The Legal Meaning of “Involvement”

This entry was originally published on the blog of our sister consulting firm, Enodo Rights. It is based on Chapter 4, “Involvement”, of Business and Human Rights as Law (LexisNexis 2019). The chapter draws on a discussion paper Enodo Rights and Debevoise & Plimpton wrote for the OECD Secretariat in 2017.


The Guiding Principles on Business and Human Rights are the authoritative standard on corporate “S” responsibility. They place corporate involvement with adverse human rights impacts at the heart of understanding the contours of a reasonable business’s human rights governance. Three terms—causecontribute, and directly linked—are critical parameters of the Guiding Principles framework: they define what a company’s relevant human rights impacts are and how the company is expected to address them. The three concepts are not simply relevant to a business once it has identified adverse human rights impacts to address. Rather, they are intrinsic to the very idea of whose rights and which rights the business should seek to identify in the first place. As the OHCHR noted in its Interpretive Guide to the Guiding Principles: “The purpose of assessing impact is to identify any adverse impact in which an enterprise might be involved.”

For such foundational concepts, however, the involvement terms remain distressingly nebulous. To the extent authoritative stakeholders have offered definitions, they have been — from the perspective of justiciability and the rule of law — vague, circular, and opaque in method. The approach risks weaving arbitrariness into corporate human rights responsibility. In this chapter, we seek to address that risk by applying a legal interpretive method to derive principled, precise, and practical definitions of the involvement terms (including omission). While the method is legal, we seek definitions that are true to the Guiding Principles as voluntary standards. We therefore expressly reject exclusively legal definitions where they are at odds with the text and overarching aims of the Guiding Principles.

We ground our discussion in a fact pattern involving modern slavery and the financial sector. The practical end of the definitions is to help shape reasonable due diligence and to guide a company on how to respond if it is involved with an adverse impact. We summarize our key findings below.


The case study involves a global toy company (ToyCo), its major lender (EuroBank), and a group of Nepalese victims of modern slavery in ToyCo’s Vietnamese facility. We propose a thought experiment based on a human rights arbitration brought by the Nepalese plaintiffs against EuroBank. The Guiding Principles are the substantive law of the dispute. Based on that framework, the workers advance three arguments to ground EuroBank’s liability:

(i) EuroBank caused or contributed to the claimants’ injuries by funding ToyCo’s expansion in Vietnam and is therefore responsible to provide remedy.

(ii) In the alternative, EuroBank caused or contributed to the claimants’ injuries by failing to conduct sufficient due diligence before funding ToyCo’s expansion in Vietnam and is therefore responsible to provide remedy.

(iii) In the further alternative, EuroBank was directly linked to the claimants’ injuries by its business relationships with ToyCo and RecruitCo and should exercise leverage over ToyCo to provide the claimants’ effective remedy in accordance with the Guiding Principles.


Drawing on this fact pattern, Chapter 4 conducts a detailed analysis of cause and contribute as they are used in the social sciences, natural sciences, and the law. (We consider the meaning of cause and contribute together because they are woven together in the Guiding Principles and in their ordinary usage.) Our overarching finding is that, in each of these disciplines, cause has come to be defined with reference to the impact on a likelihood of an event. That is, an act that increases the likelihood of a particular event — to a statistically relevant degree — may be deemed a cause of that event. To be a cause, the act should also be (i) sufficient to result in the event and (ii) part of a continuous chain to the event (i.e., unbroken by intervening acts).

Law has generally adopted the same definition to determine the factual cause of an injury. In addition, however, law considers foreseeability to determine whether a particular individual or entity should be held liable, as a matter of policy, for an injury to another. Foreseeability serves as fairness-based limitation on responsibility. That is: despite being the factual cause of an injury, an individual or business may not be legally liable for the injury if it was not foreseen or reasonably foreseeable. This element of causation is uniquely legal and expressly rejected by the Guiding Principles: “Even with the best policies and practices, a business enterprise may cause or contributeto an adverse human rights impact that it has not foreseen or been able to prevent.” (GP 22) We accordingly dismiss foreseeability as relevant to the definitions of cause and contribute.

Based on the elements of cause and contribute common to the natural sciences, social sciences, and the law, we propose the following definitions:

A business causes an adverse human rights impact when its activities (including omissions) materially increase the likelihood of the specific impact which occurred and would be sufficient, in and of themselves, to result in that impact.

A business contributes to an adverse human rights impact when its activities (including omissions) materially increase the likelihood of the specific impact which occurred even if they would not be sufficient, in and of themselves, to result in that impact.

The three-stage inquiry to assess whether a business is actually or potentially causing or contributing to an adverse impact would be:

  1. Is there an actual or potential adverse human rights impact?

  2. If so, do the company’s activities (including omissions) materially increase the likelihood of that impact?

  3. If so, would the company’s activities (including omissions) in and of themselves be sufficient to result in that impact?

Applying this analysis, the claimants’ argument that EuroBank caused or contributed to their injuries by funding ToyCo’s expansion fails at the second question. There was nothing in the lending that materially increased the risk of the claimants’ injuries, as ToyCo could reasonably have implemented far better hiring practices. ToyCo’s practices thus broke the chain of causation.


