Peter Watts KC
This is a reply to David Bullock’s response to my opinion piece in this publication earlier in August. Ideally, this reply should be read with those two earlier pieces.[1] Since those pieces were written, the relevant legislation preventing tort actions for “emissions-related climate change”, the Climate Change Response (Tort Liability) Act 2026, has been enacted.
In the following I have omitted legal citations for the cases that have already appeared in the debate. I take Bullock’s points in the order that he made them.
The Luton case and the Thames
Bullock opens by responding to my quotation of Page Wood V-C in Att-Gen v Luton Local Board of Health (1856) to the effect that the public nuisance claim cannot protect “the Thames, or any public river as it may be termed, which is always given up to be maltreated”. I had gone on to say that if that is true of the Thames, then how much more that must be the case with the global atmosphere.
Bullock argues, if I get his point right, that Page Wood V-C made that statement only because s 145 of the Public Health Act 1848 had permitted discharges of sewage into public rivers, but not private ones. But if this is Bullock’s point, I think he is wrong.
The 1848 Act and similar legislation did not permit the creation of a nuisance in respect of public rivers. Quite the opposite; the legislation preserved the law of public nuisance, including in relation to public rivers. As it happens, the UK Supreme Court recently discussed the position under the 1848 Act in United Utilities Water Ltd v Manchester Ship Canal Co Ltd [2024] UKSC 22, [2025] AC 761 at [23].
I did not intend to suggest in my first piece that no public river could be the victim of a public nuisance, although ordinarily the correct plaintiff is the Attorney-General and proceedings are subject ultimately to Crown control. I would, accordingly, accept that Page-Wood V-C may have overstated the position by referring to all public rivers in the above quotation. But the size and state of the Thames was apparently different, as might be other large rivers with multitudinous minor polluters along them.
Bullock then cites two cases that successfully established a public nuisance in relation to the Thames. I think one should concede that a particularly egregious discharge into the Thames could be found to be a public nuisance. But again the state, either by prosecution or the Attorney-General by civil proceeding, would be the correct plaintiff. In neither of the two cases Bullock relies upon was the proceeding brought by a private citizen under the “special damage” rule.
The pollution case Bullock cites, R v Medley (1834), was a prosecution for the crime of public nuisance. The other case, Att-Gen v Johnson (1819), was a relator action brought in the Attorney-General’s name. The case involved the creation by the defendant, a private citizen, of a major obstruction on the Thames, called a “purpresture”, that was interfering with public rights of movement. The case was therefore not, as Bullock says, a pollution case involving the sanitation operations of the conservators of the Thames.
Lord Eldon LC noted that, although the Crown could not directly license a public nuisance (without a statute), the certificate of the Attorney-General as to whether something was a nuisance would not be gone behind. On the facts, the Crown did not prevent the relator action from proceeding. It remains the position that the Attorney-General retains control over relator actions (see Gouriet v Union of Post Office Workers [1978] AC 435 at 477-479).
Bullock then cites Att-Gen v Birmingham Corp (1858) against me, saying this was also a case involving a public river. I do not think he is right in that. At least, the facts set out at the beginning of the report say: “The Tame is a private river, and the bed and soil of it alongside of the Plaintiff’s estate were now vested in the Plaintiff.” However, I repeat that I do not argue that public rivers cannot attract claims for public nuisance.
While we are on the Birmingham Corp case, another thing Bullock says about the case seems to me wrong, or at least doubtful. So, Birmingham Corp is one of the authorities that Bullock relies upon for the proposition that it is no defence to public nuisance for the defendant to say that it is only one among a number, or indeed many, other contributors to the nuisance.
The other cases Bullock cites may support his proposition, but I do not think Birmingham Corp does. That issue did not arise in the case. It is true that the sewage in question came from the sewers provided by the authorities by some 300,000 residents of Birmingham. But the act of discharging all the sewage into the relevant river was the act of a single person, the defendant corporation. That corporation was not acting as agent for its citizens. There could be no question of combining the (otherwise lawful) acts of the citizens in allowing the corporation to take their sewage to make the entire citizenry liable. The same can be said of the sewage of, to use the language of the court, the 2200 “inmates” of the “lunatic asylum” in Colney Hatch (1868), a case Bullock also cites.
