The Indirect Influence of Politics on Tort Liability of Public Authorities in English Law

Abstract

The scope of negligence liability of public authorities in English law has undergone significant changes in the Post-World War II period, first expanding and then, from the mid-1980s, retracting. This article tries to explain why this happened not by focusing, as is common in most commentary on this area of law, on changing doctrinal “tests,” but rather by tying it to changes in the background political ideology. My main contention is that political change has brought about a change in the law, but that it did so by affecting the scope of the political domain, and by implication, also the scope of the legal one. More specifically, I argue that Britain's Post-War consensus on the welfare state has enabled the courts to expand state liability in accordance with emerging notions of the welfare state without seeming to take the law into controversial territory. When Thatcher came to power, the welfare state was no longer in consensus, thus making further development of legal doctrines on welfarist lines appear politically contentious. The courts therefore reverted back to older doctrines that seemed less politically charged in the new political atmosphere of the 1980s.

In debates on the relationship between politics and law, two extreme views tend to occupy much of the discussion. At one end of the spectrum we see the view that law is simply a masquerade for politics. On this view, doctrinal analyses we find in judicial opinions are mere cover for what is really going on, namely politics. At the other extreme are those who think of law in political terms is to misunderstand what this law is all about (among others: Beever 2007:171–72; Oakeshott 1975:412; Stevens 2007:311).

The former approach has been very influential in analysis of the work of American courts, in particular the Supreme Court. Following on the footsteps of the legal realists who have argued that the legal considerations found in judicial decisions often fail to explain the actual motivation behind the decisions, many scholars of American courts have argued that the judges' politics provide the best explanation of judicial opinions. Thus, according to two leading proponents of this view, “[s]imply put, Rehnquist votes the way he does because he is extremely conservative; Marshall voted the way he did because he is extremely liberal” (Segal & Spaeth 2002:86).

This view is by no means universally accepted, but whatever its merits may be in explaining the United States Supreme Court, it is not clear how applicable it is to the analysis of other courts, especially those outside the U.S. American courts have long been unique in the degree to which they participate in the shaping of policy on politically divisive issues. In England, where legal realism has never had as much impact as it had in the United States (Duxbury 2005:54–69), the alternative, legalist, explanation still largely prevails. Accordingly, the legal reasons found in judicial opinions are still typically treated as the most reliable source for explaining the outcomes of cases and are closely analyzed by legal commentators. This scholarly difference reflects, I believe, an underlying difference in legal and political traditions.

While there have been some efforts to argue that English courts are influenced by political considerations (e.g., Griffith 1997; Robertson 1998), they have been few and far between, and even they have resulted in considerable criticism (e.g., Lee 1988:33–39; Minogue 1978; both directed at Griffith). Even these rare studies focused more on public law, leaving the area of private law mostly to doctrinal scholarship, which is still the dominant form of scholarship among English private lawyers. The question whether and how political ideology affects these areas of law was either unexamined,¹ or explicitly denied (Stevens 2009:651–52).

A full explanation for this difference is beyond the scope of this article, but I suspect it has to do with the different political traditions of the two countries. In England, the doctrine of Parliamentary supremacy was understood to imply that political questions are the domain of Parliament (Tomkins 2005), and as a result courts, including the House of Lords (until recently the highest court for civil cases from all the United Kingdom), have shied away from involvement in politically controversial matters (Atiyah & Summers 1987:267–68).² It is, I believe, in large part for this reason why legal realists have not had similar impact in England as they have in the United States. (After all, virtually all the important ideas associated with legal realism can be found in the work of Jeremy Bentham, who was by no means an obscure figure.) Though most of the realists said relatively little on politics, their ideas were more congenial and could be further developed in an environment in which law was increasingly entangled in political controversy.

