I. A Brief Exposition of Dworkin’s Rights Thesis

I turn to a brief description of Dworkin’s claims relevant to this inquiry. Dworkin’s rights thesis is, in large measure, a reflection of the way he sees the contrast between courts on the one hand and legislative bodies on the other. The operation of legislative bodies, he readily accepts, characteristically involves policymaking.^{Footnote 14} This is a normal and legitimate part of their institutional role and responsibility. That is not the case, however, with decisions made by courts; decisions of this type, according to him, characteristically are and should be decisions of principle, not policy. It should be added straight away, however, that the general idea of courts as a forum of principle takes, according to Dworkin, somewhat different shapes depending on whether the relevant judicial decision is common law or statute-based. This variation will be specified below. Before that, however, the exact meaning of “policy” and “principle” in Dworkin’s theory will be spelled out.

Arguments of policy are described by Dworkin as arguments that justify a political decision by showing that the decision promotes or safeguards “some collective goal of the community as a whole”.^{Footnote 15} Thus, for example, an argument calling for the introduction of a public subsidy for aircraft manufacturers on the grounds that it would contribute to national defence, or an argument calling for an increase in government spending on infrastructure on the grounds that it would stimulate the economy, are both arguments of policy.^{Footnote 16} Arguments of principle, on the other hand, are described by Dworkin as arguments that justify a decision by showing that it “respects or secures some individual or group rights”^{Footnote 17} (hence the label “the rights thesis”^{Footnote 18}). At the heart of any argument of principle, then, lies the supposition that the party on whose behalf the argument is made has a right to the outcome advocated by the argument—that it is entitled to it. It should be clarified that “rights” in the sense referred to here encompass not only the kind of human rights typically enshrined in a constitution,^{Footnote 19} but any individual entitlement the law can be deemed to have conferred on the party in question.

It is also worth clarifying that, according to Dworkin, although arguments of principle are distinct from arguments of policy, they may arise from an earlier policy decision. Consider, for example, a statute instituting a scheme whereby all development projects of renewable energy are entitled to subsidies. The decision to enact the statute was based on environmentalist policy, but if, after the enactment, a renewable energy developer (call her Claire) claims a subsidy provided by the statute, she can do so, and most probably will, by recourse to an argument of principle. Thus, Claire’s argument will not be that granting her a subsidy would help protect the environment (although this is likely to be consistent with her views), but rather that she is entitled to a subsidy pursuant to the statute. In Dworkin’s words (referring to an analogous example), her “right to a subsidy no longer depends on any argument of policy because the statute made it a matter of principle”.^{Footnote 20}

Another factor that helps explain the difference between policy and principle, according to Dworkin, is their relation to past decisions. When it comes to decisions based exclusively or predominantly on policy, the requirement for consistency with past decisions is relatively weak.^{Footnote 21} The primary question policymakers ought to ask themselves is what type of scheme would best serve the goals and welfare of society, not what type of scheme is most consistent with past decisions.^{Footnote 22} In decisions based on principle, on the other hand, there is an integral and relatively strong requirement of consistency with past decisions. A judge, for instance, characteristically does not (and should not) approach the resolution of a case as if it were simply an answer to the question: what decision would best serve society as a whole? The primary question he or she is meant to be answering, instead, is which of the parties before the court has a legal right to win the case. And one of the principal grounds for determining the parties’ legal rights is what Dworkin calls the relevant institutional history—primarily, past legislative and judicial decisions—and the requirement that like cases be treated alike.^{Footnote 23}

It is not suggested here that the sense in which Dworkin uses the term “principle” is the only sense found in common, legal, or theoretical discourse.^{Footnote 24} Quite possibly there exist other senses in some or all types of parlance—senses in which a decision might be said to be “principled” without implying that it is grounded in individual or group rights, and without excluding “policy” from its direct underpinnings. Indeed, the idea that a decision should best promote or safeguard “some collective goal of the community as a whole”^{Footnote 25} might itself be described as a type of principle in some broad sense of the word.^{Footnote 26} The sense of “principle” I am focusing on, however, is the particular sense defined by Dworkin in his above-cited statement. That is not because I think it represents an understanding of “principle” generally superior to others, but because this sense of “principle”, as contrasted with “policy”, highlights what I consider to be a genuine and important tension in adjudication that other, broader senses of principle may obscure.

