Review of ‘Compensation for Permissible Harm by Prof. Sandy Steel by Wong Yi Xuan (BCL ‘26)

I anchor my ship at your dock on a stormy night, and my ship crashes repeatedly into the wood; firefighters push a burning chimney stack onto your roof to prevent it from collapsing onto a London street; I break your arm to stop you from pushing me off a cliff. Steel’s article provides an excellent account of when and why private law imposes compensatory liability for permissible harm in some, but not all cases. I provide a brief overview of Steel’s main lines of argument, before arguing that his fairness-based account of the normative foundations of liability for permissible harm has theoretical payoffs in public law—namely, it explains the compensatory role of damages under the Human Rights Act 1998 (henceforth ‘HRA’) for permissible infringements of Article 1 Protocol 1 of the European Convention of Human Rights (henceforth ‘A1P1’).

Steel taxonomizes English legal rules which do/do not create compensatory duties/liabilities for permissibly caused harm as follows:

Private necessity (property): B poses no threat to A’s right. A infringes B’s right in order to protect only A’s property right from harm. Duty to compensate arises.

Private necessity (bodily rights or life): B poses no threat to A’s right. A infringes B’s right in order to protect only A’s right bodily rights or right to life from harm. Duty to compensate does not arise, although it does in US and German law.

Public necessity (bodily rights or life): B poses no threat to anyone’s rights (or this threat does not justify the infringement), and A infringes B’s right in order to protect the bodily rights, right to life, or property rights of others from harm. Duty to compensate does not arise.

Compensation in lieu of injunction in private nuisance. Duty to compensate arises.

Compensation for lawful action by state officials or employees. Duty to compensate arises.

Harm to the risk-imposer caused in self- and other-defence. Duty to compensate does not arise.

Harm to an innocent third party caused in self- and other-defence. Duty to compensate sometimes arises.

Harm caused by protection of one’s property under the common enemy rule. Duty to compensate does not arise.

Public necessity. Duty to compensate does not arise.

Following this, he argues that the following relationships between categories of case can (generally) be understood consistently: (1) defensive harm versus private necessity, (2) public versus private necessity, (3) private necessity and the common enemy rule, (4) public necessity and compensation in lieu of an injunction, and (5) public necessity and compensation for lawful right-infringing acts of the state.

Turning to the normative foundations of compensatory liability for permissible harm, Steel diverges from the following view:

RIGHTS: Moral (and legal) claim rights may permissibly be infringed, but since a right was infringed, this calls for compensation for harm caused within the scope of the right.

Steel’s argument, instead, is threefold. The first premise: where A has a(n all-things-considered) moral permission to Ф, A’s Ф-ing does not infringe B’s rights. Instead, the proper explanation for such cases is that A’s duty to not-Ф (correlating to B’s claim-right) is cancelled. To illustrate, ‘if you permit me to walk in your garden, I do not “permissibly infringe” your right that I not use your property without your consent’, and ‘if you are wrongfully attacking me and I use reasonable force to defend myself, I do not “permissibly infringe” your right not to be subject to force without your consent: you forfeited that right (temporarily) by your wrongful attack. The right is not transgressed at all’.

To illustrate of the premises to follow, Steel offers the following hypothetical:

Vincent 2: A anchors A’s ship on B’s dock in a ferocious storm, thereby causing damage to B’s dock. A will certainly die unless A’s ship remains anchored on B’s dock.

The second premise: in some cases such as Vincent 2, A has a temporary right to use B’s property to save A’s life. The third premise: in some cases such as Vincent 2, B’s property right is qualified by A’s temporary right. The fourth premise: where ‘morality permits harmful interference with an innocent person in order to benefit another’, it minimises the assault on the innocent person by giving rise to a duty to compensate, owed by those for whose benefit the innocent person’s right was, as a quid pro quo for morality making an exception on A’s account. The conclusion: in some cases such as Vincent 2, RIGHTS does not explain A’s compensatory duty, since (ex hypothesi) there was no ‘infringement’ of B’s rights. Rather, A’s compensatory duty is justified by the fairness of allocating the cost of A’s protection to A, not B. This is arguably stronger than Steel’s claim that RIGHTS is not falsified, but merely ‘not needed’ to explain A’s liability in Vincent 2: RIGHTS assumes that B’s rights have been infringed by A’s conduct, whereas Steel’s fairness-based account assumes that they have been qualified for A’s protection.

A further, theoretical payoff of Steel’s fairness-based account is that it explains why A does not attract (moral or legal) compensatory liability in public necessity cases such as the following:

Vincent 3: A anchors C’s ship on B’s dock in a ferocious storm, thereby causing damage to B’s dock. C is unconscious and will certainly die unless C’s ship remains anchored on B’s dock.

