5 - Standards of Liability

Summary

A threshold question for designing liability rules is the degree of fault required to impose liability. This chapter begins with a discussion of the policy considerations that underlie the policy choice between strict liability and due diligence approaches to liability and an examination of the distributive implications of this choice. The chapter then describes the approaches to standards of liability found in the law of state responsibility and civil liability regimes, before examining the specific fault requirements that structure liability in the Antarctic, deep seabed and high seas regimes.

5.1 Introduction

A threshold question for designing liability rules in any legal system is the degree of fault required to impose liability. At the core of this question are both moral and distributive considerations in determining when a loss that is suffered by one person – or in the case of environmental harm, by the community as a whole – ought to be shifted to another, usually to the person who caused the harm. The moral dimensions concern the characterization and degree of blameworthy conduct that is required to justify shifting the loss. The distributive dimensions carry with them a range of policy implications concerning the relative utility of the activity posing a risk of harm as compared to the harm itself, the ability of parties involved to bear a particular loss and practical concerns respecting the efficient and effective implementation of loss allocation measures.

One result of the complex array of considerations at play in addressing the appropriate standard of liability is that any consideration of this question is necessarily influenced by the context of its application. Where attempts have been made to develop generalized rules of international law concerning the approach to liability for environmental harm, the result has been a degree of conceptual confusion and no shortage of debate.¹ As a consequence, the rules respecting environmental liability have not developed as a unitary body of law common to all activities, but rather on a regime by regime basis with different approaches to the standard and scope of liability that respond to the regulatory setting of the activity.

With this diversity in mind, this chapter identifies the policy considerations that underlie different approaches to the standard of liability before examining the rules currently in place within the various global commons contexts. The approach to which entities are the main focus of liability is important to the issue of the standard chosen, as states and operators perform distinct functions in relation to the risk that bears on the justification for requiring fault, which has influenced state practice in this area.² As both approaches are used in the global commons contexts examined here, this chapter considers the rules that have developed in relation to both state responsibility and where liability is channelled to operators. Finally, this chapter examines the specific rules governing the standard of liability in the global commons.

In discussing the approach to liability, most legal systems distinguish between two main forms of liability: negligence, or fault-based liability, on the one hand, and strict liability, on the other.³ Negligence regimes are defined as requiring a degree of fault, usually a breach of an identified standard of care, as well as a causal link between the activities undertaken by the subject of liability and the harm, in order to impose liability for environmental harm. The standard of care for negligence can be defined variably, but it is often identified as reasonably prudent or duly diligent behaviour, as evidenced by accepted standards of behaviour in the relevant area of activity. Extensive consideration is given in this chapter to the application of the due diligence standard to environmental harm prevention obligations on states. Strict liability, on the other hand, requires no proof of fault for a finding of liability in relation to harm, but does require causation. Strict liability may still allow certain defences or exceptions to the imposition of liability, such as acts of God, acts of war, necessity and third party or contributory negligence. Where there are no exceptions or very limited exceptions, the liability is often classified as being absolute in nature.⁴ Given the limited application of absolute liability in international law,⁵ this chapter focuses on the more binary distinction between fault and no-fault (strict) liability.

5.2 Policy Considerations Underlying Standards of Liability

The basic theory behind requiring fault as an element of attributing liability is an ethical or justice-based idea that a person who causes harm should only be compelled to compensate the person who suffers an associated loss where the person who causes the harm has acted wrongly in some fashion.⁶ Where the incident in question is purely accidental, there is no moral reason for loss shifting. The requirement for fault is not punitive, since the goal is not to make the defendant worse off than they were before the incident, but rather corrective in the sense that compensation is tied to the plaintiff’s loss.⁷ Wrongfulness, of course, lies at the heart of the law of state responsibility, but only in the thin sense of arising by virtue of a breach of an international obligation.⁸ However, the requirement for fault in a subjective sense is a function of the obligation in question.⁹

The difficulty with subjective fault requirements, such as negligence, is that, in the absence of fault, there is no liability, but the victim remains harmed through no fault of their own. Thus, in the absence of fault, the policy question that arises is who should bear the loss as between two potentially non-culpable actors. Creation of risk is most often raised as a basis for imposing liability without a requirement of proof of fault.¹⁰ As a consequence, activities with higher degrees of risk are often subjected to strict forms of liability in both international and domestic law.¹¹ The presence of risk underlies the law of strict liability in common law tort regimes,¹² as well as influencing liability in civil law jurisdictions.¹³ Risk has also been raised as a basis for imposing strict liability on states where they engage in or authorize hazardous or ‘ultra-hazardous’ activities.¹⁴

Subjecting nuclear power, marine transport of oil and hazardous substances and the movement of living modified organisms to strict liability regimes reflects the risk concerns associated with those activities.¹⁵ Risk in this context is a function of both the probability of harm and the severity of harm. Goldie further expands on the concept of risk by linking it to concerns respecting the unforeseeability of harm associated with certain activities, and related difficulties in determining acceptable standards of due diligence.¹⁶ Goldie was thinking specifically about the harms arising from new technologies such as nuclear power and outer space activities. In such instances, it may be impossible for operators to reduce risks to acceptable levels through the exercise of due care, but it may nevertheless be desirable for the activities to be pursued. Thus, for Goldie, strict liability has a facilitative function, insofar as it creates conditions (indemnification of those harmed) that allow for the undertaking of activities that might otherwise not be permitted. Moreover, in the event of harm from technologically advanced and complex activities, proving negligence imposes a high evidentiary burden on injured parties.