It is this gap that this article seeks to begin to fill by examining the effect of politics on English law in one particular area—negligence liability of public authorities. In the years following World War II we see a trend toward expansion of negligence liability of the state. Starting from cases in which state agents caused harm, later cases have imposed liability on the state when it failed to supervise or prevent others, typically private individuals, from causing harm. Then, from about the mid-1980s, we see a reversal of the trend towards limitation of liability.³ Though there is vast doctrinal literature on the topic of negligence liability by the state, most of it seeks to explain changes in the law by focusing exclusively on analyzing legal concepts. The primary aim of this article is to show the limitations of doctrinal analysis in this area and to offer an alternative explanation. I do this by highlighting several aspects of the leading decisions in this area of law that received relatively little attention and explaining them in light of competing political views on the relationship between individual and state.

The second aim of this article is more general. Although the specific ways in which law and politics interact are probably unique to every jurisdiction, what I purport to offer here is a different approach to thinking about the way politics can influence law. According to the more common explanation, captured neatly in the quote from Segal and Spaeth above, the political ideology of each individual judge influences her legal analysis. Even those who reject the sweep of this position are mainly concerned to show that alongside political ideology, legal doctrine has real impact on the outcome of cases. The debate is mainly about the relative impact of legal doctrine and political ideology on legal outcomes, not the mechanism by which politics influences law.⁴

In this article I suggest that the different political traditions of England and the United States may also have an impact on the way in which politics influences law. I contend that where courts try to avoid dealing with politically contentious changes in the political scene, this can lead to changes in what counts as political—and hence impermissible—argument, and by implication, in what counts as acceptable legal argument. As I will try to show below, an argument or outcome can be considered political or non-political at different times, thereby being more or less amenable for judicial use. The second aim of this article is to demonstrate this, indirect, influence of politics on law.

I have a third, although more minor, aim in this article. I wish to draw attention to the significance of politics to the examination of private law doctrines.⁵ I am, of course, not the first to do that, yet despite a wealth of theoretical work on tort law, in recent times this perspective seems relatively unexplored, because the two most popular approaches to tort theory, moral philosophy and economics, do not give political considerations a central place. Those relying on moral philosophy for explaining tort doctrine usually try to understand and justify the rules in this area by assessing them against theories of individual moral responsibility. Such theories typically operate in relative autonomy from broader political or social contexts (as explicitly stated in e.g., Goldberg & Zipursky 2005:368, 391; Ripstein 2004:1814–15, 1830). The very different economic analysis of tort law is surprisingly similar in this regard. In most economic analyses of tort law, the legal rules are explained by extending a microeconomic market model to tort situations. Under this approach tort liability is based on assessing the costs and benefits that accrue from various activities. Of course, these costs may be different for different actors, but most economic models do not call for any special treatment of the fact that the state is involved in the activity. Thus, both approaches miss what I think is central to understanding tort law as it exists today, namely that it operates as part of (or at least alongside) the institutions of the welfare state. The significance of this fact is most acute in considering the scope of tort liability of public authorities, because one cannot adequately answer the question of what the state may be liable for in tort, unless one forms a view on what the state owes individuals. Though I do not explore this normative question in this article, the discussion here is meant to point to the need to address it.

Before proceeding, a note on methodology: Much of the American literature on the ideological influence on courts is based on statistical analysis of vast datasets. There are very few such statistical studies examining English courts, and no publicly-available datasets.⁶ Instead, for the sake of examining the hypothesis that the law on the negligence liability of public authorities has been influenced by changing political attitudes, I rely on the traditional method of examining the central cases in the field, supplemented by some extra-judicial writings of active judges. Though judicial opinions do not always lend themselves to discerning political influence on judicial decision-making, I think there is enough evidence from court cases and judges' writings to piece together an account of the influence of politics on the development of the law in this area. I will try to show how this political background helps us make more sense of some of the doctrinal arguments found in the law, as well as offer an account of the changes the law has undergone during the twentieth century.

