively; we are still in the realm of ‘exception’ and the rule remains the legal separation of entities norm. A balancing act will be called for in light of the specific institutional context.

That a responsibility to act, that some residual responsibility remains with the core company, is now legitimately established. By outlining DD steps, Ruggie is right that core companies are mainly expected to gather information and then to act on that information to address abuses in affiliates’ operations. The social expectations behind the social norm would say, I submit, that the core company is not to be ‘ignorant’, on the one hand, and once some knowledge exists, not to make an ‘unreasonable choice’, on the other hand. But ignorant about what? And choice between what and what? These two sub-responsibilities are discussed below.

2 Sub-Responsibility to Gather Information and Form Knowledge

Once there exists a responsibility for the core company to act, it is obvious that some information must be collected in order to choose between several courses of possible action. Ignorance is not bliss but a dangerous departure from the social norm and societal expectations: society would deem such wilful ignorance repugnant. Ruggie deems this ignorance unacceptable. Still, the foundations and limits of such a sub-responsibility to gather information remain somewhat raw. This argument drawing on the social norm can be further refined through a conceptual discussion drawing on negligence jurisprudence and with a pragmatic observation about the ready availability of information.

First, negligence and its ‘should have known’ standard expects the core company to gain the knowledge that a ‘reasonable person’ (a reasonable company) would have. What is reasonable depends on many factors, including social expectations. It is a moving target in law because what was deemed reasonable a generation ago might be deemed unreasonable in the light of present day realities.[159]

Secondly, it can be pragmatically observed that core companies do have some general information about the countries where they source or distribute their goods and services, and the type of abuses their affiliates might get entangled in. Some information is available from a company’s DD efforts under the ‘do no harm’ principle: risks of abuses can be uncovered as the core company tracks its own decisions prior to granting autonomy; or, when only a degree of autonomy has been granted, decisions that the company still takes. So there is a practical overlap where affiliate’s abuses are uncovered as the company tracks the impacts of its own decisions; those abuses do not require new resources to be uncovered again. Furthermore, some other information is readily available, even of notoriety, from external sources. The resources required to access such information amount to no more than casually perusing news outlets or lending their ears to NGOs communicating their concerns to the company. Currently, as distinguished from no more than 10–20 years ago, a great amount of general information is cheaply and readily available. Indeed, information overload — rather than its scarcity — consumes more resources. While specific, reliable information still might require significant expense, the information required at this stage of the RtR discussion is general and not costly. This easy availability of information is a practical observation which has another important implication: it might be it is the formation of knowledge that consumes corporate resources rather than the gathering of information as the latter might already be available within the company or easily accessible from external sources.[160]

So as a matter of practical reality, information might be available in larger quantities and more easily accessible than previously thought. What negligence jurisprudence contributes is to establish that processing the information and forming a reasonable level of knowledge is a legal obligation; it is not optional. The foundation of the responsibility not to be ignorant no longer rests on a raw argument hinging exclusively on the social norm, but on a combination of a social norm and jurisprudence, and pragmatic realities (the twin availability of information from external sources and from within the core company as it tracks the impacts of its own decisions).

The question still remains about the scope of such information-gathering efforts. More to the point, information and knowledge about what and to what level of specificity? Obviously, information gathering and processing consumes resources. The spectre of the separation of entities norm looms over this sub-responsibility to gather information as it does over the other parts of the RtR. There must be a limit to how much information the core company that grants its affiliate autonomy should have about what takes place in the affiliate’s operations. The social norm cannot give guidance as to that limit. Of course, where a core company’s own decisions lead to harm, there is no problem as the company should spend enough to understand the impacts of its own decisions. But where no such decisions exist and the affiliate acts autonomously, an invocation of the social norm leaves the argument raw and open-ended, as does Ruggie’s correct observation that only a case-by-case, factual analysis can provide greater clarity.[161]

But jurisprudence has something to say: the ‘reasonable person’ concept employed by jurisprudence offers a ready-made conceptual limitation on the responsibility to form knowledge, a conceptual limitation which is still flexible and adaptive to evolving social realities. Furthermore, another conceptual limitation comes from the purpose of information gathering at this stage of the RtR: it should be just enough information to enable the core company to decide whether abuses occurring in affiliate operations and the remedies available therein require its own intervention or not. Such intervention of the core company is captured in the sub-responsibility to act on information, as discussed below.

The analysis leads to a better-grounded responsibility to become knowledgeable in two key situations for Ruggie’s RtR. On the one hand, where affiliates have been granted autonomy, the core company should form a reasonable level of knowledge based on general information about what human rights abuses take place in the operations of its autonomous affiliates and what effective remedies are available in host countries. On the other hand, in the ‘easier’ case where affiliates are less autonomous, the core company has responsibility for its own decisions with rippling impacts and therefore should have detailed knowledge about the exact source of harm in its decision-making process, the rippling mechanisms within the business group structures, and any concrete human rights abuses generated by its own decisions. Noteworthy, the core company would have to enhance its general knowledge about autonomous affiliates depending on the choice it makes at the next stage: the sub-responsibility to act on the information and knowledge.

3 Sub-Responsibility to Follow up on Information and Knowledge

Once it has been established that in certain circumstances the core company has a responsibility to act and a sub-responsibility to collect the general or specific information (depending on the autonomy granted to its affiliate) necessary to enable it to achieve the knowledge that a ‘reasonable person’ would have, the question is how the company should follow up on that information and knowledge. In other words, is there a further sub-responsibility to act upon that knowledge?

I submit that here the conceptual emphasis should be on choice: the core company has to choose diligently between whether to ‘withdraw’ by terminating the relationship or ‘stay’ but exercise leverage. At this point it should be noted that the option of doing nothing has been abolished by a reasoning that draws on the social norm and on negligence jurisprudence. However, a distinction is necessary: ‘doing nothing’ cannot be accepted anymore as a principle following directly from the separation of entities norm, but doing nothing could be accepted once the core company chooses to ‘stay’, yet exercising leverage is not feasible for various reasons. Thus this ‘doing nothing’ option comes only after the choice ‘withdraw or stay’ has been made, not before. This will be discussed further below.

The RtR at this moment boils down to a responsibility to choose diligently which has two key implications: one for the responsibility to act and one for the responsibility to gather information. Regarding the responsibility to act, it is clear that a ‘choice’ is an act, it is a decision of the core company that comes under the ‘do no harm’ principle just as much as other business decisions covered by Ruggie’s RtR. The negative consequences of such choice, when made improperly, can be serious indeed if one considers that hastily taken decisions to withdraw can be very problematic; for example, buyers terminating contracts with non-compliant suppliers put hundreds of child labourers in the streets and leave them prey to worse abuses. Indeed, precisely for this reason CSR recommendations uniformly point towards withdrawal as a last solution, only to be used if other measures fail to bring the affiliate into compliance. The same goes for another example: investment in undemocratic countries. Here the threshold of abuses pervasive in the country must be rather high and systemic before businesses can be accused of illegitimately doing business there. Only exceptionally will companies be asked to choose in favour of non-entering or divestment.[162]

So, in practice, the choice between withdrawing and staying (and exercising leverage) will more often point to the latter — a responsibility to exercise leverage. Here, CSR and business interests converge to support companies remaining and doing business in developing countries. However if the company can withdraw without adverse consequences for the rights of local stakeholders, this is a legitimate business choice and the company’s DD efforts and resource expenditures come to an end. This freedom to choose amounts to a conceptual limitation on the DD efforts a company is required to make and is fully concordant with the ‘do no harm’ principle on which the RtR is based.