What is less clear is the degree of risk that is required to justify applying a standard of strict liability. Should, for example, no-fault liability be restricted to ‘ultra-hazardous’ activities only? ¹⁷ And, if so, what differentiates these activities from more pedestrian forms of risk? The approach of imposing strict liability for ultrahazardous activities is found across different municipal law systems, with a high degree of variance as to what activities attract strict liability.¹⁸ The US Restatement on the Law of Torts (Second) identifies the following factors:

§ 520: In determining whether an activity is abnormally dangerous, the following factors are to be considered: (a) existence of a high degree of risk of some harm to the person, land, or chattels of others; (b) likelihood that the harm that results from it will be great; (c) inability to eliminate the risk by the exercise of reasonable care; (d) extent to which the activity is not a matter of common usage; (e) inappropriateness of the activity to the place where it is carried on; and (f) extent to which its value to the community is outweighed by its dangerous attributes.¹⁹

While these factors may have some purchase in explaining the adoption of strict liability approaches within civil liability treaties, there is no generalizable approach in international law. Indeed, sub-paragraph (f), above, indicates a contextual approach where risk must be weighed against wider considerations of social utility. Thus, the determination of the approach to liability within sector-specific international regimes is more a function of state and industry policy preferences, and less a principled consideration of risk.

There is a further aspect to imposing liability on the basis of risk exposure that relates to sovereign equality and consent: the act of exposing others to risk that they cannot be presumed to have accepted justifies the imposition of liability without fault. Goldie notes that, unlike municipal legal systems that have sufficient authority to prohibit excessively risky activities, international law ‘is still largely a system of permissive and facultative norms’, which in turn justifies the imposition of strict or absolute liability.²⁰ The reasoning here is that states should not be able to unilaterally impose high levels of risk on other (equally sovereign) states without their consent. Strict liability apportions that risk by making the source state or operator responsible for the harm occasioned by its choice. Goldie’s approach is also influenced by distributive questions, particularly the degree to which the benefits from the activity are shared amongst states. Where activities involve socially beneficial outcomes, the utility structure favours the imposition of a fault-based, or at least a less stringent, approach, since there is a more balanced distribution of risks and benefits.²¹ Goldie analogizes the imposition of risk to a form of expropriation, suggesting the standard of liability is influenced by the nature of sovereign interests that affected states have in the impacted environment.²² The preferred approach is a liability rule that allows the activity to be carried out, but with payment of compensation in the event that another state’s sovereign interests are interfered with, as opposed to a rule that would prevent invasion of the interest without consent.²³

These concerns permeated the approach taken by Quentin-Baxter and Barboza in their roles as special rapporteurs in the International Law Commission’s (ILC) work on liability, where the approach was to impose liability without proof of fault, but to subject the allocation of losses to a form of equitable balancing.²⁴ In effect, the sovereign rights of both the source state – to engage in lawful but risky activities – and the affected state – to not be subjected to risk of harm without its consent – had to be reconciled, which in turn gives rise to the introduction of equity as a means of apportioning liability. While the approach was ultimately rejected as flawed and not supported by state practice, the concerns respecting exposure to risk and consent remain an important factor.

Considerations of the degree to which states may consent to activities and may benefit from those activities have some clear application to commons activities. Arguably, the sovereign interests affected in areas beyond national jurisdiction (ABNJ) are more attenuated and depend upon the characterization of the legal interest of states in the area or resource in question. In relation to activities on the high seas, which may be undertaken by states unilaterally, and the benefit of which accrues entirely to the state undertaking or authorizing the activity, the structure looks similar to transboundary harm, particularly if the interests of states in the global commons are viewed as sovereign amenities. For example, cable-laying is undertaken with little international oversight and is an activity that any state may engage in, subject to the due regard of other high seas freedoms and activities, yet may impose risks on states or on the international community as a whole, on which they have little say. Similarly, states have a wide margin of freedom to undertake scientific and tourism activities in the Antarctic, but in doing so impose risks of the international community from those activities. The ILC notes, with reference to a survey undertaken on national liability regimes, that ‘[t]he case for strict liability is strengthened when the risk has been introduced unilaterally by the defendant’.²⁵

The role of consent further complicates matters. For example, deep seabed mining is structured as a communally regulated activity, in which all parties to the 1982 United Nations Convention on the Law of the Sea (UNCLOS) have a degree of control over through their participation in the organs of the International Seabed Authority (ISA).²⁶ Not only are the risks of deep seabed mining not imposed unilaterally, but a portion of the benefits of the activity are to be equitably shared.²⁷ In this regard, the structure of the deep seabed mining regime may militate against the imposition of strict liability – at least on the basis that the allocation of risks and benefits justify shifting losses to the sponsoring state or operators under the sponsoring state’s jurisdiction.