I start my argument by describing the change in legal doctrine in the years following World War II and present some of the attempts to explain them. I show the limitations of doctrinal explanations and then consider various attempts to explain these changes by aligning them to ideological changes in the political system and, in particular, the rise of New Right ideology. In section II I begin to examine the plausibility of such explanations by describing the changing political attitudes towards public services in the years following World War II. I show that these changes were part of a broader change in attitude towards the role of the state, and that these changes have also led to a more social conception of tort liability, and in particular tort liability of public authorities. I then turn, in section III to outlining the arguments made in support of limiting tort liability of the state. I first present the arguments of the kind one would hear from someone committed to New Right ideology, the political ideology most associated with Margaret Thatcher. I then show that the actual arguments found in some of the most important court decisions concerning liability of public authorities during Thatcher's premiership reflect a different ideology, one that in some sense is directly opposed to Thatcherite ideas. Finally, in section IV I attempt to explain these developments by providing several possible reasons why the Thatcherite approach did not find favor with the judges. My central argument is that ideology can bring about legal change not merely directly, by influencing judges to favor a certain interpretation of the law, but also indirectly by changing the scope of what counts as a non-political, and hence legally acceptable, position.

Explaining the Post-War Change in English Law on Negligence Liability of Public Authorities

When trying to explain the doctrinal changes to the liability of public authorities it makes sense to begin by trying to show them as part of broader developments in negligence liability. According to the familiar story it was Anns v. Merton London Borough Council (1978), coincidentally or not a government liability case, that ushered in the “two-stage test” which was rejected thirteen years later in favor of the seemingly more restrictive “three-stage test” of Caparo Industries Plc. v. Dickman (1990).⁷

Although it is tempting to explain the change in the area of liability of public authorities as the result of a more general shift in the scope of negligence liability, I do not think this explanation works. When the House of Lords overruled Anns it declared that the latter decision “did not proceed upon any basis of established principle, but introduced a new species of liability” (Murphy v. Brentwood (1991:471). The House said this even though earlier cases, in particular Dorset Yacht v. Home Office (1970) but also McGhee v. National Coal Board (1972) (both still considered good law today), adopted a very similar formulation to the one supposedly created out of whole cloth in Anns.⁸

In fact, even the change in the “tests” is not particularly illuminating, especially if we bear in mind that the law lords themselves have frequently warned against paying too much attention to the verbal formulas that govern the question of duty of care (Custom and Excise Commissioners v. Barclays Bank plc. (2006:[35]); Caparo Industries Plc v. Dickman (1990:633)). The tests for duty of care in both Anns and Caparo are extremely vague, and it is not too difficult to reach post-Caparo outcomes using the Anns test, and vice versa. On the one hand, the House of Lords began narrowing liability for public authorities before the Anns test was officially overruled in 1990 (Governors of the Peabody Donation Fund v. Sir Lindsay Parkinson Co. Ltd. (1985); Yuen Kun Yeu v. A.G. (Hong Kong) (1988)). Similarly, the Canadian Supreme Court continues to adhere to the Anns “test,” even after its rejection in its country of origin; nonetheless, it recently moved closer to the post-Caparo English approach (Cooper (2000):[28], [37]–[39]). On the other hand, there are a few decisions from recent years that seem to have adopted a more expansive approach to duty of care in a manner more reminiscent of the spirit of Anns despite following the letter of Caparo.

Still within the bounds of doctrine but focused more closely on the question of liability of public authorities, some have argued that the reason for the legal change in the 1980s was that the House of Lords had previously failed to distinguish between causing harm, for which liability should be imposed, and the failure to confer a benefit, which should lead to no liability (Bailey & Bowman 2000:87). The change in the law, then, was simply the result of the House of Lords' recognition of “defect[s] in the … reasoning” of earlier decisions (Bailey & Bowman 2000:87), and its correction of “errors in legal analysis” (Stevens 2009:651).