Regarding the responsibility to gather information, it will crucially be informed by an emphasis on the diligence factor in the sub-responsibility to choose diligently. Therefore the core company has a responsibility to gather sufficiently detailed and reliable information and to form knowledge that enables it to make a reasonable choice. Information thus has to be more specific about the abuses happening in autonomous affiliate operations, but knowledge must also be formed also about the consequences of either ‘withdrawing’ or ‘staying’. Impact assessments might be called for at this stage to ensure diligence. It is clear that this responsibility to know reaches further than the general or more specific knowledge already formed at the previous stage. The conceptual limitation is again provided by the purpose of this sub-responsibility coupled with the ‘reasonable person’ standard.

Once the business decision to stay and exercise leverage has been taken, the question is how the responsibility to act and the responsibility to know are adapted to the ‘exercise leverage’ context. In exercising leverage, being both a choice and a business decision, the core company is again responsible for behaving diligently against the ‘reasonable person’ standard of negligence jurisprudence. From the diligence factor follows a responsibility to gather information, which in this new context (exercising leverage) requires the core company to understand not only the human rights abuse (which at this stage is well known) but also its root causes and effective solutions. Failure to gather this level of knowledge might result in wrongful conduct as the company discharges its responsibility to exercise leverage in improper ways or not at all. So the knowledge requirement gets ‘updated’ again, and limited again by the purpose of this sub-responsibility and the reasonable person standard.

Regarding the responsibility to act, that is, to exercise leverage, the standard is again the conduct of the ‘reasonable person’. As mentioned before, it might be reasonable to exercise no leverage for various reasons: for example, other actors can intervene more effectively; the company has no real leverage; the risk or size of harm is small compared with the costs of intervention and so on. Or because the solution to the root causes eludes resolution for the time being, exercising leverage might mean no more than working collaboratively with stakeholders and exploring potential solutions. What constitutes reasonable behaviour in complex or sensitive contexts might be very unclear. This reality places a conceptual limit on the resources a core company should expend on DD at this stage; minimal steps to influence affiliates might suffice to fulfil the ‘reasonable person’ standard.[163] The burden then rests heavily on the CSR community to identify effective, practical solutions that would raise the threshold of reasonability.[164] Until then, it might not be resource intensive for a duly diligent core company to discharge its RtR. But this observation should not provide false comfort to companies predisposed to passivity. Indeed, the CSR community has for the last decade made significant progress in understanding how companies can exercise leverage in cost efficient and effective ways without unintended effects. The ‘reasonable person’ standard is inherently able and indeed meant to track this progress; the thresholds of reasonability will rise accordingly.

4 The Nature of the RtR

The nature of a core company’s RtR concerning its more autonomous affiliates’ operations is clear by now: fundamentally it is a duty of care in negligence, a responsibility to act like the ‘reasonable person’. Its foundation is based on the marriage of the social norm with its raw force and negligence jurisprudence with its refined conceptual apparatus and authoritativeness. Or to put this differently, the social norm invites us not only to rely on its raw force but also gives us a golden opportunity to argue in the shadow of law, carefully enlarge exceptions in jurisprudence and reduce the thresholds of key factors in an attempt to legitimise the RtR.

Basically, the RtR thus conceived for core companies is a duty of care that is somewhere between no responsibility for affiliates based on the separation of entities norm and overreaching charges of automatic responsibility based on the social norm. The challenge is to strike a balance and persuasively argue for the imposition of a duty of care that neither overrides in practice the principle of legal separation of entities nor ends up being applied so narrowly that it ignores the position of the core company at the top of the business group. This article presents an argument supporting the RtR that resembles a makeshift structure which occupies an uneasy position in the vicinity of the separation of entities (limited liability) norm. The structure is reinforced precariously in a multitude of places; it is not based on a solid marble pillar. This makeshift structure of the duty of care remains permanently vulnerable to attacks from the separation of entities norm, with which it has an uneasy existence. If we define responsibilities too broadly, the burden becomes too heavy and the construction crumbles. Therefore, it is crucial that the burden of due diligence that the core company must carry is carefully chosen. It surely cannot carry the burden of asking the core company to ensure that all abuses are prevented or mitigated. It is a ‘reasonable care’ burden, not a strict liability one. It is based on ‘own fault’; that fault has its roots in negligence jurisprudence and is being actualised in the CSR context under the force of the social norm.

Should overambitious demands be placed at any of the key points, the whole conceptual structure will collapse. The result is a return to square one, where the RtR of core companies regarding autonomous affiliate conduct relies solely on the social norm. It follows that careful limitations of this responsibility are required not simply for the pragmatic reason of selling the case to the business world more easily, but in order not to stretch the internal logic of the scheme. It should not be overlooked that the negligence blocks on which it is built are exceptions in law and are to be interpreted narrowly. At the same time, reasonability enters the RtR at numerous key points bearing both on the responsibility to gather information and on the responsibility to act on it. Both the status of exceptions on which the responsibility to act builds and the reasonability factor are conceptual limiting factors on the RtR and DD.

Regarding limitations on the scope of RtR and the DD efforts the core company should undertake, three observations are in order. First, the exceptions in jurisprudence and the factors used in a scheme of attributing responsibility for third party misconduct both need to be relaxed with an eye to the international CSR context. This can be properly done in a focused mandate such as Ruggie’s.[165] Ruggie could incorporate negligence jurisprudence while fine-tuning the legal thresholds to meet the specific realities of the evolving business and human rights agenda, and the requirements and vision of his own mandate. This would still involve the same honesty and strategic thinking he has displayed so far in his mandate.

Second, the responsibility to collect information and form knowledge comes not overwhelmingly all at once, but in degrees at various stages: initially when the responsibility to act is established at a principled level; then when the business choice between staying and leaving in face of affiliate misconduct is made; and finally when it comes to exercising leverage. More specific information will be required accordingly. This may allow a realistic graduation of effort with more detailed information being required after the core company makes the key decisions to grant autonomy to an affiliate, and the choice between staying and leaving. This emphasis on own decisions is important because the responsibility to act will thus be based on a core company’s own choices, that is, on own actions (commissions) rather than inactions (omissions). The principle of ‘do no harm’ will be applicable to the core company’s own decisions making the core company’s RtR more consonant with the rest of Ruggie’s RtR.