As mentioned earlier, Dworkin’s conception of courts as a forum of principle takes somewhat different shapes depending on whether the relevant judicial decision is common law or statute-based.^{Footnote 27} When it comes to common-law decisions, according to Dworkin, policy arguments should be entirely precluded from the court’s reasoning process: “[j]udges must make their common-law decisions on grounds of principle, not policy”, he states.^{Footnote 28} Judicial interpretation of statutes, on the other hand, may legitimately take into account policy considerations relevant to the background justification of the statute in question.^{Footnote 29} As Dworkin puts it at one point, the judge should fix to the relevant statutory language “arguments of principle or policy [emphasis added] that provide the best justification of that language in the light of the legislature’s responsibility”.^{Footnote 30} Dworkin believes, however, that even when a judge makes this type of appeal to policy, the judge’s reasoning ultimately “remains an argument of principle”, for “he uses policy to determine what rights the legislature has already created”.^{Footnote 31} I am not certain, however, how to coherently understand Dworkin’s position on this matter. “Policy” and “principle”, as he seems to understand them, are mutually exclusive categories. How, then, can a statutory interpretation which, by Dworkin’s own admission, is sensitive to, and influenced by, policy considerations end up being classified by Dworkin in the category of “principle” just because the judge, as he puts it,^{Footnote 32} “uses policy to determine what rights the legislature has already created”? But even if Dworkin coherently accommodates policy in his conception of statutory interpretation, this cannot settle the matter I seek to address herein—i.e., the apparent conflict between the rights thesis and judicial resort to the FA—for the simple reason that the FA sometimes, or even mostly, features in common-law decisions.^{Footnote 33}

II. Is Judicial Resort to the FA Compatible with the Rights Thesis?

On the face of it, the FA appears to be a clear example of a policy argument, which, if the rights thesis is correct, has no place in civil adjudication. The FA does not focus on the conduct of the parties or the consequences of their conduct. It is not based on the merits of the claim itself in view of what the parties did or caused. What it says, instead, is that even if the lawsuit consists of a meritorious claim, it should be rejected in view of the risk of proliferation and excess of future lawsuits. The argument’s central focus, then, is the potential volume of future cases and the demands it would place on judicial resources, which is a consideration exogenous to the merits of the claim itself and not specifically linked to the parties in question. To this extent, the FA also seems to be in tension with the requirement that like cases be treated alike. For it follows from the FA that two situations substantively similar in all respects, and thus giving rise to equally meritorious claims, may nonetheless be treated differently on the sole ground that in one situation the group of potential future claimants happens to be larger than in the other.^{Footnote 34}

However, as mentioned at the outset, Dworkin himself considers his rights thesis to be reconcilable with judicial resort to the FA. Their reconcilability becomes visible, according to Dworkin, when one brings to mind a distinction between abstract rights on the one hand and concrete rights on the other.^{Footnote 35} An abstract right, Dworkin clarifies, “is a general political aim the statement of which does not indicate how that general aim is to be weighed or compromised in particular circumstances against other political aims”.^{Footnote 36} A concrete right, on the other hand, is a political aim “more precisely defined so as to express more definitely the weight [it has] against other political aims on particular occasions”.^{Footnote 37} Dworkin acknowledges that this distinction represents what is merely a difference in degree, but he does not seem to think this undermines its serviceability. He furnishes examples of rights that, he believes, lend themselves to classification in terms of that distinction. Thus, for example, the statement that “citizens have a right to free speech”^{Footnote 38} refers to an abstract right, whereas the statement that “a newspaper has a right to publish defense plans classified as secret provided this publication will not create an immediate physical danger to troops”^{Footnote 39} refers to a concrete right. According to Dworkin, when judges in a civil dispute translate the rights of the parties from abstract rights into concrete ones, they should (and do) take into account the rights of potentially affected people who are not party to the dispute. Thus:

Dworkin’s comment is framed in somewhat inconclusive terms: “might be understood”. I am not sure, however, what the word “might” means here exactly, and, since the above characterization of the decision seems essential to Dworkin’s defence of the rights thesis, I take him to endorse this characterization. Now, judicial resort to the FA is, according to Dworkin, an example of this form of balancing between competing rights:

I find Dworkin’s attempt at reconciling the FA with his account to be unpersuasive. Dworkin’s basic claim, his rights thesis, excludes policy from civil adjudication. Then, by way of clarification and qualification, the passages just quoted say: but judges in civil disputes may and sometimes do engage in a balancing exercise between, on the one hand, the disputants’ rights and, on the other hand, the more remote rights of persons (members of a general class of unnamed individuals, e.g. potential users of a public resource) who are not parties to the dispute. This qualification, however, essentially introduces through the back door much of what the thesis initially appeared to exclude. For there are many public policy goals that can be expressed in terms of rights of individuals comprising the public: public safety can be described in terms of people’s rights to life and security; healthcare policy can be couched in terms of individuals’ right to a standard of living adequate for health and well-being; and a public aim of reducing unemployment rates can be expressed as a desire to protect people’s right to work. The continuing expansion of human rights law over recent decades offers further indication of the potential compass of Dworkin’s qualification and, thus, the extent of its discordance with his basic claim.^{Footnote 42}

I do not wish to deny that public policy goals such as the ones just mentioned are genuinely connected to rights. When Tom sues Bill for assaulting him, the legal right Tom is invoking has of course some relation to the policy goal of public safety, in that, for example, respect for this right is likely to contribute positively to public safety. But there is a crucial difference between this situation and a situation in which a court considers possible future implications of its judgment for the safety of a general class of unnamed third parties—as, for example, when a court, in adjudicating a dispute between a police officer and his or her employer over the terms of employment, incidentally takes into account possible future implications of the judgment for the safety of potential users of police services. The best way of capturing this difference is to say that in the latter type of situation considerations of safety figure in the court’s reasoning as a public policy concern. In denying just that, Dworkin offers what is not only a far less intuitive characterization of the situation, but also one that blurs a normatively important distinction between two different functions of the court: that of adjudicating a specific dispute between particular parties in accordance with the law, and that of helping to steer the law in desirable directions in terms of its wider implications for a larger class of people, members of society at large.^{Footnote 43} As will become clearer in the following section, I do not deny that these two functions are sometimes simultaneously entangled in a single judicial decision. But this does not make them one and same, nor does it speak in favour of adopting a conceptual construction that only makes it harder to tell the difference between them. On the contrary, it reinforces the reasons for an analytical framework that tends to facilitate awareness of, and sensitivity to, the distinction.

To conclude this section, Dworkin’s characterization of judicial resort to the FA as a policy-free decision seems to me to be far less cogent than the alternative, that is, a plain acknowledgement that judicial resort to this argument is an exercise of reasoning involving policy. I remain, therefore, of the view that the FA and the rights thesis are in conflict. Now, if this observation is accepted, the next likely question is: How should the conflict be resolved? Or, more specifically, should we endorse the rights thesis, and thus conclude that judges should not resort to the FA? Or, should we instead approve of judicial resort to the FA, and thus reject the rights thesis? I will advocate an intermediate approach. I will argue that, while Dworkin’s thesis contains an important insight, it is making a hyperbolic claim that requires considerable qualification. Applying these observations to the specific question of the FA, I will put forward an operative suggestion as to the proper conditions for, and constraints on, judicial resort to the FA.

III. The Rights Thesis: Insight and Hyperbole

I will start with what I think is the insight in the rights thesis. One important fact that the thesis usefully highlights is that a judge characteristically does not approach the task of deciding a case by asking the question: would it be best for society if the claimant wins the case, or if the defendant wins? Rather, the judge characteristically embarks on the task by asking, firstly, what the law is on the matter at hand,^{Footnote 44} and which of the adversaries is entitled to win the case in accordance with the law. Unlike Dworkin, I am inclined to think that questions of policy may nonetheless feature in the decision, but—and this is a fact which, I believe, Dworkin’s account usefully casts light upon—they are not the starting point of judicial reasoning. The characteristic starting point of judicial reasoning consists, instead, in an effort to derive the current decision from norms founded (at least in part) on past legislative or judicial decisions, and an attempt to draw a coherent link between these decisions.^{Footnote 45}

That judges approach their task in this way is no mere happenstance. It is a method of decision embedded in their understanding of the role of a judge and rooted in some of the basic values underpinning the function of law in social life. Thus, for example, adherence to the requirement that like cases be treated alike helps render adjudication more impartial, and introduces a measure of fairness between litigants. A commitment to coherence serves to reduce the degree of arbitrariness that judicial decisions would otherwise involve, and makes an essential part of law’s commitment to reason.^{Footnote 46} And the judicial effort to ground the present decision in past decisions minimizes, as far as possible, elements of retroactive imposition by law—which echoes one of the central desiderata of the rule of law.^{Footnote 47} Living up to these cherished standards is not just a worthy aim, but is in fact a prerequisite for the legitimacy of a legal system and its entitlement to demand citizens’ compliance with its courts’ rulings.