It is hard to accept such explanations as the whole story. The supposed legal errors were made by a unanimous panel of the House of Lords and were then adopted by other courts in the Commonwealth. That would suggest that these courts did not make a doctrinal error, but that, for a while at least, the courts did not accept this distinction as relevant. Furthermore, English and Commonwealth courts commonly read each other's decisions and are happy to borrow from each other. If this were a case of legal error, we would not expect the error to persist in other Commonwealth jurisdictions after English courts have pointed it out; and yet that is what we see.⁹ More to the point, the distinction between duty and failure to confer a benefit is one that presupposes a certain baseline of what the state is required to do, one that doctrine itself does not supply, and without which the distinction is empty.

Doctrine alone is equally unhelpful if we wish to support the overruled decisions. True, to many contemporary commentators, and of course to the House of Lords itself, both Dorset Yacht and Anns seemed natural applications, or at least justified extensions, of Donoghue v. Stevenson (1932); but such determinations (or re-determinations) of the scope of earlier decisions are never required by the earlier decisions, so something else outside the cases must have been at work.

My alternative suggestion is that we can gain some insight into this question by examining shifts in the dominant political ideology at the time. Surprisingly, there has been no detailed discussion of tort liability of public authorities, an area in which the impact of political ideas would seem more immediately relevant. One finds occasional comments that acknowledge the impact of political ideology on the law (Cane 1982:62; Lunney & Oliphant 2010:542; cf. Stapleton 1995:820), and there are also brief comparative comments that seek to attribute the difference between English and French law in this area to the difference between English (or British) and French attitudes towards the state (Fairgrieve 2002:265–66; Monti 1999:772–73). But though a step in the right direction, these throwaway remarks—quite often not more than a sentence or two—are all too brief and general to explain the doctrinal puzzles mentioned above and in particular to the change in the attitude towards liability of public authorities in the 1980s.

A more specific idea that falls firmly within this approach and that superficially provides a plausible answer to our puzzle, links the changes in the law to Margaret Thatcher's rise to power and more specifically to the ideas now associated with her name. Perhaps most famous for this claim is Patrick Atiyah. Atiyah suggested that the law in the 1970s reflected a time “in which people still believed the role of the state is to take care of people ‘from the cradle to the grave’.” He suggested that some of the restrictions on tort liability against the state that came in the following decade reflect the rejection of this view (Atiyah 1997:140–41, 176; cf. 1987:1027–28).¹⁰ Besides Atiyah, the influence of Thatcherism has been suggested by other prominent tort scholars as an explanation for various changes in tort liability that took place in this period, although, again, none of these claims has been developed in any detail (Cane 1996:483; Howarth 1991:65–66; Markesinis & Fedtke 2007:12–13; Stanton 1991:84).

Despite its appeal, I will argue below that this suggestion is unsatisfactory. Though the change in the law may be attributed to “conservative judges” (Markesinis and Fedtke 2007:37, 64), when their views are actually examined, we find in them attitudes quite different from those one would expect to see from judges seeking to implement Thatcherite ideology. In the next section I begin to look at the political changes in Post-War Britain, changes that I will subsequently argue are relevant for understanding the changing law.

The Changing State in Post-War Britain

World War II forced many changes on British society. Among others, it was instrumental to long-lasting changes in attitudes on the proper “size” of government. In part, the War was instrumental to increases in social expenditure that began during the war years and which post-War governments found difficult to roll back to their pre-War levels (Dryzek & Goodin 1986:11–21); in part it was the disillusionment with laissez faire policies for their perceived responsibility for the War (Middleton 1996:§11.3), and which helped legitimate the new tasks public authorities were now called to perform. These changing attitudes were aided by a stronger sense of solidarity between members of the different classes, as well as higher levels of trust in the state and its institutions (Marshall 1975:82–84; Marwick 2003:80).¹¹