Third, at the beginning of this article the suggestion was made that the social norm and jurisprudence could be used together to strengthen the foundation of the RtR. Their interaction is important. Indeed, there are some reinforcing loops where the social norm weighs in heavily and specifically: for example, when asking core companies to take responsibility for the indirect, remote impacts of their own decisions;[166] when enlarging the exceptions laid down in negligence law; when reducing the thresholds of factors relevant in a scheme for attributing responsibility, such as complicity and negligence schemes; and when defining reasonability in knowledge and conduct in light of present days realities. Arguments relying on the social norm need to tread carefully and not lose sight of the scope of RtR issue: imposing too heavy a weight causes the conceptual structure to crumble, leaving the social norm to carry the whole burden. But if the contrary happens, and companies, lawmakers and remedy providers begin to institutionalise a duty of care drawing on the social norm and evolving in shadow of law, the social norm is only reinforced and the standard of reasonableness gets clarified and modified upwards in time in a reinforcing loop.

Indeed, due to its many joints and reinforcement loops, the aforementioned makeshift structure can be stealthily strengthened, including through targeted legal interventions. Negligence jurisprudence has delivered, in the ‘reasonable person’ concept and in the multitude of exceptions generating a positive duty to act, valuable entry points for evolving business practices and for shifting societal perceptions of fairness. Therefore companies ignore at their own peril both the raw force of the social norm and the makeshift structure being quietly erected.

The last observation is about possible follow up of Ruggie’s RtR in the regulatory sphere. On the one hand, Ruggie has consistently specified that states and regulations have a key role to play in narrowing governance gaps which some companies exploit to infringe human rights. On the other hand, I submit that the current conceptualisation of the RtR is consistent with, and will be followed up by, only some types of policy and regulatory intervention. This is due to Ruggie’s treating the separation norm as an issue relevant to remedy in Pillar Three and furthermore as a legal issue relevant to legal remedies only. Deliberately, Ruggie attempts to shield the RtR in Pillar Two from the conflict of norms. This is neither desirable nor truly necessary as argued in this section.

The implications for the practical institutionalisation of the RtR following the conclusion of the SRSG mandate are powerful: Ruggie’s RtR will be institutionalised only when the clash of norms does not arise. The following situations qualify: voluntary application of the RtR by the core company; soft law interventions; contractual conditions imposed by financial institutions such as the IFC, export credit agencies, banks, public procurement bodies and so on; and transparency regulations asking core companies to report on the human rights risks and abuses taking place in affiliate operations. What all these institutionalisation options have in common is voluntariness. This is obvious in the cases of a core company’s voluntary adoption of the RtR and soft law. Contractual clauses are legal between the parties, but entering the contract remains voluntary; contracts are indeed legally binding but only after opting in. Transparency regulations require companies to disclose information, but do not in themselves ask companies to prevent or mitigate abuses; that latter part remains voluntary and dependent on social pressure building up to compel companies to act. The commonality is clear — as long as the core company can opt out of assuming responsibility for compensating victims of abuse and/or for changing affiliates conduct, the conflict of norms does not arise. Ruggie’s RtR with its DD emphasis can suffice despite its weak foundations and insufficiently defined scope.

The other side of the coin is that other attempts to institutionalise the RtR will be impossible or very difficult to achieve because they will run straight into the conflict of norms that Ruggie chose not to address. The following options are relevant here: regulations that make the core company liable to compensate victims of abuse, such as tort laws; regulations requiring the core company to prevent abuses or their reoccurrence through exercising leverage on affiliates, for example through procedures and managerial systems;[167] and regulations in company law requiring directors of the core company to exercise their managerial duty of care owed to the company in a way that spots and addresses risks of abuse in affiliate operations.[168] It is clear that asking core companies to compensate victims requires a principled explanation of why the limited liability principles should be waived.

Not only substantive regulations, but also procedural regulations, remain vulnerable. It seems highly probable that any regulation asking the core company to expend significant amounts of resources to influence affiliates will run into legal separation objections. Should these resources spent on DD efforts be insignificant, the problem could be overlooked as a trivial matter in practice, but in the vast majority of cases resources — in terms of funds, time and missed opportunities — cannot be treated as insignificant. The imposition of such expenses, when not insignificant, is also likely to exert pressure even on disclosure regulations because reporting on affiliate impacts will be deemed costly[169] and unjustified in the absence of a normative responsibility for affiliate operations; these pressures will be exerted either at the law-making or enforcement stages and severely curtail the effectiveness of less interventionist regulatory strategies. Indeed, governments can adopt procedural regulations to create a culture of diligence[170] but their scope is questionable. It is the conflict of norms that drains legitimacy from such regulations by raising the spectre of open-endedness. An RtR presented as a duty of care as informed by negligence jurisprudence contains limiting concepts and thus better addresses charges of open-endedness. In short, it gives a fairer chance to such regulations to be adopted and function properly afterwards.

Furthermore, leaving regulations aside, the current concept of RtR is not helping advocates of CSR to persuasively use jurisprudence and properly capture its normative force. Neither are CSR academics given a sufficiently clear intellectual foundation nor guidance in how to navigate the treacherous waters around the conflict of norms and steer clear of artificially polarised treatments of CSR based on denying the separation norm and obscuring profitability considerations. The battle for making a persuasive case for the RtR requires a principles-based, conceptual and intellectually compelling framework, as Ruggie himself rightly noted.

V CONCLUSIONS

Ruggie has proposed a properly broad RtR that covers core companies in business groups and networks. Such a company’s responsibility arises when its own decisions have contributed to harm in affiliates’ operations and also when the company is merely associated to affiliates infringing human rights. This article has dealt with the latter situation: none of the core company’s business decisions contributes to harm though either direct or indirect impacts on affiliates’ harmful operations. The question is whether the core company has a responsibility to act in such a situation, or whether it can legitimately remain passive. This is a foundational question. Ruggie’s compelling elaborations of DD hinge entirely on a responsibility to act being established in this particular situation. Ruggie’s reliance on the raw force of the social norm is well justified but insufficient, while references to soft laws and non-legal complicity notions are not helpful upon closer examination.

Ruggie has not acknowledged that his RtR and the social norm on which it is based are in practice on a collision course with the separation of entities (limited liability) norm. This will confine the institutionalisation of his RtR to corporate voluntarism in various forms: a voluntary CSR commitment, contractual agreements, soft law interventions, and transparency regulations. Other types of regulations are not facilitated by the current elaboration of the RtR, particularly procedural regulations asking the core company to influence affiliates to address (risks of) abuses rather than simply issue reports, not to mention substantive regulations holding the core company accountable for repairing affiliate abuses. Beyond the regulatory sphere, advocacy and academic treatments are not facilitated in the battle of ideas in many different arenas where CSR debates take place and leaves CSR proponents open to charges of illegitimacy and open-endedness against the RtR.