Risk is not only a function of the nature of the activity but is also affected by the nature of the receiving environment. Where the potentially affected environment is fragile or less resilient, the risks of harm posed by activities carried out in those areas are heightened. The absence of scientific knowledge respecting impacts may also be viewed as a source of risk since the environmental outcomes are more challenging to predict. In these circumstances, reasonable steps may be difficult to determine ex ante, providing further justification for strict liability approaches in environmental sensitive ecosystems or receiving environments characterized by high levels of uncertainty. Such concerns have been raised in connection with the deep seabed and the Antarctic environment.²⁸

There is also an intergenerational element to risk allocation insofar as future generations neither consent to nor benefit (directly) from risky activities, but where unforeseeable or non-negligent environmental harm arises, the costs of addressing that harm is often borne by future generations through unremedied harm.²⁹ This may particularly be the case in relation to commons resources where the victim of the harm is the international community. The intergenerational dimensions of liability have been an emerging trend in atmospheric trust litigation.³⁰

Standards of liability may also respond to other shared objectives in international law. Viewed in light of an environmental harm prevention objective, strict liability may be justified as a means to promote deterrence of risky behaviour by providing greater incentives for operators to take steps to prevent accidental damage. This rationale applies equally, if not more, to fault-based liability, since what is sought to be deterred most often is intentional, reckless or negligent behaviour. In a no-fault context, the rationale of deterrence focuses on the imposition of a higher standard of care than mere non-negligence in order to avoid harms that are viewed as socially undesirable. In the case of pollution, deterrence also reflects the notion that harm prevention is preferred to compensation, given that some environmental harms may be difficult or impossible to restore, and that the full measure of harm is not easily quantifiable. As a regulatory matter, operators are much better positioned to take risk minimization measures, and therefore placing a higher standard facilitates greater care, as the law requires that the operator take all steps to prevent harm, not just those that are reasonable. In the absence of strict liability, operators are able to externalize the costs of measures taken to protect the environment that go beyond mere negligence.

In relation to states in their oversight role, it may be argued that strict liability might result in more vigilant oversight of operators. However, accidents that are causally connected to weak oversight would likely result in liability under a due diligence standard, and a higher standard would not prevent unforeseen or purely accidental harm. Strict liability for states has some potential to make more funds available for addressing harm since the responsible state effectively becomes the insurer of the operator, but this would depend on the financial capabilities of the state in question, and there may be more effective ways, such as pooled insurance and compensation funds, to achieve that goal.³¹

As a matter of environmental protection, and as a reflection of economic efficiency, cost internalization is often cited as a desirable policy goal.³² Cost internalization may promote more efficient methods of loss sharing through insurance or compensation schemes, which spread the risk amongst operators and better protect against unfunded harm due to insufficient funds. No-fault regimes may also provide for simplified dispute settlement processes, since the claimant is relieved of the burden of proving fault and may therefore be preferred on efficiency grounds; a goal that might be seen as being present under international law in the requirement for ‘prompt’ compensation.³³

Cost internalization is reflected in the inclusion of the polluter-pays principle in international declarations and treaties.³⁴ The polluter-pays principle has some clear purchase in the area of marine pollution,³⁵ and is identified as a relevant principle in relation to marine pollution from oil transport.³⁶ Outside the marine pollution area, it has been linked to strict liability under the Lugano Convention on Civil Liability for Damage Resulting from Activities Dangerous to the Environment.³⁷ The polluter-pays principle has been referenced as a core principle in both the deep seabed mining regime and the negotiations of a new international legally binding instrument (ILBI) on marine biodiversity beyond national jurisdiction.³⁸ As an allocation rule, the polluter-pays principle favours placing costs associated with environmental harm on the operator, not the victim, and in this regard favours strict liability.³⁹ The principle reflects the goal of deterrence and harm prevention, as well as recognizing that responsibility should follow those actors who benefit from activity. The principle is not without qualification and provides room for policy choices respecting exceptions and limitations on liability.⁴⁰

5.3 Approaches to Standards of Liability in International Law

The two principal approaches to addressing liability for environmental harm in international law involve leaving states as the primary subjects of liability through rules of state responsibility, or by channelling liability directly to operators.⁴¹ The approaches are not mutually exclusive. Where states opt to develop a civil liability regime, they may not necessarily divest themselves of responsibility, but rather make operators liable in the first instance.⁴² The policy choice regarding which party shall be primarily responsible is severable from the decision respecting the standard of liability, but in practice, states have opted to couple strict liability with civil liability regimes that channel liability to the operator, while maintaining a requirement for wrongful activity (fault-based liability, at least in the sense of a breach of international obligation) in relation to state responsibility for environmental harm.

5.3.1 State Responsibility

The default rules for state liability for environmental harm combine two fundamental rules. The first establishes the primary obligation on states to prevent transboundary harm. This obligation applies to activities under state control and includes harm to both the territory of other states, as well as harm to areas or resources beyond national jurisdiction. The crucial feature of the no-harm rule for current purposes is that it is a rule of due diligence; that is, the standard of liability is negligence-based, not strict. The second is the basic rule of state responsibility that maintains that states are responsible for the harm that flows from breaches of their international obligations. Thus, this rule requires the responsible state to make reparations for the injury caused by wrongful acts that are attributable to the state.⁴³ Reparations include restitution and compensation by way of damages.⁴⁴

The due diligence obligation to prevent harm is well established in international law. The rule has been recognized in numerous decisions of international courts and tribunals,⁴⁵ and finds expression in numerous treaties,⁴⁶ as well as in Principle 21 of the 1972 Stockholm Declaration and Principle 2 of the 1992 Rio Declaration.⁴⁷ Ultimately, after consideration of state practice, the ILC adopted a due diligence standard in relation to the obligation of states to prevent transboundary harm,⁴⁸ while leaving the precise contours of liability to be determined in accordance with the obligation to provide recourse for victims of harm through domestic or other agreed upon mechanisms.⁴⁹