It was during these years that new welfare institutions, most notably the National Health Service and Industrial Injuries Scheme, were created, and others that were already in place, like National Insurance, were expanded. Importantly, these developments were the work of both Labour and Conservative governments. In what are known as the consensus years following World War II there was relative similarity on many fundamental issues regarding the structure of the economy. Successive governments from both major parties supported a form of mixed economy, Keynesian economic policies, and commitment to the protection of the institutions of the welfare state (Kavanagh 1997:45–47, 71–72; Kavanagh & Morris 1994:76–77).¹² Conservative politician Harold Macmillan serves as a useful example. In 1938 he published The Middle Way (Macmillan 1938), a book inspired by the country whose economic model has over the years come to symbolize benign socialism, Sweden.¹³ When twenty years later he became Prime Minister, Macmillan reiterated his commitment to the book's main tenets.¹⁴ For him, patriotism meant that the state had parent-like obligations to care for its citizens and to make sure that the less fortunate among them were not left behind (Green 2002:160–62).

The consensus years and the expansion of the welfare state were reflected not only in the creation and expansion of new social institutions, but also in a slow but steady expansion of tort liability against the state and employers. This expansion in effect reoriented tort law towards becoming a mechanism for the optimal allocation of risks among different groups in society.¹⁵ The Beveridge Report, which laid down the intellectual foundations and provided a blueprint for the welfare institutions that were soon afterwards created, contained a short discussion of the role of tort liability as one of the means for dealing with industrial accidents and proposed to replace it with an administrative procedure that was part of a more comprehensive welfare scheme (Beveridge Report 1942:130–31). Shortly afterwards, legislation was introduced to abolish the defense of common employment and to eliminate the immunity of state bodies from private law claims.

The change in political mood, and the statutory developments it begat, have been noticed in academic commentary. Academic lawyers writing at the time started talking about this area of law in terms of what it “does” (instead of what it “is”), and increasingly this question was answered by considering its role within the welfare state (Denning 1949:72–75, 77–81; Friedmann 1972:ch. 5; Williams 1951). Towards the end of this period, many found tort law wanting and have suggested more or less radical reforms to it, from no-fault liability for accidents to the replacement of the law with social insurance schemes (Atiyah 1970:611–14; Ison 1967; and from judges writing extra-judicially: Kilbrandon 1966:32–33, 42–43; Parker 1965:9–11). Even though these reforms were not adopted, the fact that they were suggested is a good indication of what was by then the mainstream academic view as to the appropriate way of thinking about tort law in the modern state.

These changing political and social attitudes have also affected judicial attitudes on doctrines within tort law. In the area that is the focus of this article we can begin with East Suffolk River Catchment Board v. Kent (1941), a case in which a majority of the House of Lords rejected a claim against public authorities for being unreasonably slow in fixing a broken wall, which the plaintiffs contended resulted in losses from high tides. The majority's opinion was premised on a dichotomy between powers and duties: whenever public authorities are given powers to act it follows that they cannot be under a duty to act. As the authority only had a power to act, its failure to act more quickly could not give rise to liability. In dissent Lord Atkin rejected the dichotomy: being given a public (statutory) power does not entail that one is not also under a common law duty to take reasonable care not to injure others, a duty that existed “whether a person is performing a public duty, or merely exercising a power which he possesses either under statutory authority or in pursuance of his ordinary rights as a citizen” (East Suffolk 1941:89).¹⁶