The claimants’ second allegation is that EuroBank’s due diligence failure caused or contributed to their injuries. Analyzing that claim turns on the meaning of omission. At a conceptual level, the complexity lies in the fact that inaction, to constitute omission, implicitly suggests that the risk originated other than from the company itself. To understand what constitutes a relevant omission under the Guiding Principles, we are therefore asking: When does the responsibility to respect include an obligation to protect?

A principled and precise understanding of omission under the Guiding Principles would need to explain in what circumstances a failure to mitigate risk can be deemed a contribution to risk. That is necessarily tied to reasonable expectations. In order for a failure to act to be an omission capable of causing injury, that failure to act must itself have violated reasonable expectations of behavior. In other words, there must have been a pre-existing duty to act.

Based on our survey of comparable legal constructs and the structure of the Guiding Principles, we conclude:

A company’s failure to act may be deemed an omission that causes or contributes to an adverse human rights impact if, and only if, the company failed to fulfill a duty (1) under law or (2) under a special relationship defined by contract, authority, control, or other legitimate expectations. Any special relationship must exist independently of the company’s responsibilities under the Guiding Principles.

Applying this concept to EuroBank, the claimants’ arguments fail because (as we stipulate) EuroBank complied with all relevant law and there was no special relationship created between EuroBank and its client’s workers simply by virtue of lending.


The third ground advanced by the claimants is that EuroBank is directly linked to their injuries and is therefore obliged to apply leverage to ensure ToyCo provides them remedy. Directly linked is rather more complex than the other involvement terms. We have found no identically phrased precedent in the social sciences or law to ground a definition of the term.

We therefore focus on the structure and objectives of the Guiding Principles to home in on a definition. These suggest that directly linked responsibility is independent of fault. The aim is to leverage as many avenues as possible to encourage responsible business practice and advance the pursuit of effective remedy for stakeholders. In this, the best analogue may be the legal concept of vicarious liability — a form of no-fault liability under which an entity may be held accountable for the acts of an employee or business partner irrespective of whether the entity itself caused the injury. Vicarious liability turns on two elements: business relationship and wrongdoing in the scope of that relationship.

Based on our analysis of the ordinary meaning of the term in the context of the Guiding Principles — with specific reference to the analogous concept of vicarious liability — we propose the following definition of directly linked:

A business is directly linked to an adverse human rights impact when it has a relationship for mutual private benefit with a state or non-state entity, and, in performing activities within the scope of that relationship, the state or non-state entity materially increases the likelihood of the impact which occurred.

The cornerstone of this definition is for mutual private benefit. The link underpinning a business’s responsibility to conduct due diligence and seek leverage to avoid or mitigate an adverse human rights impact, even when it has not contributed to it, is the benefit the business derives from the adverse impact. Directly then conditions the type of benefit provided and received through the value chain rather than the number of intermediaries through which it passes. For mutual … benefit is essential to avoid capturing “extremely loosely connected associations”, such as might extend, for instance, from infrastructure projects to all who rely on them. That is, for a business relationship to exist, there must be a mutual intention between the businesses to benefit one another’s operations, products, or services.

Lastly, private is essential to avoid capturing the activities of a state acting in a public capacity (as opposed to when it is conferring a private benefit). Thus, for instance, while a state has a relationship for mutual benefit with every business to the extent it collects taxes which it invests in infrastructure, education, and the rule of law, that relationship is in the exercise of the state’s public and general responsibilities; it would not ground a direct link. The mutual benefit would need to be from the state’s exercise of commercial powers or through the exercise of public authority to accord the business a private benefit not accorded to the public at large.

The three-stage inquiry for impact or risk assessment would be:

  1. Does the business have a relationship for mutual private benefit with the state or non-state entity?

  2. Does the benefit provided by the state or non-state entity retain consistent form as it is transmitted to the company’s products, operations, or services?

  3. When acting to provide the benefit that is the object of the relationship, did the state or non-state entity materially increase the likelihood of the adverse human rights impact which occurred or may occur?

Only if the answer to all three questions is “yes” can the business be said to be directly linked to the adverse impact and be expected to exercise or seek leverage to prevent or mitigate the impact to the extent possible.

Applying the test to this case suggests that EuroBank was directly linked to the claimants’ injuries through its relationship to ToyCo. First, the relationship with ToyCo was for clear mutual private benefit. ToyCo received funding to expand in Vietnam; EuroBank received the interest on its loans. Second, the type of private benefit received by EuroBank was monetary and retained that form. Third, the injury to the claimants occurred as ToyCo sought to build out its Vietnamese operations, the success of which was the very object of this lending facility.


Our hope with this chapter is not to advance definitive interpretations of the involvement terms. Rather, it is to suggest an objective and transparent method for courts, regulators, businesses, and stakeholders to assess corporate human rights responsibility in a way that aligns with the rule of law. Our findings do not align with various stakeholder proposals. But, as we explain in the chapter, involvement may be the clearest indicator of paradigm incompatibility between the Guiding Principles as voluntary corporate responsibility norms and their emerging role as bases of legal liability.