The extension of the no harm principle to areas beyond national jurisdiction is explicitly recognized in Principle 21 and Principle 2, and is reflected in treaty commitments concerning commons resources, such as the deep seabed,⁵⁰ the high seas,⁵¹ as well as the Antarctic environment.⁵² The Seabed Disputes Chamber (SDC) considered the nature of the due diligence obligations owed by sponsoring states in the context of mining activities in the Area. The SDC described the nature of the sponsoring states obligations flowing from the specific provisions within Part XI of UNCLOS as follows:

The sponsoring State’s obligation ‘to ensure’ is not an obligation to achieve, in each and every case, the result that the sponsored contractor complies with the aforementioned obligations. Rather, it is an obligation to deploy adequate means, to exercise best possible efforts, to do the utmost, to obtain this result. To utilize the terminology current in international law, this obligation may be characterized as an obligation ‘of conduct’ and not ‘of result’, and as an obligation of ‘due diligence’.⁵³

The reasoning of the SDC concerning due diligence was subsequently adopted by the International Tribunal for the Law of the Sea (ITLOS) and applied to the obligations of flag states in relation to fisheries activities in the exclusive economic zone, and more broadly to obligations to conserve living resources with the marine environment.⁵⁴

As discussed earlier, a number of commentators, inside and outside the ILC’s work on ‘international liability for injurious consequences arising out of acts not prohibited by international law’, have argued in favour of imposing a no-fault standard in relation to those activities that can be classed as ultrahazardous in nature.⁵⁵ The principal justification relates to the role of the source state in authorizing the risk. In such circumstances, the source state voluntarily creates a risk, which is involuntarily borne by the affected state. Despite the broad acceptance of the underlying logic, the support for such a principle in international law is weak. The regimes respecting nuclear facilities, oil pollution and other hazardous activities have all channelled liability to the operator, and thus, do not speak to state liability. The only example of strict liability imposed directly on states is the 1972 Convention on International Liability for Damage Caused by Space Objects,⁵⁶ and the Cosmos 954 claim that was filed under that treaty.⁵⁷

The approach taken in relation to space objects can be distinguished from other state activities in the commons on the basis of the role of the state in the activity in question. Unlike the placement of space objects, which may be understood as an activity (at least until recently) requiring direct state involvement, in other commons activities, such as fisheries or deep seabed mining, the sponsoring state is only involved in the activity in its oversight role. Thus, the issue of control, which is fundamental to the deterrence justification,⁵⁸ is indirect. Liability for oversight activities is unquestionably fault-based. Where states are acting as operators, for example as contractors in relation to activities in the Area, they will typically be subject to the same liability requirements applicable to other (non-state) operators. This is also evident in the Liability Annex adopted under the Antarctic Treaty system, where state and non-state operators are subject to strict requirements to respond to environmental emergencies, albeit with each subject to different procedural requirements.⁵⁹

The issue of risk was addressed by the 2011 Activities in the Area Advisory Opinion, where in the context of considering the content of due diligence, the SDC notes:

The content of ‘due diligence’ obligations may not easily be described in precise terms. … It may change over time as measures considered sufficiently diligent at a certain moment may become not diligent enough in light, for instance, of new scientific or technological knowledge. It may also change in relation to the risks involved in the activity. As regards activities in the Area, it seems reasonable to state that prospecting is, generally speaking, less risky than exploration activities which, in turn, entail less risk than exploitation. … The standard of due diligence has to be more severe for the riskier activities.⁶⁰

The approach to risk, therefore, is not to alter the standard of liability, but to vary the content of due diligence. States may be held to a high standard of vigilance in relation to riskier activities, but the basis of liability will be the failure of the state to meet the standards of conduct that the particular context requires, and not causation alone. The SDC also considers the relationship between due diligence and the precautionary principle,⁶¹ noting that precaution is ‘an integral part of the general obligation of due diligence’.⁶² This then requires that the sponsoring state incorporates prudential risk assessment in exercising due diligence. Consistent with the approach cited above, the precaution does not operate to alter the standard of liability, but rather informs the content of the standard of care.

5.3.2 Civil Liability

5.3.2.1 Approach to Liability

While international law does not support strict liability for states in connection with transboundary (including impacts to areas beyond national jurisdiction) environmental harm, states are under an obligation to take measures to ensure prompt and adequate compensation in the event of harm occurring.⁶³ These measures may be taken in their domestic legal systems or through collective measures, such as sector-specific civil liability regimes.⁶⁴ The current sectors where civil liability regimes have been negotiated include nuclear facilities,⁶⁵ oil pollution, carriage of hazardous and noxious substances by sea,⁶⁶ bunker oil,⁶⁷ hazardous waste,⁶⁸ transboundary movement of living modified organisms⁶⁹ and Antarctic activities.⁷⁰ The standard of liability for operators under these civil liability regimes is strict but not absolute.⁷¹ The standard of strict liability is implemented through a provision that indicates that the responsible party ‘shall be liable for any pollution damage’ caused by the activity in question.⁷² In order to recover damages, the claimant need only prove a causal link between the activity and the damage.

The policy justifications for imposing strict liability, as discussed above, include the objective to ensure prompt and adequate compensation, including available compensation for remediation and reinstatement of environmental harm; the desire to encourage a high standard of care and deter pollution; the polluter-pays principle; the recognition of the fairness of having the creator of risks (as opposed to the victim) bear losses associated with that activity; and the greater efficiency of providing for compensation without proof of fault. As the entity that directly controls the activity, the policy justification for a strict standard is stronger for operators than for states. Similarly, the polluter-pays principle is more clearly applicable to operators (who are directly causally responsible for harm).