To address this weakness, this article tries to reinforce the RtR by combining the raw power of the social norm with jurisprudence. Negligence law has recognised a number of exceptions to the principle of no responsibility for third party abuses. The normative force behind these exceptions could be put to good use in support of the RtR. Drawing on the social norm, careful extensions of legal exceptions can be pursued to establish the legitimate existence of the RtR. Thus a responsibility to act would fall on a core company that set in motion a chain of events (prior conduct), possibly following non-diligent outsourcing to irresponsible contractors (negligent selection) or at least being in a relationship with the irresponsible affiliate (special relationship), conduct that created a situation of peril (situation of peril), even more so if there is an inherently dangerous activity (peculiar risk), with risks that were foreseeable (foreseeability) and affecting a vulnerable population (vulnerability).

The result is an RtR resembling a duty of care in negligence law. It can support the two sub-responsibilities that make Ruggie’s DD: responsibility to gather information and form knowledge, and responsibility to act on that knowledge. The conceptual structure thus erected occupies nevertheless a precarious position in the vicinity of the separation of entities norm. This requires increased attention to drawing realistic limitations on what core companies are asked to undertake in terms of DD; otherwise, the burden will make this makeshift structure crumble. The conceptual limitations on the RtR are drawn from the law of negligence, particularly legal exceptions that give rise to positive duties to act and the ‘reasonable person’ standard.

Ruggie is invited to adapt jurisprudential insights to the international CSR context by carefully enlarging the applicability of such exceptions and lowering some legal thresholds used in jurisprudence to affix liability for third party misconduct. The expectations behind the social norm invite him to do that. Although Ruggie has elaborated with increasing insight on the key DD steps companies should take and what factors a company should pay attention to, the triad of legitimate existence of the RtR, content in the shape of DD steps, and limiting concepts for responsibility (scope) might be needed for a fully fledged treatment of the core company’s responsibilities. So far the content of DD has received overwhelming attention from the SRSG team. Ruggie cannot rest content with the current conceptualisation of the RtR and concentrate pragmatically on the operationalisation of the responsibility as required by his second mandate (2008–11). That would be premature and unhelpful. It would be premature because the concept of responsibility to respect has a hole in it that undermines its legitimacy, and unhelpful because it deprives him of accessing a body of jurisprudence in which substantive wisdom has accumulated through decades of judicial practice and academic commentary.

If Ruggie reconsidered the need for a principled treatment of RtR foundations and in the process redefined his relationship with jurisprudence,[171] the discussion could then naturally move to how to make this duty of care consequential in practice or, in other words, look at ways of institutionalising it. Here Ruggie could elegantly pass the baton to his successor after correctly identifying the distinct situation of core companies, the special nature of the responsibility when a core company’s own decision did not contribute to the harm, the relationship of the RtR with key institutional elements such as the separation of entities norm with the economic, legal and social considerations supporting it, the complex balancing act required to successfully institutionalise this duty given conflicting norms and the absence of silver bullets, and the necessity to continue accounting simultaneously for legal and non-legal forces converging on corporate DD in a global governance context.


[1] Scott Jerbi, ‘Business and Human Rights at the UN: What Might Happen Next?’ (2009) 31 Human Rights Quarterly 299.

[2] Human Rights Council, Mandate of the Special Representative of the Secretary-General on the Issue of Human Rights and Transnational Corporations and Other Business Enterprises, 8th sess, 28th mtg, UN Doc A/HRC/RES/8/7 (18 June 2008).

[3] A cursory reading of his reports and the topics addressed through multi-stakeholder meetings and special expert papers gives a flavour of the many and diverse aspects relevant to the business and human rights discussion. See the complete list of documents prepared by or submitted to John Ruggie: Business and Human Rights Resource Centre, UN Secretary-General’s Special Representative on Business and Human Rights <http://www.business-humanrights.org/SpecialRepPortal/Home> .

[4] John Ruggie, Protect, Respect and Remedy: A Framework for Business and Human RightsReport of the Special Representative of the Secretary-General on the Issue of Human Rights and Transnational Corporations and Other Business Enterprises, 8th sess, Agenda Item 3, UN Doc A/HRC/8/5 (7 April 2008) paras 51–81 (‘Protect, Respect and Remedy’).

[5] ‘Core company’ covers entities such as parent companies, buyer companies, and participants in joint ventures wielding significant influence over their partners.

[6] The terms ‘affiliates’ and ‘partners’ will be used throughout this article to refer to entities such as subsidiaries, contractors, suppliers, joint venture partners and so on, which are involved in a core company’s operations and might be under the latter’s influence or even control.

[7] Richard Locke, Fei Qin and Alberto Brause, ‘Does Monitoring Improve Labor Standards?: Lessons from Nike’ (Working Paper No 24, Corporate Social Responsibility Initiative, Kennedy School of Government, Harvard University, July 2006).

[8] Human Rights Watch, ‘The Price of Oil — Corporate Responsibility and Human Rights Violations in Nigeria’s Oil Producing Communities’ (Report, January 1999). Actually, Shell was subjected to litigation in the US with plaintiffs alleging acts of assistance rather than mere corporate silence and inaction. The allegations were that the companies provided monetary and logistical support to the Nigerian police, bribed witnesses to produce false testimonies, and colluded with the Nigerian government to bring about the arrest and execution of the Ogoni Nine. The case was settled in 2009: Center for Constitutional Rights, Wiwa et al v Royal Dutch Petroleum et al <http://ccrjustice.org/ourcases/current-cases/wiwa-v.-royal-dutch-petroleum> .

[9] Ruggie noted that ‘[t]he issue of business and human rights burst into global public consciousness in the 1990s. Some of the early cases have acquired iconic status: Shell accused of complicity for standing by silently as the Nigerian military government executed a leader of community groups demonstrating against the company’s environmental degradation of the Delta region; BP accused of being responsible for alleged acts of murder, disappearances, torture, rape, and forced displacement of communities by a Colombian army brigade protecting its installation; allegations of sweatshop conditions and child labor in Nike’s Indonesian, and the GAP’s Salvadorian, suppliers.’: John Ruggie, ‘Next Steps in Business and Human Rights’ (Speech delivered at the Royal Institute of International Affairs, Chatham House, London, 22 May 2008) <www.reports-and-materials.org/Ruggie-speech-Chatham-House-22-May-2008.pdf>.

[10] Sithole v Thor Chemicals Holdings Ltd [1999] All ER (D) 102; Connelly v RTZ plc [1997] UKHL 30; [1998] AC 854; Lubbe v Cape plc [2000] UKHL 41; [2000] 4 All ER 268.

[11] Richard Meeran, ‘Corporations, Human Rights and Transnational Litigation’ (Lecture delivered at the Castan Centre for Human Rights Law, Monash University Law Chambers, Melbourne, 29 January 2003) <www.law.monash.edu.au/castancentre/events/2003/meeranpaper.html>.

[12] John Ruggie, Business and Human Rights: Towards Operationalizing the ‘Protect, Respect and Remedy’ Framework — Report of the Special Representative of the Secretary-General on the Issue of Human Rights and Transnational Corporations and other Business Enterprises, 11th sess, Agenda Item 3, UN Doc A/HRC/11/13 (22 April 2009) para 56 (‘Towards Operationalizing’).