Are public rights the rights of the state?
Bullock next criticises my suggestion that the public nuisance action is prima facie a right enforceable by the state. Amongst the sources he cites against me is the statement in R v Rimmington (HL 2005) that a public nuisance is an act that obstructs “the public in the exercise or enjoyment of rights common to all Her Majesty’s subjects”.
However, that passage from Rimmington merely describes, at a high level, the scope of the concept of public nuisance. It says little about the standing needed to bring an action, unless the word “rights” be read as giving a personal right. The phrase should not be so read in this context. Notably, the same phrase is used in relation to the crime of public nuisance, which is undoubtedly a state action.
Bullock also cites an influential article of William Prosser (Private Action for Public Nuisance (1966) 52 Virginia LR 997) in support of his views. However, Professor Prosser simply omits to address the possibility that the public nuisance action is a state civil action as well as, where there is special damage, a private action. The focus of Prosser’s article is solely on when individuals have standing to bring a tortious claim.
Prosser goes on to state that when a nuisance “becomes so general and widespread as to affect a whole community, or a very wide area within it, the line is drawn [against individual action]” (at 1015). He had earlier discussed a line of cases where it can be inferred that a large number of people each suffered very substantial pecuniary loss yet could not establish special damage (at 1010). As a hypothetical, Prosser instanced a mountain village being cut off by the negligent destruction of a bridge; no private action would be possible even if one’s livelihood were severely affected.
It would not always follow that the state, through the Attorney-General, might not be able to bring a civil claim in public nuisance in such circumstances. There is every reason why, in relation to public nuisance, the state’s legal rights have not been, and should not be, confined to the use of criminal punishment. Hence, it would have been unsatisfactory not to add injunctive and compensatory rights to the armoury of the state.
Contrary to Bullock’s argument, it seems that the injunction as a remedy for public nuisance was in fact available only to the Crown right into the 19th century. Relator actions were possible, but they could be controlled by the Attorney-General. So, an injunction against the commission of a public nuisance was expressly held not to be available to private litigants in Baines v Baker (1752) Amb 158, 27 ER 105. This was a case where the plaintiff was attempting to prevent a smallpox hospital from being set up near his property. He alleged that, in the face of the proposed hospital, his tenants had departed. Lord Hardwicke LC stated:
If a public nuisance, it should be an information in the name of the Attorney-General, and then it would be for his consideration, whether he would file such an information or not.
It was Lord Eldon LC in Crowder v Tinkler (1816) 19 Ves Jun 617 at 621, 34 ER 645 at 646 who, in obiter dicta, moved the dial in favour of the possibility of enjoining a public nuisance at the behest of a private person. But even here Lord Eldon stated:
Where the subject of complaint is matter of public nuisance, the Attorney General alone can sue; but it is going too far to say, particularly without more materials than can be had on motion, that, if a plain nuisance is attended with particular and special injury to an individual, producing irreparable damage, that individual shall not be at liberty to come here, unless the Attorney General chooses to accompany him.
The question then becomes “what is ‘particular and special injury’?” That question will be addressed in the next section, but it is a fair inference that Lord Eldon did not have in mind the test favoured by Bullock, namely anything more than “a theoretical injury to rights”. In any event, the relator action in the name of the Attorney-General remains to this day the standard process for obtaining an injunction for public nuisance.
Special damage and standing
Bullock next says I am wrong to state that “the concept of special damage is a vague term of art” and to say that the special-damage rule operates as an exception to public nuisance as a state claim. He says “special damage” is a clear and “well settled” concept that merely requires a plaintiff to “show that they have suffered an actual harm to them from the nuisance, not just a theoretical injury to rights”. He cites a Canadian case and two textbooks in support.