The exclusion of environmental damage to areas beyond national jurisdiction under civil liability regimes is a reflection of the uncertainty of standing to recover for harm in areas beyond national jurisdiction, but also points to the incompleteness of civil liability regimes in responding to the preventive and remedial aims of compensation.⁷³ The 2001 Bunker Oil Convention, for example, cites both articles 194 and 235 of UNCLOS in its preamble, indicating an intent to address ‘all damage caused by pollution of the marine environment’, but goes on to exclude damages in areas beyond national jurisdiction.⁷⁴

While states have adopted a consistent approach to operator liability in international civil liability regimes, it remains an open question whether the duty to take measures to ensure prompt and adequate compensation requires the adoption of a strict liability standard in all cases involving hazardous or ultrahazardous activities. The argument is different from the discussion above on whether states are themselves strictly liable for ultrahazardous activities under their jurisdiction or control, as the subject of liability here is the operator not the state itself, and therefore flows indirectly from the obligation to provide prompt and adequate compensation. This is the approach of the ILC in the Draft Principles,⁷⁵ although the ILC’s position appears to reflect more of a policy preference than a recognition of an established or emerging requirement in international law.⁷⁶

Generalizing from the practice derived from civil liability regimes is challenging because each regime reflects the particular constellation of interests amongst the states, operators and providers of financial assurance, as reflected in a variety of approaches to exceptions and liability caps. Nonetheless, the consistent imposition of strict liability on operators indicates a high degree of consensus amongst states that strict liability best serves the multiple objectives of liability regimes, and, as such, creates a high burden of justification on states that seek to use a fault-based approach. While the civil liability conventions tend to exclude harm to the commons as compensable damage, the considerations that have informed the preference for strict liability approaches apply equally to harm in areas beyond national jurisdiction.

5.3.2.2 Exceptions to Liability

Liability is said to be strict, not absolute, because each civil liability regime contains exceptions to the imposition of liability, which range in their breadth.⁷⁷ Even the 1997 Vienna Convention, which identifies the imposed standard as ‘absolute’,⁷⁸ contains a narrow set of exonerating circumstances, namely incidents due to armed conflict and ‘a grave natural disaster of an exceptional character’.⁷⁹ More typically, treaties contain a longer list of exceptions,⁸⁰ which include:

• Armed conflict;

• Intentional damage by a third party;

• Contributory negligence (the incident resulted from the intentional or negligent actions of the claimant);

• Damage caused by government negligence;

• The result of ‘a natural phenomena of exceptional, inevitable, unforeseeable and irresistible character’; or⁸¹

• Damage caused as a result of compliance with a compulsory measure of a public authority.⁸²

The presence of exceptions moves away from a rigid application of polluter-pays and appears to be largely driven by issues of fairness and control. For example, exonerations based on governmental contributory negligence or compliance with governmental measures respond to the inequities of imposing liability on an operator where the fault lies elsewhere. The natural phenomena exception can be justified on the basis that the exonerating circumstances are limited to those instances where the event is unforeseeable and the resulting damage cannot be guarded against. Given that the liabilities are typically insured against as part of the scheme, the exceptions may also reflect the unwillingness of insurers to cover wholly unforeseeable events.

5.3.2.3 Liability Caps under Civil Liability Regimes

The other common feature of civil liability regimes is the practice of limiting liability to identified compensation caps on a per incident basis. The presence of liability caps responds to the practicality of insurance and pooled compensation funds, since insurers and funds cannot take on limitless liability.⁸³ The approach is a further derogation from the application of the polluter-pays principle, as it may result in victims of environmental harms having to bear some of the losses themselves. Coupling liability caps with strict liability approaches reflects the greater acceptability from an ethical standpoint of relieving a non-negligent, but causally responsible, party from the obligation to provide full compensation. Of course, the absence of a fault requirement does not necessarily mean that the responsible party did not act without requisite care. However, the efficiencies associated with a more simple, strict liability approach counterbalance the desirability of holding negligent parties fully responsible. However, the 1999 Basel Liability Protocol provides for unlimited liability where the harm is a result of ‘wrongful intentional, reckless or negligent acts or omissions’,⁸⁴ which is consistent with imposing a higher degree of responsibility (in terms of compensation) for morally wrongful acts.

The amount and structure of the caps is highly variable, but there are some evident attempts to match the amounts to reasonably anticipated claims. For example, where the maximum liability amounts in the nuclear regime appeared insufficient following the Chernobyl incident, the amounts were raised.⁸⁵ A similar reaction has been seen in the oil pollution regime where severe incidents led to concerns about unfunded damages, which in turn led to higher overall ceilings. Thus, despite the practical considerations surrounding insurability, there remains an evident desire to prevent loss shifting to victims of harm.

5.4 Standards of Liability in ABNJ

5.4.1 Antarctic

While not in force, the Convention on the Regulation of Antarctic Mineral Resource Activities (CRAMRA) provides a useful example of the structure of liability rules in the Antarctic in the face of risky activities. CRAMRA, which sets out the rules and procedures governing mineral resource exploitation in the Antarctic, contains specific liability rules that impose strict liability on the operator for broadly defined environmental harms and economic losses arising from its resource activities.⁸⁶ The operator’s liability is subject to very narrow exceptions, namely damage caused by ‘a natural disaster of an exceptional character’ and armed conflict.⁸⁷ The liability rules are incomplete, and a further protocol was contemplated (but never negotiated), which may have contained further provisions placing limits on liability, in conjunction with financial assurances and the establishment of a fund.⁸⁸ The liability provision of CRAMRA also addresses state liability in article 8(3) which provides for sponsoring state liability, where the sponsoring state’s failure to carry out its oversight obligations under the Convention contributed (from a causation standpoint) to compensable damage.⁸⁹ The distinction between operators and sponsoring states is functional in that states, when involved directly as resource operators, will be subject to the strict standard of liability. Fault-based liability will apply to states in their oversight capacity.