[13] Radu Mares, The Dynamics of Corporate Social Responsibilities (Martinus Nijhoff Publishers, 2008).

[14] Protect, Respect and Remedy, UN Doc A/HRC/8/5, para 24.

[15] Towards Operationalizing, UN Doc A/HRC/11/13, para 75.

[16] Protect, Respect and Remedy, UN Doc A/HRC/8/5, para 24.

[17] Towards Operationalizing, UN Doc A/HRC/11/13, paras 46–7.

[18] Commission on Human Rights, Human Rights and Transnational Corporations and Other Business Enterprises, CHR Res 2005/69, 61st sess, UN Doc E/CN4/RES/2005/69 (20 April 2005).

[19] John Ruggie, Promotion and Protection of Human Rights: Interim Report of the Special Representative of the Secretary-General on the Issue of Human Rights and Transnational Corporations and Other Business Enterprises, 62nd sess, Agenda Item 2, UN Doc E/CN.4/2006/97 (22 February 2006) para 11.

[20] John Gerard Ruggie, ‘Business and Human Rights: The Evolving International Agenda’ (2007) 101 American Journal of International Law 819.

[21] Protect, Respect and Remedy, UN Doc A/HRC/8/5, para 13.

[22] An SRSG report writes: ‘A recent study conducted by the Office of the High Commissioner for Human Rights (OHCHR) for the Special Representative, which maps allegations against companies, documents that 41 per cent of the 320 cases (from all regions and sectors) in the sample alleged indirect forms of company involvement in various human rights abuses’: John Ruggie, Clarifying the Concepts of ‘Sphere of influence’ and ‘Complicity’: Report of the Special Representative of the Secretary-General on the Issue of Human Rights and Transnational Corporations and Other Business Enterprises, 8th sess, Agenda Item 3, UN Doc A/HRC/8/16 (15 May 2008) para 29 (‘Clarifying the Concepts’).

[23] Email from John Ruggie to Radu Mares, 26 August 2010.

[24] This can be seen clearly from the cautionary tone taken by some business organisations, a message that seems to boil down to a troublesome ‘hands off the core companies!’. ‘We also see a potential danger in the focus on multinational companies and foreign investment in this section, which reduces the attention on the vast majority of enterprises in the world which operate at the local and national level. If the goal is to reach down into the global supply chain, a large part of the focus should be on the suppliers and domestic companies themselves and the framework conditions in which they operate’: International Organisation of Employers (‘IOE’), International Chamber of Commerce (‘ICC’) and Business and Industry Advisory Committee to the OECD (‘BIAC’), ‘Joint Initial Views to the Third Report of the Special Representative of the UN Secretary-General on Business and Human Rights’ (Report, May 2008) 2.

[25] Clarifying the Concepts, UN Doc A/HRC/8/16, para 12.

[26] John G Ruggie, ‘Remarks by John G Ruggie’ (Speech Delivered at the Forum on Corporate Social Responsibility, Bamberg, Germany, 14 June 2006) <www.reports-and-materials.org/Ruggie-remarks-to-Fair-Labor-Association-and-German-Network-of-Business-Ethics-14-June-2006.pdf>.

[27] Clarifying the Concepts, UN Doc A/HRC/8/16, para 17.

[28] Protect, Respect and Remedy, UN Doc A/HRC/8/5, para 69.

[29] Mandate of the Special Representative of The Secretary-General on Human Rights and Transnational Corporations and Other Business Enterprises, ‘The Corporate Responsibility to Respect Human Rights in Supply Chains’ (Discussion Paper, 10th OECD Roundtable on Corporate Responsibility, 30 June 2010) (emphasis added).

[30] Email from John Ruggie to Radu Mares, 26 August 2010.

[31] Protect, Respect and Remedy, UN Doc A/HRC/8/5, para 69.

[32] Clarifying the Concepts, UN Doc A/HRC/8/16, para 39.

[33] John Ruggie, Corporate Responsibility: Comment on FIDH Position Paper by John Ruggie, UN Special Representative on Transnational and Human Rights (20 March 2006) <http://www.fidh.org/Comment-on-FIDH-Position-Paper-by-John-Ruggie-UN> .

[34] Protect, Respect and Remedy, UN Doc A/HRC/8/5, para 24.

[35] The Norms spoke of a corporate ‘obligation to promote, secure the fulfilment of, respect, ensure respect of and protect human rights’: Commission on Human Rights, Sub-Commission on the Promotion and Protection of Human Rights, Commentary on the Norms on the Responsibilities of Transnational Corporations and Other Business Enterprises with Regard to Human Rights, 55th sess, Agenda Item 4, UN Doc E/CN.4/Sub.2/2003/38/Rev.2 (26 August 2003) para 1 (‘Commentary on the Norms’).

[36] Protect, Respect and Remedy, UN Doc A/HRC/8/5, para 56.

[37] Ibid paras 60–4.

[38] Towards Operationalizing, UN Doc A/HRC/11/13, para 71 (emphasis added).

[39] Protect, Respect and Remedy, UN Doc A/HRC/8/5, paras 60–4. Ruggie further considers that grievance mechanisms also perform a key monitoring function: ‘they serve as early warning systems, providing companies with ongoing information about their current or potential human rights impacts from those impacted’: Human Rights Council, Business and Human Rights: Further Steps Toward the Operationalization of the ‘Protect, Respect and Remedy’ FrameworkReport of the Special Representative of the Secretary-General on the Issue of Human Rights and Transnational Corporations and Other Business Enterprises, John Ruggie, 14th sess, Agenda Item 3, UN Doc A/HRC/14/27 (9 April 2010) para 92 (‘Further Steps’).

[40] Protect, Respect and Remedy, UN Doc A/HRC/8/5, paras 60–4.

[41] Andrew Clapham and Scott Jerbi, ‘Categories of Corporate Complicity in Human Rights Abuses’ (2001) 24 Hastings International and Comparative Law Review 339. The ICJ conducted a two-year project finalised in 2008 with a 3-volume Final Report: ICJ, ‘Report of the International Commission of Jurists Expert Legal Panel on Corporate Complicity in International Crimes’ (Final Report, 2008).

[42] Protect, Respect and Remedy, UN Doc A/HRC/8/5, para 73.

[43] Alien Tort Claims Act, 28 USC §§ 1350, 1789.

[44] Protect, Respect and Remedy, UN Doc A/HRC/8/5, para 74. Subsequently, the US courts controversially narrowed even further the legal definition of complicity by raising the threshold of mens rea. Thus knowledge was deemed insufficient: the accomplice must act ‘with the purpose to assist the government’s violations of customary international law’: David Glovin, Talisman Court Upholds Sudan Genocide Suit Dismissal (2 October 2009) Bloomberg <www.bloomberg.com/apps/news?pid=20601082&sid=av47UyrplgEQ>.