Again, I would deny that I am wrong on this. Among the many unsettled points that surround the special-damage rule is whether the test is not more onerous where the claimant is seeking an injunction as opposed to damages for past loss. We have seen already that the remedy of injunction for public nuisance was until the 19th century a state-only remedy. It was then extended to private citizens in their own right but subject to a special-damage requirement. Seeking damages for past loss is apt to be a much less invasive remedy than one that is seeking to enjoin activity on an ongoing basis.
It may well be that a number of academics and some judges favour Bullock’s minimalist test. But that does not get one very far when, plainly generalising, academics are notoriously interested in pushing, and indeed are incentivised to push, the boundaries of their subject, and the appetite for jurisdiction of some judges is avaricious.
I do not suggest the particular case on which Bullock relies, the decision of the Court of Appeal of Newfoundland in George v R (2016), is an example of judicial avarice. This was a class action in public nuisance for deaths and personal injuries alleged to have resulted from the government’s failure to control the numbers of moose (non-native) wandering onto public roads. The court declined to recognise the nuisance.
It is true that the court appears to have accepted the academics’ test. It is not, however, a strong precedent to rely upon. Quite apart from the fact that the relevant statement was obiter because no nuisance was established, one can note that: the court accepted the test without reasoning of its own; special damage was scarcely a relevant factor since the claimants had suffered death or personal injury which will usually qualify for special damage; and it was not an injunction case.
Moreover, the court quoted, without apparent disagreement, Professor Spencer’s comment, in Public Nuisance – a Critical Examination [1989] Cambridge LJ 55 at 74 that “few points in civil law are more obscure than the meaning of ‘special damage’.” The court had also cited the decision of the Supreme Court of Canada in British Columbia v Canadian Forest Products Ltd 2004 SCC 38, [2004] 2 SCR 74 for the proposition that “the primary right to sue [for public nuisance] lay with the Attorney-General”. That scarcely seems consistent with Bullock’s thesis.
In fact, the Supreme Court in the Canadian Forest case said a number of things about public nuisance that are difficult to square with Bullock’s views. In this case, the defendant was held to have committed a public nuisance by starting a forest fire. It was confirmed that the Crown could recover damages for its own forests, but there was also obiter discussion as to whether the Crown could recover compensation for damage to non-commercial public land.
The defendant argued that the Crown’s rights as public enforcer were limited to obtaining an injunction and did not extend to damages. This argument was rejected by Binnie J, giving the lead judgment, who stated ([63]-[68]):
[63] The Crown sued and recovered compensation in the courts of British Columbia in its capacity as landowner of the Crown forests. [64] Can the Crown also sue as a representative of the public to enforce the public interest in an unspoiled environment? If the Crown cannot do so, who (if anyone) can?….[66] Historically, of course, the Attorney General, representing the Crown, has been the appropriate party to sue for abatement of a public nuisance… [68] It is true that the role of the Attorney General has traditionally been to seek a stop to the activity that is interfering with the public’s rights. This has led to a view that the only remedy available to the Attorney General is injunctive relief. Some commentators regard the injunction as the “public remedy” obtained by the Attorney General, while damages are a “private remedy” available to those private citizens who have suffered a special loss such as personal injury or damage to private property…The reality, of course, is that it would be impractical in most of these environmental cases for individual members of the public to show sufficient “special damages” to mount a tort action having enough financial clout to serve the twin policy objectives of deterrence to wrongdoers and adequate compensation of their victims: Bazley v Curry, [1999] 2 SCR 534. Class actions will have a role to play but, as Professor Klar notes, “[w]hat has made public nuisance a particularly ineffective private law remedy is the special damages requirement” (L N Klar, Tort Law (3rd ed 2003), at p 647).
When one turns to look at English law after Crowder v Tinkler (above), one finds relatively little discussion of the special-damage requirement in injunction cases. This is not because the issue was seen as unimportant, but rather most cases for an injunction continued to be brought as relator actions, which the Attorney-General could, in principle, control.