The subsequent negotiation of the 1991 Protocol on Environmental Protection to the Antarctic Treaty (1991 Antarctic Protocol) prohibits extractive activities in the Antarctic in accordance with the objective to prioritize the conservation and protection of the Antarctic environment, effectively denouncing the objectives of the CRAMRA.⁹⁰ The structure of the principal obligations respecting environmental protection under the 1991 Antarctic Protocol are conduct, not result, based.⁹¹ Liability rules and procedures are governed by a separate Annex, the negotiation of which was anticipated under the 1991 Antarctic Protocol.⁹² As discussed in Chapter 2, the approach has an administrative posture in that the primary obligation in the face of an environmental emergency is a ‘response action’. Liability flows from the failure to take appropriate response actions, and damages are related to the costs actually incurred or estimated for the response action.⁹³ Operator liability is strict but subject to exceptions and liability caps.⁹⁴ Although, the liability caps do not apply to damage arising from acts that are committed with intention to cause an emergency or recklessness.⁹⁵ The state liability provision adopts a fault-based standard, with states only being held liable for the failure of an operator to take response actions where the state failed to take ‘appropriate measures within its competence, including the adoption of law and regulations, administrative actions and enforcement measures, to ensure compliance with the Annex’.⁹⁶ The content of due diligence will reflect the obligations on states to protect the Antarctic environment as specified in the 1991 Antarctic Protocol such as carrying out environmental assessments, and monitoring activities,⁹⁷ but also reflects general customary legal requirements respecting harm prevention.

5.4.2 Deep Seabed

The rules governing liability from activities in the Area are a combination of general provisions within UNCLOS, provisions dealing with deep seabed mining specifically found within Part XI and Annex III of UNCLOS, and requirements found within the regulations enacted by the ISA. There is no special liability regime that has been developed by the ISA to date, although the need for such rules has been acknowledged by the ISA.⁹⁸ In keeping with other sector-specific liability regimes, the liability rules for deep seabed mining distinguish between the liability of contractors (operators) and of sponsoring states. UNCLOS also recognizes that the ISA, which shares oversight duties with sponsoring states, may also be liable for damages arising from its own activities.⁹⁹

For contractors, liability for damage arising from their activities in the Area is addressed in Annex III, article 22, which provides that ‘contractors shall have responsibility or liability for any damages arising out of wrongful acts in the conduct of its operations …’.¹⁰⁰ The phrase ‘wrongful act’ should not be interpreted as requiring fault-based liability. ‘Wrongful’ in this context should be taken to mean that liability will flow from a breach of legal requirements to which the contractor is subjected to. Article 22 of Annex III is analogous to the basic rule of state responsibility that recognizes that liability flows from breaches of international law attributable to the state.¹⁰¹ The requirements for fault will be determined by the specific requirements imposed on contractors by UNCLOS and the rules enacted by the ISA.

The obligation on contractors to prevent environmental harm in relation to exploration activities is set out in the ISA’s regulations. As it stands under the Exploration Regulations,¹⁰² the standard of liability imposed on contractors requires a failure of due diligence. Regulation 31(5) of the Regulations on Prospecting and Exploration for Polymetallic Nodules in the Area (PMN) is framed as a duty of conduct to take ‘necessary measures to prevent, reduce and control pollution and other hazards to the marine environment arising from its activities in the area as far as reasonably possible, applying a precautionary approach and best environmental practices’.¹⁰³ Accidental damages that arise despite all reasonable measures being taken or damages that are unforeseen are not currently ‘wrongful’, and therefore, not compensable under the ISA’s rules. Although, where the failure to comply with a direct, primary obligation results in harm – for example, failing to comply with an emergency order – non-compliance ought to be viewed as wrongful, with liability consequences flowing from the non-compliance.¹⁰⁴ It is open for the ISA to impose a strict liability standard on contractors through the enactment of further rules that entail obligations of result. The potential for the development of rules further specifying compensation obligations is expressly contemplated in UNCLOS.¹⁰⁵ The ISA’s Draft Exploitation Regulations (DER) takes a similar administratively oriented approach, whereby contractor liability flows from its wrongful acts, which could include obligations of conduct and result.¹⁰⁶

The liability of sponsoring states was addressed comprehensively by the SDC in its 2011 Advisory Opinion.¹⁰⁷ The principal obligations on sponsoring states are contained in article 139(1) and article 4(4) of Annex III of UNCLOS, both of which impose a ‘responsibility to ensure’ that mining activities are carried out in accordance with the requirements of Part XI.¹⁰⁸ The focus of these provisions is on the oversight functions of the sponsoring states; although to be clear, the wording of article 139 is directed towards all states parties, not just sponsoring states. These obligations were characterized by the SDC as being those requiring adherence to standards of conduct, namely ‘due diligence’.¹⁰⁹