Here, B’s right is qualified by C’s temporary right to use B’s property to save C’s life, and A merely gives effect to C’s right. Fairness requires the costs of C’s protection to be allocated to C, not to A or B.

Call Steel’s fairness-based account of compensatory liability for permissible harm FAIRNESS. Steel’s distinction between RIGHTS and FAIRNESS has a further theoretical payoff in the public law context: it explains the different roles played by compensation in the context of HRA claims brought for violations of A1P1.

Many ECHR rights, including the A1P1 right to ‘peaceful enjoyment of possessions’, are qualified. To determine whether there has been a HRA/ECHR breach, the courts engage in a proportionality assessment, to determine (inter alia) whether a ‘fair balance’ has been struck between the interests of individual right-holders and the public writ large. Section 8(3) of the HRA empowers UK courts to award damages for successful human rights claims, where ‘necessary to award just satisfaction’ to the claimant. In most cases, breach and compensation are considered at distinct stages in the courts’ reasoning: the courts first ask whether an HRA/ECHR breach has occurred, and then ask whether compensation should be awarded.

A1P1 claims are distinct, because compensation frequently features in both stages above. Consider, for example, Mott v Environment Agency. Mott operated a ‘putcher rank’ fishery as his full-time occupation, catching and selling around 623 fish a year. In 2012, the Environment Agency served notice to Mott that his annual licence to catch salmon would be limited to 30 fish for the 2012 season—this number was lowered to 23 and 24 fish in 2013 and 2014 respectively. The Agency justified these measures on the basis of protecting the ‘at risk’ salmon stock in the local rivers, which are designated as Special Areas of Conservation in EU law. The Agency paid Mott compensation for lost profits on various previous occasions, but not in respect of the restrictions imposed from 2012-2014. Mott successfully challenged the limit conditions, which made his fishery ‘wholly unprofitable’ to operate, for breach of A1P1. Compensation featured at two stages of the court’s analysis: first, whether the Environment Agency had struck a ‘fair balance’ between Mott’s and the public interest (answer: it had not), and second, having found an A1P1 breach, whether damages should be awarded to Mott (answer: they should).

These two stages at which compensation is considered reflect FAIRNESS and RIGHTS specifically. At the breach stage, the court ex hypothesi has not yet found that the claimant’s A1P1 rights have been breached, and is thus (in Steel’s language) seeking to determine whether the beneficiary has offered the claimant a fair quid pro quo for the qualification of their property rights in the public interest—as per FAIRNESS. At the remedies stage, the court has found a breach of the claimant’s A1P1 rights. The breach, inflicted (in Mott) for ‘necessary environmental controls’, is made permissible by the award of damages—as per RIGHTS. This divergence also maps onto Steel’s statement that RIGHTS and FAIRNESS may coexist as explanations for compensatory duties in different cases.

One further implication of applying Steel’s explanatory framework to the public law context is of interest. Where compensation features in the courts’ analysis of A1P1 breach, A1P1 rights are arguably qualified because the public writ large has a moral/legal permission to constrain an individual’s (exercise of) property rights, but not a moral/legal right. Steel pairs permissions and rights together when he states:

‘[Morality] effects a change in the normative situation of others for whose benefit these exceptional permissions or rights to harm arose.’ (emphasis mine)

In the cases that Steel considers (eg Vincent 2 and 3 above), there is a clearly defined person or group who enjoys the permission/right against B. In cases such as Mott, however, it is not clear who the relevant right-holder would be. The Environment Agency exercises special powers against Mott—to restrict his fishing license, to demand that he comply with those restrictions, etc, as per will theories of rights. Yet, it is not in the Environment Agency’s interest that Mott comply, as the Agency has no intrinsic interest in the number of salmon caught or left behind. One may turn to the more plausible candidate of the public writ large, which bears the burden of compensation since damages awarded against the Environment Agency are paid for using public funds. However, five potential issues arise. First, in terms of legal rights, cases such as Mott are not reasoned on the basis that the public (or any subset thereof) has a claim-right against the claimant. Second, in terms of moral rights, it is not clear whether the public (or any subset thereof) is a collective agent capable of being a moral right-holder properly-so-called. Third, it is not clear which subset of the public is best understood as the relevant right-holder—the public writ large, since ‘the quality of the natural environment is of legitimate concern to everyone’? Other salmon fishermen? Other fishermen in general? Salmon consumers? Fourth, it is not clear whether the public (or any subset thereof) exercises relevant ex ante powers (eg waiver) or ex post powers (eg calling for justification) against Mott, contra will theories of rights. Fifth, so far as the public (or any subset thereof) has an interest in Mott’s compliance with fishing restrictions, it is not clear whether that interest is ‘sufficient’ to ground a right, on interest theories of rights. In FAIRNESS, cases in which B’s right is qualified by C’s rights versus by C’s permissions may thus come apart.

Professor Sandy Steel’s article can be accessed in full here.