In cases where the remedy sought was damages, one finds that the 19th century cases rather see-sawed on what was required for “special damage”, though as far as I know none came close to saying that anything more than a theoretical loss will suffice.
There was curial debate as to whether mere damage to the economics of a particular business, as opposed to adverse effects on the value of the claimant’s land, could qualify as special damage. A narrow approach to that issue was taken by the House of Lords in Ricket v Metropolitan Railway Co (1867) LR 2 HL 175. But even where a broader approach had been applied, as in Wilkes v Hungerford Market Co (1835) 2 Bing (NS) 281, 132 ER 110, it was thought the position may be different if a whole street of shops had had their trade adversely affected (see the judgment of Bosanquet J at 297-298). The judges spoke of “peculiar damage” not just “more than nominal damage”. Wilkes, an obstruction case, was disapproved in Ricket but has since had something of a revival.
I accept that the much more recent English case that Bullock cites, Jan du Nul (UK) Ltd v NV Royal Belge (2000), appears to take a more lenient approach to the special-damage requirement. But the context of that case is important. Again, it was not an injunction case. The relevant nuisance, siltation resulting from dredging a river, involved unwanted physical accretions to the land of a number of people. Undoubtedly a public nuisance had occurred.
The claim was one under a liability insurance policy, where the creator of the nuisance was suing its insurer to cover the costs of fixing the problem. There was no need for a broad approach to be taken to the special-damage issue. Since some persons had certainly suffered special damage, remedying the problem needed to be done just to make them whole, irrespective of how many other members of the public were affected. Moore-Bick J also stated (at [43]): “In the end the question whether the plaintiff’s injury is sufficiently ‘special’ and ‘direct’ must depend very much on the facts of the case”. Again, that is not Bullock’s test.
In my submission, the special-damage rule has always had a major role in keeping private proceedings for public nuisance under some sort of control, particularly injunction proceedings. The rule’s scope and application will vary with the type of activity in question and the type of loss. The need to have a means to control private suits is all the greater when the action for public nuisance has such vague boundaries, extending to “annoyances” to the public, and acts that “endanger the life, health, property, morals or comfort of the public, or obstruct the public in the exercise or enjoyment of rights common to all Her Majesty’s subjects”, to adopt the definition used in the George case. A similar definition is found in the 2026 Act.
In relation to climate change, the financial effects of global warming are vastly more ubiquitous even than Prosser’s hypothetical of the town cut off from the outside world. Not only is all coastal land apt to be adversely affected, but drought and heat may devalue inland land and adversely affect many inland businesses. Under Bullock’s test, not only might a very large number of people be able to prove “a more than theoretical injury to rights”, but any single one of them can seek to enjoin all businesses in the country that are not net-free of carbon and methane emissions.
A strong special-damage requirement is not needed just to protect the courts from a multiplicity of claims, but also to hinder judges from blundering into tackling, without government oversight, environmental or other problems that are beyond their competence to deal with.
Public nuisance: tort and crime
I can be rather briefer with the next section of Bullock’s argument. I had argued that New Zealand’s replacement of the common law crime of public nuisance with a statutory one did not free the civil action from being a public, indeed state, claim. Bullock has argued that I was advocating for “a false mutual exclusivity between criminal and tortious wrongdoing for the same conduct”.
I was not making such an error. I was simply arguing that there was no reason to treat the abolition of the common law crime of public nuisance as the basis for revisiting the nature of the civil action. I reiterate my argument that at heart, and not just in its origins, the civil action is not a normal private law action. As seen above, the action has never freed itself from its public law roots. If the special-damage rule were as relaxed as Bullock says it is, the relator action, still a mainstay, would have shrivelled on the vine a long time ago.
It may be that, as a matter of the history of civil procedure, the Crown’s civil claim for public nuisance, including the relator action, was not strictly a tort, whereas the direct private claim for damages was based on the action on the case. However, the tortious claim also had a special-damage requirement. With the statutory abolition in the 19th century of the forms of action, these procedural differences may have ceased to matter. One tends now to think of a single civil action in public nuisance.