The content of due diligence is driven by the treaty language and context, but also appears to be influenced by more general customary rules governing state obligations to prevent environmental harm. The primary requirements of due diligence require states to adopt ‘reasonably appropriate’ laws and regulations and to take measures to secure compliance.¹¹⁰ The SDC enumerates a set of further obligations that must be complied with as a measure of due diligence:

• The obligation to assist the Authority in the exercise of control over activities in the Area;

• The obligation to apply a precautionary approach;

• The obligation to apply best environmental practices;

• The obligation to take measures to ensure the provision of guarantees in the event of an emergency order by the Authority for protection of the marine environment;

• The obligation to ensure the availability of recourse for compensation in respect of damage caused by pollution; and

• The obligation to conduct environmental impact assessments.¹¹¹

The relationship between these obligations and due diligence is complex. As noted, they are constituent factors that contribute to the state’s general obligation to take reasonable steps to prevent harm to the marine environment. However, these obligations are separate or direct obligations on the sponsoring state (each of which is identified under the rules applicable to deep seabed mining), and as such, states are required to comply with each of these obligations independently from their general obligation to ensure contractor compliance.¹¹²

The SDC also considered the question of whether the development status of the sponsoring state is a relevant factor in determining the reasonableness of the oversight steps taken by the sponsoring state. In holding that the obligations apply equally to all states regardless of development status, the SDC relies on the specific wording of Part XI, which discloses no intent to differentiate oversight obligations on the basis of development status.¹¹³ The SDC also observes that, were responsibilities to be differentiated between developing states and developed states, there may be incentives for contractors to seek sponsorship from states that are subject to a lesser set of oversight obligations, linking the uniform content of due diligence to ‘the highest standards of protection of the marine environment, the safe development of activities in the Area and the protection of the common heritage of mankind’.¹¹⁴

The obligation on sponsoring states to provide recourse for victims of harm within their domestic legal systems under article 235 is also viewed as part of that state’s due diligence obligations. This obligation requires states assure ‘prompt and adequate compensation’. As discussed above, the standard of ‘prompt and adequate compensation’ supports, but does not require, the imposition of strict liability. The SDC points out that article 235(2) ensures that the contractor can live up to its obligation to provide reparation for damages caused by its wrongful acts. It appears open for the sponsoring state to impose domestic rules that provide for strict liability, regardless of the approach taken by the ISA, although this may turn on whether imposing a strict standard is seen as being ‘inconsistent with Part XI’.¹¹⁵ This provision does allow sponsoring states to adopt rules that are ‘more stringent’ than those adopted by the Authority, which may provide greater latitude for sponsoring states to impose a strict liability standard.¹¹⁶ Article 235(3) recognizes that states may also address this objective through the development of a specialized international (civil) liability regime, which could include compulsory insurance or compensation funds.¹¹⁷

As the first of the enumerated direct obligations indicates, sponsoring states do not have sole responsibility for oversight of mining operators. These responsibilities are shared with the ISA. Consequently, the ISA is also liable for damages arising out of its own wrongful conduct.¹¹⁸ The standard of liability for the ISA is not addressed in the SDC’s 2011 Advisory Opinion, but flows from its obligations in article 153 to ‘exercise such control over Activities in the Area as is necessary for the purpose of securing compliance with the provisions of this Part’.¹¹⁹ While the wording of the obligations of the ISA does not match with that of sponsoring states (‘responsibility to ensure’), the thrust of the obligation to oversee is the same, and, ought, therefore, to be understood as requiring due diligence.

5.4.3 High Seas

The standard of liability for activities causing harm within the high seas area will again be a function of the specific obligations to prevent harm. The nature of the commitments within UNCLOS to protect the marine environment and its resources was the subject of the Sub-Regional Fisheries Commission Advisory Opinion (SRFC Advisory Opinion) issued by the ITLOS in 2015,¹²⁰ as well as the South China Sea Arbitration.¹²¹ The SRFC Advisory Opinion was concerned with the obligations of flag states to prevent illegal, unreported and unregulated fishing activity in the exclusive economic zones of other states, but the reasoning of the ITLOS applies equally to activities in the high seas.¹²² In reviewing these obligations,¹²³ the ITLOS notes that flag states are required to exercise ‘effective jurisdiction and control in administrative matters’ over fishing vessels subject to their jurisdiction.¹²⁴ This requires flag states to adopt appropriate laws and to take measures to ensure compliance with those laws.¹²⁵ As a set of oversight obligations, the standard of liability is due diligence.¹²⁶ The ITLOS adopts the reasoning of the SDC in its 2011 Activities in the Area Advisory Opinion as to the variable and contextual nature of due diligence. Due diligence obligations extend to international organizations, such as the European Union, that exercise jurisdiction over aspects of the activities in question.

One important clarification respecting the nature of due diligence provided in the SRFC Advisory Opinion relates to the relationship between harm and due diligence. The issue was framed in terms of whether flag states could be found to have breached their due diligence obligations in the event of isolated illegalities or whether a breach required a more sustained pattern of illegality. In holding that the frequency of illegal fishing activity is not a relevant consideration, the ITLOS centres the analysis of due diligence on the adequacy of the measures taken, not the frequency of the illegal activity.¹²⁷ In principle, the reasoning is sound; liability will flow where a causal relationship between environmental harm and insufficient oversight can be demonstrated. In practice, however, determining the reasonableness of the oversight will be influenced by the degree of compliance that the measures are likely to bring about.