To see how the law changed in the years following East Suffolk we need to identify two different developments, since it is only by understanding the way both have operated together that we can make sense of the expansion of state liability. The first (and more familiar) change is the emergence of a general duty of care and a general tort of negligence. Though this development is rather familiar now, it is easy to forget how long it took. Donoghue, now read to have established a new general tort of negligence, did not become a leading case overnight.¹⁷ It was probably only in the 1960s that it began to be understood in this way (e.g., Diplock 1965:2), with the definitive statement found in Dorset Yacht v. Home Office.¹⁸ In holding for the plaintiffs Lord Reid said that though “the well-known passage” from Lord Atkin's opinion in Donoghue “should not be treated as if it were a statutory definition … it ought to apply unless there is some justification or valid explanation for its exclusion” (1970:1027). This view was further entrenched in Anns v. Merton London Borough Council (1978:757).¹⁹ Writing for a unanimous panel Lord Wilberforce explained the apparent difference between the decision and the majority's opinion in East Suffolk by stating that “the conception of a general duty of care … had not at the time [of East Suffolk] become fully recognised.” This was a polite way of overruling East Suffolk in favor of Lord Atkin's dissent (an opinion thought, unsurprisingly, to have had a better grasp of the principle established in Donoghue).

Next to this familiar development, there was a second, much less familiar, change in judicial attitudes about the relationship between the individual and the state. Perhaps because they felt freer to express their broader view on developments in the law in extra-judicial writings, it is mostly there that we can identify this change. As early as 1953 Lord Radcliffe explained that this was “the Century of the Plaintiff,” because of “[t]he widespread use of insurance, so that people have come to feel that there is no loss or mischance that ought not somehow to be made good to the sufferer—by someone else.” He added further that as “so much of industry and public activity is now conducted by large impersonal corporations with large impersonal resources … it hardly seems even unkind to make them pay for every sort of damage that an individual may have met with at their hands.” It was, he concluded, “humanitarianism,” the sense that makes people “indignant” to learn that “there can be grievances in all the dark and irresponsible record of human affairs which the law is not equipped with a remedy to put right” (Radcliffe 1968:32–33; cf. Macmillan 1937:6–8, 279–83; Nettleship v. Weston 1971:699), that led to the change in legal attitude.

It is important to recognize that this new conception of the relationship between the individual and the state was something that judges at the time were fully aware of. The more influential judges of the time quite clearly recognized that these political changes had implications not only for public law; they also required reshaping tort law to better fit the new reality of the modern welfare state which was at the heart of the post-War consensus.²⁰ Their refashioning of tort law, and especially the tort of negligence, was a conscious attempt to make sure the law remained in line with the changes that were taking place outside the law. The main tool was an increasingly frequent acknowledgement of the significance of “policy” considerations (Dorset Yacht Co. Ltd. v. Home Office 1969:426–27; Reid 1972:27; Radcliffe 1960:40–41). In one particularly clear statement Lord Radcliffe (1960:57) criticized the English courts for taking the wrong kind of considerations into account when appealing to public policy. What they should consider is not so much “what the State or the public requires in its own interest,” but rather “what the public should guarantee to the individual for the protection of his essential dignity.”

On the more traditional understanding of tort law, it belongs firmly within “private law”; as such it is only concerned with redressing private wrongs and any suggestion that it should be affected by the changing relationship between individual and the state would be misplaced. But the combined effect of the two developments just described has been to provide a radically new understanding of tort law as part of the machinery of the welfare state. Thus, in Hedley Byrne and Co. Ltd. v. Heller and Partners Ltd. (1964:536) Lord Pearce said that the scope of the duty of care had to be based “upon the courts’ assessment of the demands of society for protection from the carelessness of others.”²¹ Speaking more generally in Broome v. Cassell & Co. Ltd. (No. 1) (1972:1114), Lord Wilberforce, who would go on to deliver the main speech in Anns only a few years later, rejected the view that it is always the “criminal law, rather than the civil law [that is] … the better instrument for conveying social disapproval, or for redressing a wrong to the social fabric. …” And around the same time Lord Denning, who has cast considerable influence on the development of tort law in this period, wrote that the “reason why the law of negligence has been extended so as to embrace nearly all activities in which people engage … is that, when severe loss is suffered by any one singly, it should be borne, not by him alone, but be spread throughout the community at large.” Even the requirement of fault that was still retained in the law was not explained by appeal to any moral or legal notion of rights, but simply because “compensation without fault would make society bankrupt” (Denning 1979:280).