Dworkin is also correct, in my judgment, to point out that even when past legislative and judicial decisions are ambiguous and varied enough to support more than one possible outcome, adjudicative discourse does not simply cut loose from its aspiration to a principled judgment on the parties’ entitlements.^{Footnote 48} Even then lawyers and judges do not simply discard the legal materials as irrelevant. They continue to consider and debate distinctions and analogies between the present case and preceding ones, and persist in their attempt to show a coherent and principled link between their decision and what has been decided in the past. To appreciate this fact, it suffices to consider the sheer proportion of discussion devoted to past cases in even the most path-breaking tort judgments, such as those in Donoghue v Stevenson,^{Footnote 49} Hedley Byrne v Heller & Partners,^{Footnote 50} McLoughlin v O’Brian,^{Footnote 51} MacPherson v Buick Motor Co.,^{Footnote 52} and Sindell v Abbott Laboratories.^{Footnote 53} Many times judges follow that mode of reasoning without explicitly stating so. At other times, however, it receives explicit manifestation in the judgment. Thus, for example, in Hunter v Canary Wharf [1997] AC 655, a leading English case in the area of private nuisance, Lord Hoffmann remarks:

And in Fairchild v Glenhaven Funeral Services Ltd [2003] 1 AC 32, a notable case in the context of causation in negligence, Lord Nicholls comments:

What, then, is the exaggeration in Dworkin’s rights thesis? The exaggeration is his denial that anything of what judges characteristically do and should do in the relevant legal areas is properly described as policymaking.^{Footnote 56} Although policy arguments should generally not be the starting point or primary component of judicial reasoning, to deny that they have an essential role to play in adjudication is a step too far. There are cases that inextricably involve wide social and economic implications that are just too significant to be disregarded by the court—and consideration of those wider implications for society is most readily understood as an exercise that introduces an element of policymaking into the decision.^{Footnote 57} Dworkin, as noted in the preceding section, did not suggest that courts should never consider those wider implications of their decisions, but he did insist that their doing so can be understood as an exercise involving only arguments of principle and no policy. As argued in the previous section, however, his attempt to square those instances with his rights thesis extends his notion of “principle” to the point of obscuring even the element of truth in the thesis. It is better, instead, to acknowledge that such instances of judicial reasoning involve an element of policymaking.

Tort law, an area of law particularly pertinent to this inquiry,^{Footnote 58} offers some of the clearest illustrations of the essential part of policymaking in adjudication. If tort law was reducible to a set of definite norms, framed in precise terms which lend themselves to clear-cut application, much more (though still not all) of the policymaking involved in this area could be left to the legislature. But the competing values and interests underpinning, or bound up with, tort law are often finely balanced and fact-sensitive to a degree that necessitates the use of open-ended and abstract normative concepts—e.g. “foreseeability”, “proximity”, and “fair, just, and reasonable”,^{Footnote 59} “non-natural use”^{Footnote 60}, or the like—which gradually come to have more concrete content through their interpretation and implementation in court judgments, but which also retain a measure of elasticity and indeterminacy in their application.^{Footnote 61} A court’s interpretation and implementation of such normative terms may, in turn, have significant impact on the shape of law with wide-ranging social and economic implications. Thus, when a court determines the limits of liability by reference to the duty tests in negligence,^{Footnote 62} the elements of private nuisance, or the rule in Rylands v Fletcher, its decision may have knock-on effects such as driving up insurance premiums and imposing precautionary burdens that may inhibit socially valuable or even essential activities and/or affect the price of related services or products. If, for example, a court is to hold a local housing authority liable for failing to prevent a resident’s assault on another resident, future forms of overly defensive behaviour of housing authorities in the operation of estates may result;^{Footnote 63} and if the court is to hold a non-negligent apartment block owner liable under the rule in Rylands v Fletcher for damage caused to a gas supplier by escaping water kept on the premises for domestic use, landlords may have to scale up their insurance to cover risks beyond those normally associated with domestic or commercial plumbing;^{Footnote 64} and so on. In such cases, judges must consider policy lest their decisions steer the law—and social and economic affairs dependent on it—in consequentially wrong directions.^{Footnote 65}