[45] Protect, Respect and Remedy, UN Doc A/HRC/8/5, para 73. See also Clarifying the Concepts, UN Doc A/HRC/8/16, para 27.

[46] ‘A company is complicit in human rights abuses if it authorises, tolerates, or knowingly ignores human rights abuses committed by an entity associated with it, or if the company knowingly provides practical assistance or encouragement that has a substantial effect on the perpetration of human rights abuse’: United Nations Global Compact Office and the Office of the United Nations High Commissioner for Human Rights, ‘Embedding Human Rights into Business Practice’ (Report, 2004) 19.

[47] ICJ, ‘Corporate Complicity and Legal Accountability: Facing the Facts and Charting a Legal Path — Report of the International Commission of Jurists Expert Legal Panel on Corporate Complicity in International Crimes’ (Report, Vol 1, 2008) 3 (emphasis added).

[48] Clapham and Jerbi, above n 41.

[49] United Nations Global Compact, Global Compact Principle Two (10 January 2010) <http://www.unglobalcompact.org/AboutTheGC/TheTenPrinciples/Principle2.html> .

[50] International Organization for Standardization, ‘Guidance on Social Responsibility ISO 26000’ (Report, 2010) [6.3.5.1].

[51] Clarifying the Concepts, UN Doc A/HRC/8/16, para 58.

[52] Protect, Respect and Remedy, UN Doc A/HRC/8/5, para 78.

[53] Clarifying the Concepts, UN Doc A/HRC/8/16, para 41. See also at para 70; Protect, Respect and Remedy, UN Doc A/HRC/8/5, para 81.

[54] Clarifying the Concepts, UN Doc A/HRC/8/16, Summary.

[55] Email from John Ruggie to Radu Mares, 26 August 2010.

[56] Similarly, the ICJ uses ‘companies’, ‘businesses’ and ‘corporations’ interchangeably: ‘Although the title of the Panel’s report uses the phrase “corporate complicity”, throughout its inquiry the Panel has considered all business entities irrespective of structure or composition, of whether they are large or small, multinational, transnational or national, state or privately owned. The Panel’s analysis and findings are intended to apply across the board to all business entities and throughout its report the Panel uses the terms company and business interchangeably in order to capture the extent of its inquiry’: ICJ, ‘Facing the Facts’, above n 47, 4.

[57] For example, in the case of supply chains Ruggie took note of ‘brand-induced problems’, such as flexible production, fast turnaround, surge orders, changed orders and so on: Ruggie, ‘Remarks’, above n 26.

[58] For examples of both responsibilities, see Mandate of the Special Representative of The Secretary-General on Human Rights and Transnational Corporations and Other Business Enterprises, above n 29.

[59] See Part II(C) above.

[60] Protect, Respect and Remedy, UN Doc A/HRC/8/5, para 24.

[61] See above n 17.

[62] Organisation for Economic Co-Operation and Development, OECD Guidelines for Multinational Enterprises (OECD Publishing, revised ed, 2008) 14 (‘OECD Guidelines’).

[63] International Labour Organization, Tripartite Declaration of Principles Concerning Multinational Enterprises and Social Policy (International Labour Office, 4th ed, 2006) 3 (‘ILO Tripartite Declaration’).

[64] Draft UN Code of Conduct on Transnational Corporations, 23 ILM 626 (1984).

[65] Commentary on the Norms, UN Doc E/CN.4/Sub.2/2003/38/Rev.2, para 20.

[66] ILO Tripartite Declaration, above n 63, 2; OECD Guidelines, above n 62, 12; ibid; Draft UN Code of Conduct on Transnational Corporations, 23 ILM 626 (1984) [1].

[67] Being voluntary by definition and intent, soft law instruments are addressed to the entire MNE. Legal separation is not an issue to be discussed because it is entirely up to companies to comply or not and thus to disregard the legal advantage of legal separation or not, if they so wish.

[68] OECD Guidelines, above n 62, 12 (emphasis added).

[69] ILO Tripartite Declaration, above n 63, 2.

[70] Organisation for Economic Co-Operation and Development, ‘OECD Guidelines for Multinational Enterprises: Text, Commentary and Clarifications’ (Report No DAFFE/IME/WPG(2000)15/FINAL, 31 October 2001) 9 (‘OECD Clarifications’). As the OECD explains, ‘[t]he clarifications provide interpretations of how certain provisions of the Guidelines should be understood as a result of deliberations by the Committee on International Investment and Multinational Enterprises prior to the 2000 Review. These clarifications, however, do not modify the authoritative texts, which are found in the verbatim language of the Guidelines and the relevant decisions of the Committee on International Investment and Multinational Enterprises’: at 2.

[71] Ibid 10 (emphasis added).

[72] Ibid (emphasis added).

[73] Ruggie takes note of fault-based liability and no-fault liability based on public policy in Further Steps, UN Doc A/HRC/14/27, para 106.

[74] OECD Guidelines, above n 62, 15.

[75] OECD Clarifications, above n 70, 10.

[76] Ibid 9 (emphasis added).

[77] See, eg, Karen Vandekerckhove, Piercing the Corporate Veil: A Transnational Approach (Kluwer Law International, 2007).

[78] See, eg, Sithole v Thor Chemicals Holdings Ltd [1999] All ER (D) 102; Connelly v RTZ plc [1997] UKHL 30; [1998] AC 854; Lubbe v Cape plc [2000] UKHL 41; [2000] 4 All ER 268.

[79] OECD Guidelines, above n 62, 12.

[80] OECD Clarifications, above n 70, 9.

[81] Ibid.

[82] Ibid.

[83] OECD Guidelines, above n 62, 14.

[84] See generally Radu Mares, ‘The Limits of Supply Chain Responsibility — A Critical Analysis of CSR Instruments’ (2010) 79 Nordic Journal of International Law 193, 199–203.

[85] OECD Committee on International Investment and Multinational Enterprises (‘CIME’), Scope of the Guidelines and the Investment Nexus: Statement by the Committee (April 2003) OECD <http://www.oecd.org/document/3/0,3343,en_2649_34889_37356074_1_1_1_1,00.html> .

[86] Ruggie wrote that the ‘investment nexus’ approach ‘reflects the link between the [OECD] Guidelines and the OECD Declaration on International Investment; however, it significantly limits NCPs’ utility as a grievance mechanism for rapidly expanding segments of global value chains’: Further Steps, UN Doc A/HRC/14/27, para 99.

[87] ICJ, ‘Corporate Complicity and Legal Accountability: Civil Remedies — Report of the International Commission of Jurists Expert Legal Panel on Corporate Complicity in International Crimes’ (Report, Vol 3, 2008) 47.

[88] OECD Clarifications, above n 70, 9.

[89] OECD, ‘Terms of Reference for an Update of the OECD Guidelines for Multinational Enterprises’ (Terms of Reference, 4 May 2010) 3–4 <www.oecd.org/dataoecd/61/41/45124171.pdf>.