It is interesting to see how the two developments were used to overcome doctrinal hurdles that stood in the way of developing the law in what was thought to be the desired direction. The broad reading of Donoghue in the context of Dorset Yacht and Anns established the idea that as far as the citizen is concerned the state is its “neighbour” for much of what it does. Typically, Lord Denning was clearest on this front. In Dutton v. Bognor Regis Urban District Council, decided a few years before Anns, and likewise dealing with the liability of public authority for failure to inspect poor construction by private contractors, he wrote: “[The local council] were entrusted by Parliament with the task of seeing that houses were properly built. They received public funds for the purpose. The very object was to protect purchasers and occupiers of houses. Yet they failed to protect them. Their shoulders are broad enough to bear the loss” (1972:398). This was the political consensus on the welfare state translated into legal doctrine.

Dutton is important for another, and perhaps more significant, doctrinal innovation that clearly reflects the second development mentioned earlier. As we have seen, the majority in East Suffolk has taken the existence of a statutory power as excluding the possibility of a duty, dooming the claim. Lord Denning overcame this conflicting (and formally binding) case by dismissing the distinction drawn there between power and duty. The distinction was a “mistake,” he said, because when considering “the functions of a local authority” there is a more applicable “middle term”: the public authority had “control” over “every stage of the work” (Dutton 1972:391–92).²² Control is the kind of relation that exists between individuals and the state, and it is one that is markedly different from the relationship that exists between individuals who are more-or-less equally situated. State authorities have “control” because they can compel compliance with their requirements, and this control implies a duty of care. Because the state had control over the situation, the distinction drawn in East Suffolk between harms that were the result of risks caused by actions of the defendant and cases of harm resulting from risks created by natural causes could be rejected. When the state was given (or has taken) control for the sake of reducing risks, it mattered little whether the risk was created by the state.²³

Of the two developments just described—the emergence of a general duty of care and the reorientation of the relationship between individuals and the state in the context of tort claims—most academic attention has been directed to the first. It is the development that could more easily be identified, could be thought relevant to “private” law and, whether liked or not, could be explained in terms of familiar legal categories. The second development, by contrast, challenged the separation between private law and public law and hovered close to the boundary between law and politics.

Ignoring this second development, it is not surprising that some commentators found Anns and the decisions following it difficult to understand and ended up explaining it as the result of “errors in legal analysis” (Stevens 2009:651). But with the change in the relationship between individual and the state as part of the picture Anns makes perfect sense. It reflects a concerted effort to adapt the law to the new state responsibilities brought about by the new institutions of the welfare state.

The Thatcher Years

By the time Margaret Thatcher became Prime Minister there was a growing sense that something had gone wrong with Britain, that it became “ungovernable.” More and more people have come to accept the view that it was the welfare state, hitherto untouchable, that was to blame for many of the ills that befell British society in the late 1970s. Before Thatcher came to power major reforms in public services were sometimes mooted, but not much was done (Lowe 1994:48–49). Thatcher, backed by growing public discontent, sought radical change. Informed by an ideology that looked at all government institutions with mistrust, she was eager to reorganize public authorities and in particular the institutions of the welfare state. She disliked government because she thought it tended to divert talented people away from wealth-creating business to business-stifling bureaucracy; she particularly disliked the institutions of the welfare state for encouraging a culture of dependency instead of a culture of responsibility (Harris 1988:22–23).

It was not just the welfare state that Thatcher wanted to change, it was also the political consensus that brought it about. As she put it “consensus seems to be the process of abandoning all beliefs, principles, values, and policies” (quoted in Kavanagh & Morris 1994:2). The decline in solidarity and trust of government (Harris 1986:256–57) suggests she was not alone. The question I explore in this section is whether these shifts had an impact on the judiciary in the area of liability of public authorities, and if they did, what it was.