To be sure, adverse social and economic policy repercussions of a court decision are something that might be set right or mitigated by a later legislative action. And the knowledge that such corrective action can be taken should arguably make it easier for judges to focus their attention on legal principles and leave wider policy concerns to the legislature. I do not wish to dispute the validity or relevance of this last consideration. But it must be borne in mind that its force will be limited if and insofar as there are grounds to suspect that no legislative action would actually be taken or that the time lag of a statutory reaction would be considerable, whether in view of present legislative priorities, a backlog of parliamentary work, or other factors.^{Footnote 66} With this in mind, it seems unwarranted to assign invariable force, let alone invariably conclusive force, to the potential of ameliorative statutory response.

The observations made in the previous paragraphs combine to support the following conclusion: although the primary task of civil judges is to render principled decisions on the parties’ entitlements according to the law, policy arguments sometimes do and should feature in their decisions. Arguments of principle and arguments of policy both have essential roles to play in civil adjudication. The need to resort to policy arguments, in at least some judicial decisions, cannot be eliminated, but nor can the reasons that militate against judicial policymaking and call for adherence to principle. Civil law adjudication, and tort adjudication in particular, is an arena of ineliminable tension (or, at any rate, potential tension) between principle and policy.^{Footnote 67}

I wish to consider a possible response to my argument. The response puts forward the following construction of Dworkin’s position by way of defence against my criticism. According to Dworkin, legal interpretation involves an assessment on two dimensions: fit (the degree to which a given interpretation fits the practice) and justification (the degree to which that interpretation presents the practice in an appealing light from an evaluative perspective of political morality).^{Footnote 68} When the assessments on these two dimensions pull in different directions—i.e., when a given interpretation has considerable justificatory appeal but a comparatively low level of fit, or vice versa—some form of trade-off between them may be required:^{Footnote 69} the interpretation adopted should be such that it reflects an optimal balance between “fit” and “justification”.^{Footnote 70} With this in mind, the opponent might suggest that what I argued to be the “ineliminable tension between principle and policy” can actually be assimilated into Dworkin’s conception of the interplay between “fit” and “justification”. On this understanding, the aspect of “justification” in the interpretive exercise takes into account (inter alia) policy considerations for or against a given interpretation of the law, whereas the constraint of “fit” secures the type of consistency with past decisions that characterizes arguments of principle—it turns what would otherwise be an exercise of policymaking into a decision of principle.

The above response is, I suggest, misguided on a number of counts. Perhaps most pertinently for the present analysis, the response misconceives Dworkin’s position on the type of justifications that may appropriately figure in interpretation of common law. Dworkin does not accept that the justifications figuring in this type of interpretation may take into account considerations of policy.^{Footnote 71} As was noted earlier, he distinguishes in this respect the interpretation of statute and the interpretation of common law. Regarding the latter, he unequivocally bars recourse to policy justifications.^{Footnote 72} Thus, for example, Dworkin states that when Hercules, his hypothetical ideal judge, interprets judicial enactments, as opposed to statute, “he will fix to the relevant language only arguments of principle, because the rights thesis argues that only such arguments acquit the responsibility of the ‘enacting’ court.”^{Footnote 73}

Now suppose the opponent concedes the point just made and offers again his or her initial response with the only difference that it is no longer presented as Dworkin’s position. In this variant, the opponent accepts that, by allowing policy justifications to figure in common-law interpretation, he or she has departed from Dworkin’s position. But (as in the original response) the opponent insists that adherence to the “fit” requirement imbues the decision (including any element of policy it would otherwise contain) with a quality of principle. This variant response is no longer accountable as a description of Dworkin’s position, but it is guilty of another mistake. It expects the “fit” requirement to perform a job beyond its ability. Observance of the “fit” requirement cannot wipe out elements of policy that had been taken into account by the decision-maker, nor can it transform them into something that is purely and only an argument of principle. By this I do not mean to suggest that the “fit” requirement does not impose a genuine constraint on the decision-maker. However, in many instances the available body of past precedents exhibits a degree of heterogeneity, incompleteness, or uncertainty that leaves the judge faced with more than one alternative resolution defensible in terms of the “fit” requirement. In such cases, even if the “fit” requirement contributes to the decision, it is illusory to suppose that its contribution somehow eliminate from the judgment policy factors that the judge had taken into consideration.^{Footnote 74}