[90] OECD Clarifications, above n 70, 10.

[91] Towards Operationalizing, UN Doc A/HRC/11/13, paras 46–7.

[92] See text accompanying above n 71.

[93] Frank H Easterbrook and Daniel R Fischel, The Economic Structure of Corporate Law (Harvard University Press, 1991).

[94] For two recent treatments on the history of the legal separation principle, see especially Paddy Ireland, ‘Limited Liability, Shareholder Rights and the Problem of Corporate Irresponsibility’ (2010) 34 Cambridge Journal of Economics 837; Peter Muchlinski, ‘Limited Liability and Multinational Enterprises: A Case for Reform?’ (2010) 34 Cambridge Journal of Economics 915.

[95] This efficiency is obtained by permitting an efficient division of labour between different agents (employees, shareholders, lenders, suppliers), by reducing monitoring costs within the firm, by promoting a stock market in which shares reflect the value of firms and by allowing for efficient diversification and investors’ spreading risk: Phillip I Blumberg, Multinational Challenge to Corporation Law: The Search for a New Corporate Personality (Oxford University Press, 1993).

[96] Protect, Respect and Remedy, UN Doc A/HRC/8/5, para 13.

[97] Further Steps, UN Doc A/HRC/14/27, para 104.

[98] Ibid para 105.

[99] Ibid.

[100] Ibid para 106 (emphasis added).

[101] Towards Operationalizing, UN Doc A/HRC/11/13, para 48.

[102] OECD Clarifications, above n 70, 10.

[103] Clarifying the Concepts, UN Doc A/HRC/8/16, para 22 (emphasis added). The full quotation reads: ‘A company should ensure that it is not implicated in third party harm to rights through its relationships with such parties. This possibility can arise from a company’s business activities, including the provision or contracting of goods, services, and even non-business activities, such as lending equipment or vehicles. Therefore, a company needs to understand the track records of those entities with which it deals in order to assess whether it might contribute to or be associated with harm caused by entities with which it conducts, or is considering conducting business or other activities’.

[104] Protect, Respect and Remedy, UN Doc A/HRC/8/5, para 73 (emphasis added).

[105] Clarifying the Concepts, UN Doc A/HRC/8/16, para 30; Protect, Respect and Remedy, UN Doc A/HRC/8/5, para 74.

[106] Jonathan Clough, ‘Punishing the Parent: Corporate Criminal Complicity in Human Rights Abuses’ (2007–08) 33 Brooklyn Journal of International Law 899, 909.

[107] The ICJ, for example, refers to culpable conduct as being when ‘the company or its employees contribute to specific gross human rights abuses, whether through an act or failure to act, and whatever form of participation, assistance or encouragement the conduct takes’: ICJ, ‘Facing the Facts’, above n 47, 9.

[108] Information-gathering through monitoring, conducting HRIAs, and even employing grievance mechanisms is a key DD step for Ruggie. Should the core company fail to do that and remain ignorant, it will be deemed complicit in its partners’ abuses: ‘the relationship between complicity and due diligence is clear and compelling: companies can avoid complicity by employing the due diligence processes’: Protect, Respect and Remedy, UN Doc A/HRC/8/5, para 81.

[109] ‘Knowledge may be inferred from both direct and circumstantial facts’: Protect, Respect and Remedy, UN Doc A/HRC/8/5, para 79. See also Clarifying the Concepts, UN Doc A/HRC/8/16, para 43: ‘The knowledge requirement can be established through direct and indirect or circumstantial evidence. Therefore objective facts can be used to infer the subjective mental state of the accused, and constructive knowledge could be inferred even where the accused has not explicitly expressed that they had such knowledge or in fact denied they had knowledge’.

[110] See above n 41.

[111] Ibid. See also text accompanying above n 54.

[112] Extreme cases of complicity by omission exist, but they do not resemble the vast majority of CSR situations. For an example of complicity by omission, see Prosecutor v Anto Furundzija (Trial Judgement) (International Criminal Tribunal for the Former Yugoslavia, Trial Chamber II, Case No IT–95–17/1, 10 December 1998) [235]. See also Clarifying the Concepts, UN Doc A/HRC/8/16, para 40, where Ruggie points out that courts deemed conduct by omission as aiding and abetting when individuals were silently present at the scene of a crime and had some form of superior status. Reference is made to Prosecutor v Miroslav Kvocka (Trial Judgement) (International Criminal Tribunal for the Former Yugoslavia, Trial Chamber I, Case No IT–98–30/1, 2 November 2001) [257]–[261].

[113] We could talk of criminal enterprise where those associated for criminal purposes are held accountable for that association per se irrespective of being passive and ignorant regarding specific crimes. But this is different from complicity and most likely inapplicable to CSR in most cases given that core companies and affiliates associate for doing business only, a legitimate activity. Ruggie took note of the doctrines of criminal enterprise and superior liability in Clarifying the Concepts, UN Doc A/HRC/8/16, para 33.

[114] Protect, Respect and Remedy, UN Doc A/HRC/8/5, para 79.

[115] See above nn 49–51.

[116] The initial mandate on the SRSG (Commission on Human Rights, Human Rights and Transnational Corporations and Other Business Enterprises, UNCHR Res 69, 61st sess, 59th mtg, UN Doc E/CN.4/RES/2005/69 (20 April 2005)) actually asked him to clarify the concept of complicity in the CSR context. All his reports since have mentioned this concept.

[117] Towards Operationalizing, UN Doc A/HRC/11/13, para 76.

[118] Conduct by omission is where a core company’s own decisions did not assist the affiliate/perpetrator in any way, or put differently, they did not have direct or indirect impacts on the harm occurring: see above n 112.

[119] The term ‘scope’ as used herein refers to scope or extent of DD efforts, which must be conceptually limited to prevent charges of open-endedness. Ruggie uses ‘scope’ differently as scope or coverage of the RtR which is properly broad to include direct and indirect corporate impacts. He thus aims to impress that not only direct impacts should be considered by a company. ‘Scope [of the RtR] is defined by the actual and potential human rights impacts generated through a company’s own business activities and through its relationships with other parties, such as business partners, entities in its value chain, other non-State actors and State agents’: Further Steps, UN Doc A/HRC/14/27, para 58. In his 2009 Report, Ruggie refers to the scope of DD processes to explain the same idea of broad coverage: Towards Operationalizing, UN Doc A/HRC/11/13, para 50. However, commenting on the different extent of DD efforts, Ruggie only succinctly notes that ‘[t]he [DD] process inevitably will be inductive and fact-based ... How far or how deep this process must go will depend on circumstances’: Protect, Respect and Remedy, UN Doc A/HRC/8/5, para 57. This might be insufficient as along the way Ruggie loses the crucial importance of setting or at least discussing conceptual limitations on DD efforts.

[120] American Law Institute, Restatement (Second) of Torts (1977) § 315.

[121] Ibid § 409.

[122] Ibid § 315 (emphasis added).