Multiple defendants
Bullock next challenges my proposition that in ordinary private law proceedings, a valid claim will have a finite number of actual or potential defendants. He says what this argument ignores is that the authorities are replete with examples of claims in public nuisance for pollution arising from the contribution of many parties and it is not necessary for the plaintiff to sue them all. Then he suggests I have argued that it is unfair that a plaintiff can choose to sue only some emitters.
I think Bullock here miscomprehends what I was, and am, saying. I do not deny that a plaintiff who has suffered special damage as a result of a public nuisance of a recognised type is entitled to choose who, and how many, among many polluters in the relevant public domain that the plaintiff sues. Nor did I suggest that a party who commits a public nuisance can defend the claim solely on the basis that others, even many others, have committed the same or similar wrongs to the plaintiff.
I would, however, say three things. First, each contributor to the relevant pollution must have committed a prima facie wrongful act (such as pouring blood and other animal products into a stream). At least in the absence of a conspiracy to injure, two or more lawful actions cannot in combination become a wrong.
Second, in deciding whether some new type of activity can amount to a public nuisance it is relevant to consider how widespread the activity is and what the consequences of stopping it would be.
Third, and this is the point I made in my earlier article, where Parliament is intervening to make clear that certain conduct is not wrongful at common law, or that certain wrongs are no longer to be actionable, it would indeed be unfair for Parliament to withhold the resulting freedom from people who just happen to have been sued in the past or are presently being sued.
I want to elaborate a little on the first two of these points. The highly variable nature of public nuisance entails that one has to consider each time whether a type of act constitutes a nuisance. For this reason, it cannot validly be argued that the proceeding in Smith v Fonterra involved the simple application of established principles. The claim was novel. What was being asserted is that it is prima facie wrongful (with exceptions perhaps for de minimis instances and defences for offsets) to put carbon and methane into the atmosphere. These are not established types of wrong.
That there is no established wrong is not in itself a reason for letting a trial to go ahead to test the question. Nothing that the plaintiff could say in Smith v Fonterra, in my view, could make it sensible for the courts to declare, without legislative backing, that the emission of carbon or methane into the atmosphere is prime facie unlawful. This is so even if it were only businesses that were to be liable and they were each entitled to bring into account any offsets they may have.
One needs to take account of how the action for public nuisance operates. It is clear from the authorities that if something is found to be a nuisance, the relevant conduct is immediately deemed unlawful, both retrospectively and for the future. That off-setting activities might be taken into account can operate only as a defence. In other words, it is emissions that are the nuisance, not net-positive emissions.
The courts may then withhold an injunction from the plaintiff while the defendant takes steps to cease its unlawful activity. But the conduct remains unlawful. Ultimately, therefore, the public-nuisance solution to climate emissions involves, on the one hand, the prospect of enforcement through the blunt law of contempt of court and the granting of indulgences, on the other. Parliament, in contrast, has a much wider armoury of responses. These include the ability to discriminate amongst industries, to set clear time limits by which conduct will become prohibited, to specify known penalties, and to use taxes (and apply them to suppliers rather than emitters) and incentives.
This leads to the second of the points. Carbon and methane emissions are ubiquitous and at this point intrinsic to the economy. It is quite unrealistic for the courts to declare that the commercial production of ruminant meat is unlawful, necessary as that might in due course become if solutions cannot be found. Merely delaying enforcement does not solve the overreach. The same can be said for declaring unlawful the use of aeroplanes for tourism, whether in-bound or outbound. This is to say nothing of all the other businesses with significant such emissions. Yet this was the logic of Smith’s claim. It would be even worse for the courts, as opposed to Parliament, to involve themselves in picking and choosing which businesses, and which business activities, are lawful and which not.