IV. What Implications Follow for the FA?

Above I have argued for an understanding of civil law adjudication as an arena of ineradicable (potential) tension between principle and policy. As part of this argument, I have highlighted some general (though defeasible) reasons against judicial policymaking. I now wish to utilize these observations as a framework of analysis in addressing the particular problem of the FA. The first thing to note as I approach this task is that, although the reasons against judicial policymaking are general in scope, their force is sensitive to factors that change with the context and the type of policy argument involved. I will draw attention to one such factor which has particular relevance to an analysis of the FA: namely, the question of how tight or loose the connection is between the policy argument considered by the court and the substance of the lawsuit at hand;^{Footnote 75} or, in other words, how remote the argument is from material facts that bear on the merits of the claim. The more remote it is, the wider the departure of the court from what has been identified here as its paradigmatic task—that is, providing a principled resolution to the dispute between the parties—and, thus, the harder it becomes to justify and defend the court’s resort to that policy argument.

Consider an example of a policy argument that exhibits a fairly close connection to the substance of the lawsuit in question. The English case JD (FC) v East Berkshire Community Health NHS Trust ^{Footnote 76} involved parents who brought actions for negligence against healthcare and local authorities, claiming damages for alleged psychiatric harm caused to them as a result of false allegations that they had abused their children. The House of Lords dismissed the parents’ appeal on the basis that no duty of care was owed to them by the defendants. The main reason was the following policy consideration. When dealing with suspected child abuse, the relevant authorities should act in the best interest of the child, and while the child’s interest is normally congruent with the parents’, that is not so where the possibility of parental abuse arises. To recognize a duty of care towards the parent may thus unduly inhibit the authorities in fulfilling their main function and may compromise the protection of children against abuse. Hence, while a duty of care is owed to the children, it is not owed to their parents. This consideration is best characterized as a policy consideration, because it is focused on the collective goal of protecting children from abuse (a goal which does not serve in any direct way the children of the appellants, as the accusations against the appellants proved to be unfounded). At the same time, however, it is directly and inherently connected to substantive characteristics of the lawsuit at hand: namely, the facts that the case involved suspected child abuse, that the defendants were public authorities in charge of dealing with such suspicions, and that the claim was for negligence towards the parents rather than their children. These factors go to the heart of the claim. When a policy consideration is so closely intertwined with substantive and material facts of the claim, it is hard to see how a court could or should sever it from the rest of the claim and disregard it.

Now contrast the above-described policy consideration with the FA. The FA bears a much looser connection, if any, to the substance of the lawsuit in question, and a connection that is far more contingent. The FA, to reiterate, is an argument that calls for the rejection of a legal claim due to the fear of a large volume of future claims placing excessive demands on the court system. This description of the argument includes no reference to the particular facts of a case, and it is a natural description, not one artificially cast at a high level of abstraction to avoid reference to concrete facts. Indeed, the potential of a large number of future claimants can exist in different cases that share nothing in substance. And, vice versa, the potential number of future claimants can widely diverge between two substantively similar cases, with the only difference that, say, one involves a popular and socially widespread activity while the other involves an activity undertaken only by few.^{Footnote 77} The FA’s distinctive lack of connection to the substance and merits of the claim makes the resort to it, ceteris paribus, harder to defend against an attack on grounds discussed earlier, such as the value of substantive fairness and coherence in law.

The following objection might be raised at this point. My claim that the FA is distinctively extraneous to the substance and merits of the lawsuit is mistaken, it might be argued, because when a judge considers the FA he or she is likely to take into account, among other factors, the gravity of the interests or rights involved in the relevant legal action. Thus, for example, when the substantive interests or rights at stake are of relatively little weight, all else equal, the judge is likely to be more receptive to the FA, and vice versa. Now, the first point I should make in reply is that I am quite uncertain about the extent to which the facts stated in the objection are correct—that is, the extent to which judicial response to the FA actually depends on the importance of the interests or rights involved in the claim. One body of decisions that may give rise to such doubts is the English case law on negligence liability for psychiatric injury: the rather stringent limitations on the scope of liability that largely endure in this area of law result, at least in part, from floodgates concerns. Yet, few would deny (at least presently and in recent times^{Footnote 78}) that the claimant’s protected interest, his or her mental integrity, carries great importance, and few would suggest that the type of harm involved, psychiatric illness, is of little significance—as reflected, e.g., in Lord Steyn remarks in White v Chief Constable of South Yorkshire Police ^{Footnote 79} that “psychiatric harm may be more serious”^{Footnote 80} and “may be far more debilitating than physical harm”.^{Footnote 81} Now, the example of psychiatric harm does not go as far as disproving the objection’s factual veracity, because the extent to which the FA has contributed to judicial decisions in this area is itself unclear, and because the proviso used in the objection, “all else equal”, may not have applied in those cases. But the example does, at the very least, place a question mark over the facts stated in the objection.