[123] ‘The actor’s prior conduct, whether tortious or innocent, may have created a situation of peril to the other, as a result of which the actor is under a duty to act to prevent harm, as stated in §§ 321 and 322’: ibid § 314.

[124] Ibid § 316–20.

[125] Ibid § 409.

[126] Ibid § 302 (emphasis added).

[127] Ibid § 441(1): ‘An intervening force is one which actively operates in producing harm to another after the actor’s negligent act or omission has been committed’.

[128] Ibid § 302B (emphasis added).

[129] Ibid § 409.

[130] Ibid § 283. See also at § 289.

[131] Ibid § 290.

[132] Ibid.

[133] Ibid.

[134] Ibid.

[135] ICJ, ‘Facing the Facts’, above n 47, 19–20 (emphasis added).

[136] For a more detailed discussion of negligence jurisprudence insights in the CSR context, see Radu Mares, ‘Defining the Limits of Corporate Responsibilities against the Concept of Legal Positive Obligations’ (2009) 40 George Washington International Law Review 1157, 1178–97.

[137] John Ruggie, Report of the Special Representative of the Secretary-General on the Issue of Human Rights and Transnational Corporations and Other Business Enterprises: Addendum — Summary of Five Multi-stakeholder Consultations, 8th sess, Agenda Item 3, UN Doc A/HRC/8/5/Add.1 (23 April 2008) paras 49–53.

[138] Clarifying the Concepts, UN Doc A/HRC/8/16, para 52.

[139] Ibid paras 52–3.

[140] ‘[I]n the context of negligence, proving knowledge might require showing only that the company should have known that it was taking a foreseeable risk of contributing to an abuse as opposed to needing to prove that there was actual knowledge’: ibid para 53.

[141] See Clapham and Jerbi, above n 41.

[142] See the definition at above n 56.

[143] ICJ, ‘Facing the Facts’, above n 47, 6.

[144] Ibid 7.

[145] Mares, ‘Defining the Limits’, above n 136.

[146] James Anaya, Report of the Special Rapporteur on the Situation of Human Rights and Fundamental Freedoms of Indigenous People, 15th sess, Agenda Item 3, UN Doc A/HRC/15/37 (19 July 2010).

[147] ICJ, ‘Facing the Facts’, above n 47, 24.

[148] Ibid 25.

[149] Ibid.

[150] Ibid.

[151] Ibid.

[152] Protect, Respect and Remedy, UN Doc A/HRC/8/5, para 8.

[153] Ibid para 107.

[154] CSR-relevant initiatives ‘must cohere and generate an interactive dynamic of cumulative progress — which the framework is designed to help achieve’: Further Steps, UN Doc A/HRC/14/27, para 5.

[155] Clarifying the Concepts, UN Doc A/HRC/8/16, para 21.

[156] Now, if the company not only has total control over the affiliate but also exercises it, then it arguably takes all the decisions for it; the law will either deem the core company automatically liable under strict liability theories or will charge the core company with negligent exercise of control in order to deem its controlling decisions negligent. Although this scenario is exceptional in CSR, it is real and the law illustrates perfectly Ruggie’s prescription for the core company assuming responsibility for its own decisions.

[157] See the concerns of a well-known US law firm regarding the SRSG’s due diligence concept that observed that due diligence can become an expansive undertaking, although the commentary ran adrift afterwards: Martin Lipton and Kevin S Schwartz, ‘A United Nations Proposal Defining Corporate Social Responsibility for Human Rights’ (Memo, Watchell, Lipton, Rosen and Katz, 2008) <amlawdaily.typepad.com/amlawdaily/files/wachtell_lipton_memo_on_global_business_human_rights.pdf>.

[158] For example, benefiting from abuse, with such a benefit appearing morally repugnant. CSR literature has made this argument by references to beneficial complicity. Public policy may impose strict liability for both moral and utilitarian reasons given that the large, wealthy core company may be in the best position to absorb and manage risks generated by affiliate misconduct. Tort law sometimes lays down strict liability regimes by arguing in this way.

[159] ‘The reasonable person conceived of by the law of civil remedies does not represent the lowest common denominator, but instead is a responsible, careful actor, ‘a good member of society’. ... [A]s societal expectations develop and expand, so too will the expectations placed on the reasonable person by the law of civil remedies, and the requirements of careful conduct today, will always be higher than they were yesterday’: ICJ, ‘Civil Remedies’, above n 87, 16.

[160] Processing information in order to form knowledge can be treated as part of the sub-responsibility to act on information as opposed to leaving it lay idle. For reasons of simplicity, information and knowledge are treated here as part of the same sub-responsibility to not remain ignorant.

[161] Ruggie noted that ‘the [due diligence] process inevitably will be inductive and fact-based’, and that ‘how far or how deep this [due diligence] process must go will depend on circumstances’: Protect, Respect and Remedy, UN Doc A/HRC/8/5, para 57.

[162] See, eg, United Nations Global Compact Office and the Office of the United Nations High Commissioner for Human Rights, above n 46, 22.

[163] For example, companies disseminating their codes of conduct to affiliates or providing some training to affiliate managers.

[164] ‘The principles of human rights due diligence and its core elements should be internalised by all businesses, regardless of their nature or size. But the specific activities that companies must undertake to discharge this responsibility will vary in ways not yet fully understood’: Towards Operationalizing, UN Doc A/HRC/11/13, para 72. For difficulties raised by doing business in conflict zones, see Further Steps, UN Doc A/HRC/14/27, paras 44–5. For situations where international standards and national laws conflict see Further Steps, UN Doc A/HRC/14/27, para 68.

[165] Compared to assigning responsibility in a national context, difficulties get magnified with a MNE’s global operations; for example, causation becomes more indirect, proximity more remote, and public policy priorities are differently articulated when the costs and benefits of imposing liability on the parent company accrue to different jurisdictions and so on.

[166] Clarifying the Concepts, UN Doc A/HRC/8/16, para 17. See the full quote accompanying above n 27.

[167] Such as those laid out in the US Sentencing Guidelines: United States Sentencing Commission, ‘2010 Federal Sentencing Guidelines Manual and Supplement’ (2010) s 8B2.1(a), quoted in Further Steps, UN Doc A/HRC/14/27, para 42.

[168] Ruggie took note of the directors’ duty under company law and suggests incorporation of DD notions in there: Further Steps, UN Doc A/HRC/14/27, para 41.

[169] The OECD is mindful of such considerations: ‘Disclosure requirements are not expected to place unreasonable administrative or cost burdens on enterprises’: OECD Clarifications, above n 70, 7 [18].

[170] See Ruggie on rights-respecting corporate cultures in: Further Steps, UN Doc A/HRC/14/27, paras 33–43.

[171] Currently, jurisprudence informs remedies in Pillar Three and state obligations in Pillar One but not the RtR itself.


AustLII: Copyright Policy | Disclaimers | Privacy Policy | Feedback
URL: http://www.austlii.edu.au/au/journals/MonashULawRw/2010/26.html