The SRFC Advisory Opinion does not address the issue of the relative capabilities of states in relation to their due diligence obligations, but there may be reasons to consider whether the approach of the SDC on this issue is generalizable to activities in the high seas. There are provisions within UNCLOS that may be relevant to the determination of the standard of oversight required. For example, article 194(1) requires states to take measures to prevent marine pollution ‘in accordance with their capabilities’. The 1996 London Protocol to the Convention on the Prevention of Marine Pollution by Dumping qualifies the obligation to protect and preserve the marine environment in a similar fashion.¹²⁸ Notably, however, the recognition of the relevance of differentiation capabilities is not present in other key provisions on oversight obligations in the high seas, including the obligation to conserve living resources,¹²⁹ and the provisions requiring states to ‘take all necessary measures to ensure activities under their jurisdiction’ are conducted so as not to cause pollution in areas outside of their jurisdiction.¹³⁰ In this regard, the approach of the SDC is more important than the result. What is required is a careful consideration of the specific obligations and the context of their application, as well as the minimal reasonable requirements for ‘vigilance, employment of infrastructure and monitoring of hazardous activities’ that are expected.¹³¹ Given that developed states are under obligations to share technologies and contribute to the capacities of developing states, the availability of such mechanisms and support to developing states is a further salient consideration.¹³²

As with fisheries, there is no specialized regime for liability for environmental harm arising from shipping activities in the high seas. The result is that the default standard of liability will reflect general obligations of due diligence for states and domestic negligence standards for operators. States have the ability to impose strict liability on ships operating (flagged) under their jurisdiction, but the incentives to do so are minimal in the absence of international cooperation to impose a uniform standard. The SRFC Advisory Opinion has relevance for the standard of conduct that applies to flag states, including those states that maintain open registries, in relation to oversight of shipping activities. Article 94 requires any state to ‘exercise its jurisdiction and control in administrative, technical and social matters over ships flying its flag’.¹³³ A failure to exercise due diligence exposes the flag state to liability for environmental harm that is causally connected to oversight failures.

Standard limitations of liability under the Convention on Limitation of Liability for Maritime Claims (1976 LLMC) may still serve to limit liability regardless of the standard imposed.¹³⁴ While the 1996 Protocol to amend the 1976 LLMC (1996 LLMC Protocol) exempts claims arising under the oil pollution liability regime, the non-application of the oil pollution liability rules to areas beyond national jurisdiction means that the 1976 LLMC limits will have broad application to environmental claims in the high seas.¹³⁵ The one exception to the fault-based standard is the transportation of nuclear materials by sea that is covered under the 1997 Vienna Convention, which applies to damage ‘wherever suffered’, including the high seas.¹³⁶

Due diligence, as a standard of required state behaviour, has a broad application to other high seas activities or to matters affecting the high seas, with implications for the development of new liability rules for emerging ocean activities and concerns, including marine genetic resources,¹³⁷ ocean acidification¹³⁸ and ocean fertilization.¹³⁹ The difficulty is not with extending the general obligation to emerging circumstances, but rather with identifying the content of the standard of care. In relation to state responsibility, what amounts to reasonable oversight steps will depend upon the surrounding normative environment. In the deep seabed mining context, the SDC draws on the regulatory framework with Part XI, as elaborated upon in the ISA’s regulations. While the SDC focused on the application of due diligence to sponsoring states, it is equally clear that the contractor’s standard of care will be assessed in light of the regulatory requirements of the regime. Similarly, state oversight of ocean fertilization activities is likely to be assessed in light of the requirements of the 1996 London Protocol.¹⁴⁰ Generally accepted international rules and standards (GAIRS) will also play an important role in defining the standard of care, although the lower density of regulations and guidance in the high seas that could structure the content of state and private actor due diligence will pose some challenges in determining applicable standards of conduct with precision.

5.5 Conclusions

At the heart of the policy question concerning standards of liability is the distribution of losses following an event which causes harm to third parties, be they states or individuals. The prevailing approach within the law of state responsibility is not to impose a strict standard on states in relation to activities under their jurisdiction or control. States are simply unwilling to become the insurers of environmentally risky activities, preferring instead to oversee these activities with due diligence. It is unsurprising that in relation to activities affecting the commons environment that states have not been more open to moving towards a strict standard. The distributive calculus of risk in the commons does not favour a strict standard since states are not required to bear the full risk of environmental harm themselves, but are to share that risk with all states, and in many cases, with future generations. Restricting loss-shifting to failures of state due diligence subjects the international community to risks that they neither consent to nor control, although states can influence the content of the standard of care through their own oversight actions and through the development of international standards of duly diligent conduct. The concerns, first raised by Goldie, with foreseeability of harm for emerging activities and where there is an absence of clear standards of behaviour, have ongoing purchase in commons activities, where the risks to the environment are often less well understood.

The shifting of losses to third parties and to the international community as a whole can, however, be substantially mitigated through the imposition of strict liability standards on operators. The greater willingness to use strict liability standards in civil liability regimes is again a reflection of risk distribution, where the victims of environmental harm, which often include states themselves, are less diffuse. There is a growing association between the obligation on states to provide prompt and adequate compensation and the imposition of strict liability on operators. This association reflects a policy preference, as opposed to being an obligation. However, the consistent adoption of strict liability standards for activities that pose clear transboundary risks, including risks to commons resources, raises a strong presumption in favour of strict operator liability, albeit accompanied by liability caps. Insofar as strict liability incentivizes higher standards of care and reduces the risk of environmental harm that is not remediated, the rationale supporting no-fault liability applies with equal force to the global commons.