But let us now suppose arguendo that the facts stated in the objection are correct. Even so, the objection should be rejected on other grounds. There is a crucial difference between the way material facts of a case figure in ordinary judicial reasoning and the way they might figure in judicial consideration of the FA. In ordinary judicial reasoning (in the context of civil disputes) the primary question is whether the claimant has a valid cause of action, and, correspondingly, whether the defendant is liable. If the answer is positive, the claimant is deemed to have a legal entitlement (to a remedy whose exact type and size remain to be determined). The meritorious nature of the claim is not a question of degree or relative importance: the claimant is not required to establish, for example, that his or her cause of action is very important or more important than some other causes of action which other claimants might have, but only that he or she has a cause of action. When it comes to judicial consideration of the FA, on the other hand, matters are very different: here, according to the objection, the court takes into account the relative importance of the rights at stake. This assessment of relative importance forms an unusual hurdle for the claimant, over and above the ordinary need to establish that his or her claim is meritorious. Why is the claimant required to overcome this unusual extra hurdle? Because of the fear of a potential flood of litigation—a reason that has nothing to do with the merits of his or her claim. This means that the claimant in such cases is subject to treatment different from claimants in other cases (where there is no floodgates concern), and the ground for this different treatment is entirely extraneous to the merits of his or her claim. We can see now that, even when the facts stated in the objection are assumed to be correct, there remains a crucial, and problematic, sense in which judicial resort to the FA rests upon extraneous grounds. The objection, therefore, fails to address the normative concerns highlighted here against judicial resort to the FA.

Another consideration is worth noting at this point. Invokers of the FA often seem to implicitly regard the available amount of judicial recourses as static. But is that assumption always necessarily correct? An increase in the volume of lawsuits may prompt a government to respond by channelling more public funds to the judiciary so as to augment existing judicial personnel. To ignore this possibility is, in effect, to put the cart before the horse: namely, to shape the law according to the existing amount of judicial resources, instead of allowing that amount to be changed in response to the law.^{Footnote 82} The intuitively correct order is the latter. Of course, one should not be naïve enough to assume that additional judicial resources would always be made available in response to growing need; and when one has strong enough reason to believe otherwise, it may appropriately inform the background factual assumptions relevant to an assessment of the FA—as will be seen in the solution I will shortly propose. But the basic point remains that, in what generally and intuitively seems to be the sounder direction of reasoning, the amount of judicial personnel should be determined by substantive needs generated by the law, not the other way around.

What, then, is the upshot for the FA? The analysis in this section, if correct, shows that there exist strong normative considerations against judicial resort to the FA. These derive primarily from the general considerations against judicial policymaking, but they carry particular force in the specific context of the FA, whose exogenous nature stands in an especially sharp contrast to values associated with principled adjudication. Does it follow that courts should never accept the FA? No. The considerations highlighted here against the FA are strong, but their strength is a quantitative attribute that remains relative. It does not turn them into absolute considerations, which would entail an approach as sweeping as the proverb: “Let justice be done even though the heavens fall”. And it is conceivable that the pragmatic concern about a looming flood of litigation might, on occasion, be so compelling and well grounded that it counterbalances and even overrides the normative considerations discussed above. It would be irresponsible for the court system to entirely divest itself of the possibility to resort to the FA on those occasions, rare and exceptional though they may be.