This is all the more obvious when it is known that global warming is a global problem and that even if the courts were to stop all emitters of carbon and methane in the country, that would not stop the warming of the climate in the country. The New Zealand courts cannot sensibly extend their jurisdiction to emitters overseas.
Ironically, if the courts are to have a role it might best be used in ensuring that the executive does not itself attempt to use the public-nuisance action to short-circuit public participation, via Parliament, when decisions are made. To use my earlier examples, we would not want the Attorney-General seeking to stop the commercial farming of cattle and sheep, or in-bound and outbound tourism, without involving Parliament.
If it is not obvious enough, my concern in the current debate is to deprecate the courts from involving themselves in matters that are beyond their competence. I am a supporter of strong governmental responses to climate change under legislative mandate. This should extend, for instance, to the protection of urban trees. I am not in the pocket of any big business, think tank or political party.
De minimis
The last topic Bullock takes issue with is my argument that ordinarily the common law draws no distinction between the small and large, the rich and poor, and corporations and individuals. He points out that at common law, users can take water from a stream for domestic purposes when large users wanting to use it for major irrigation or manufacturing cannot. Only the larger users are subject to the public nuisance claim.
I accept that Bullock is right to point this out. However, I consider the generalisation I made holds true, including in relation to public nuisance. Once an activity is viewed as prima facie wrongful, such as introducing blood and offal or other pollutants into streams, it is wrongful for anyone to do it. I consider it is only on this basis that one can (partially) defend the view of Kay J in Blair v Deakin (1887) that a nuisance is committed where “a number of persons contribute what I will call in each case an infinitesimal share to that which becomes in the aggregate a grievous nuisance.” The de minimis concept does not operate here.
Insofar as Kay J was indicating that two non-wrongs can in combination become a wrong, I respectfully consider he was in error. I am aware that other dicta also suggest a non-nuisance can in combination with other non-nuisances become a nuisance. However, the better view is that each person’s actions need to be independently unlawful. Most, perhaps all, cases where these dicta are to be found could have been decided on the presence of independent wrongs. So, in Sadler v Great Western Railway Co [1895] 2 QB 688 it was prima facie wrongful for anyone to park carts on the highway even if, initially, the claimant could negotiate around them. The judgment of AL Smith LJ in that case was correct on this point.
In the absence of a conspiracy, it is hard to see how lawful action can become unlawful because of the actions of others over whom one has no control. The opposing view is contrary to general principles in the law of wrongs (see, for instance, Crofter Hand Woven Harris Tweed Co Ltd v Veitch [1942] AC 435 at 487; and JSC BTA Bank v Ablyazov (No 14) [2018] UKSC 19, [2020] AC 727 at [7]), and also not easy to reconcile with the rule that a downstream user of water has no action where the lawful extraction of water from a stream by a series of upstream landowners collectively causes the stream to run dry (McCartney v Londonderry & Lough Swilly Railway Co Ltd [1904] AC 301 at 306).
Applying the foregoing reasoning to carbon emissions, I accept that using domestic fires for heating would not be actionable and that would remain so even if, collectively, pollution results. But why would the use of the combustion engine, including to undertake recreational travel, not be caught by the climate-change tort where no, or an inadequate, offset is undertaken by the traveller?
Conclusion
In conclusion, I reiterate the basic argument I made in my first contribution to this debate.
It is not inappropriate for Parliament to have precluded the use of the action of public nuisance as a means of making prima facie unlawful the commercial emission of carbon and methane into the atmosphere. Such emissions engage no established common law wrong.
Moreover, it was right for Parliament to have included Smith and any other existing claimant within the preclusion. First, there was little to stop Smith from adding all net-positive emitting businesses in the country to his existing proceeding, thereby thwarting Parliament’s aim. Second, it would be unfair to leave the currently named defendants as the only ones whose businesses might be declared unlawful.
Peter Watts KC
Peter Watts KC: Senior Research Fellow, Harris Manchester College, Oxford; Emeritus Professor, University of Auckland; barrister, Bankside Chambers.
0 Comments