These comments gesture to a solution that can conveniently be characterized as a presumption against judicial resort to the FA: a decisional tool that inclines the decision-maker against acceptance of the FA, but, as is normally the case with presumptions, remains susceptible to override when the case for using the FA is particularly strong.^{Footnote 83} In determining whether this presumption should be overridden in a given case, I suggest that specific consideration should be given to the following questions and factors. (1) Overwhelming magnitude: is the volume of additional lawsuits expected to ensue if the court finds for the claimant of overwhelming magnitude (relative to the resources available to the judicial system)? The suggested presumption against the FA should not normally be overridden when the answer is negative. (2) Evidential support: the fear that such volume of lawsuits would ensue should be based not on mere surmise or speculations, but on concrete evidence—for example, factual data indicating the potential number of future claimants, the likely frequency of claims, and the existing volume of claims dealt with by the system. (3) The absence of alternative caseload management solutions: it should be established that alternative caseload management solutions—including, for example, feasible and moderate adjustments to the resources allocated to the judicial system and/or probable recourse to efficient modes of litigation such as class action—are not likely to be effective. The foregoing list is not meant to exhaustively capture all relevant factors, but merely to highlight some particularly relevant ones.

A couple of additional comments should be made before this article is brought to a close. First, it is worth reemphasizing the distinction drawn at the outset between the version of the floodgates argument discussed here (the FA) and other arguments that sometimes appear under, or in connection with, the label “the floodgates argument”.^{Footnote 84} A discussion of those other arguments must await another opportunity, but it can nonetheless be noted here that at least some of those arguments appear, prima facie, to differ significantly from the FA in terms of their normative status. Take, for example, the concern not to impose on the defendant a disproportionately large or indeterminate amount of liability, and the related concern to avoid an undue inhibition of a socially beneficial activity pursued by the defendant. Reliance on such considerations may be easier to justify than reliance on the FA, because the former are closer in focus to the substance of the claim and its material facts. Indeed, they seem to be inextricably linked with factors such as the type and degree of risk generated by the defendant’s activity, whether and how well the defendant is placed to take effective precautions, and whether and how the imposition of liability would affect any valuable function fulfilled by the defendant. Such considerations appear, therefore, to be considerably less susceptible to the objection levelled against the FA due to its exogenous nature.

Second, it is worth distinguishing the objection discussed here against the FA from another type of critical response to it. Critics of the FA have, inter alia, called into question its factual premises, suggesting that the fear of multitudinous claims flooding the court system is often a factually unwarranted or an exaggerated fear. Thus, for example, Deakin, Johnston, and Markesinis have pointed out, among their several different reservations about the floodgates argument, that it

While this type of argument expresses scepticism about the factual component of the FA, the arguments focused on in this article, and thus the main grounds for the suggested presumption, are normative. They are arguments from the role and value of principled adjudication. Having said that, the two types of argument—the factual and the normative—do not exclude one another and may jointly support the suggested presumption.

A final clarification should be made about the relationship between my argument and past judicial treatment of the FA. There are several court judgments wherein different expressions of scepticism have been made regarding the FA and its force as a ground for a decision.^{Footnote 86} The conclusions arrived at here are not at odds with the general tenor of these judicial remarks. Instead of calling for a radically different approach, this article’s main contribution to existing discourse on the FA, as I see it, lies elsewhere and can be summed up as follows. (1) One of the distinctive attributes of my discussion of the FA is that it rests upon a jurisprudential analysis. It thus offers a theoretically underpinned argument that, if accepted as sound, may serve to test, reinforce, change, or refine (as the case may be) one’s otherwise held view regarding the FA. (2) It is hoped that the present analysis may help create a clearer framework of deliberation about the FA and other versions of the floodgates argument. Despite what was said above about past judicial treatment of the FA, the legal position remains somewhat unclear. Part of the uncertainty results from the existence of multiple versions of the floodgates argument, as noted at the outset of this article; and the fact that different versions—which, if my analysis is correct, seem to merit different treatment—may be intertwined in one decision with insufficiently clear and explicit differentiation between them. With this in mind, part of my purpose has been to promote a more discriminative discussion by singling out for a separate analysis one particularly problematic version of the floodgates argument. (3) Scepticism towards the FA can take different forms. Here I have advocated one, which implies a rejection of some others. Take, for example, the stance adopted by certain judges who have cast their disfavour of the FA in absolute terms, such as Holt CJ’s categorical statement in Ashby v White ^{Footnote 87} that “it is no objection to say that it will occasion a multiplicity of actions”.^{Footnote 88} The approach advocated here is different in that, instead of categorically barring the FA, it adopts an overridable presumption against it. It leaves open the possibility of judicial resort to the FA but reserves it to occasions where the case for doing so is especially compelling in terms of the scale of the anticipated caseload, the level of certitude with which it is foreseen, the absence of alternative solutions for the expected problem of case backlog, and other relevant factors.