New Right in the House of Commons: The State Can Only Do Wrong

Commentators disagree on the extent to which Thatcher (or Thatcherism) is a continuation of conservative ideas or a departure from them.²⁴ It is clear, though, that Thatcher had little patience for Burkean incrementalism or the protection of institutions just because they withstood the test of time. Nor did she harbor Oakeshottean suspicions of “rationalist” attempts at “political science.” Thatcher believed that one could glean from the works of the likes of Friedrich von Hayek or Milton Friedman guiding principles for reform of government and society. If those called for major changes to, or even the elimination of, old institutions, so be it.

How does this ideology translate to liability of public authorities? At first, it may seem to favor the expansion of liability of public authorities. According to this view, public authorities should be treated just like private service providers, and therefore should be liable for “breach” whenever they fail to live up to promised standard (for such suggestions coming from the Institute of Economic Affairs, the most prominent British New Right think-tank, see Mather 1991:73–75, 82–84, 87–88). But in reality, the attitude has been quite different, for reasons that are not difficult to see. This may have something to do with worries that broader liability may lead to undesired expansion of state institutions needed to handle all those claims. Mostly, however, the objection to such liability can be explained by moral arguments that have to do with the perceived detrimental impact that expanding liability of public authorities would have on the “Victorian values” of personal responsibility (cf. Travers 1977:163, 175–81).²⁵ Expanding the scope of tort liability on the state effectively turns it into an insurer of last resort, one that potentially covers all possible activities. Thatcher's famous dictum that “there is no such thing as a society” is often said by her defenders to have been taken out of context. When brought into context it is clear that she was concerned about a society in which people do not take responsibility for their actions, and are given by an over-pampering state disincentives for self-improvement.²⁶ It is thus an individualistic call for greater self-reliance and responsibility, and—what is merely the other side of the same coin—the demand that they stop looking to “blame others for their misfortunes” (Atiyah 1997:138; cf. Cane 2006:462).

This link between welfare and morality makes familiar New Right ideas easy to translate into arguments against the expansion of tort liability of public authorities. If tort law is understood as a system of personal responsibility,²⁷ then it would be a mistake, moral and conceptual, to introduce welfarist considerations into it by expanding liability of public authorities. A basic premise of this view is that a responsible, planning person should take possible future misfortunes into consideration when deciding on future action. On this view it is the injured person's failure to consider possible mishaps and to protect herself against them (by reducing the level of activity or investing in safety measures to reduce the risk of harm, or by purchasing insurance to reduce the impact of the harm) that is responsible for her misery. On many occasions it is the “victim” who will be in a better position than anyone else, both epistemically and morally, to consider potential harm that might befall her. At its most extreme, this view contends that there is nothing to distinguish the risk of harm from, say, an earthquake from the risk of harm brought about by the actions of other people: both are foreseeable setbacks for which one often can, and therefore should, adequately prepare in advance (Epstein 1996:293; Kaplow 2003:177; cf. Spigelman 2002:433). Even if the plaintiff is not similarly situated to prevent an injury from occurring, she is typically in a position to decide for herself whether and to what degree she wishes to deal with its potential impact.²⁸ In any case, the state has no role in being involved in what are private relations between individuals.

A related but somewhat different way of getting to similar conclusions emphasizes the importance of choice. Thatcher is reported to have said that “Choice is the essence of morality” (quoted in Jordan 1989:19 n.1; similar quotes are found in Finlayson 1994:358–59) and expansion of tort liability might be thought to remove this choice. The rational, planning person should be given the choice between protecting herself against potential risks and running the risk, as well as the choice, if she elects to protect herself against the risk, on how to do so. By imposing tort liability on the state for failure to protect from the acts of others, the law removes potential victims' ability to choose between engaging in an uninsured but cheaper activity or an insured but more expensive one, and effectively adds compulsory insurance to the costs of the activity to which all have to pay through their taxes (Huber 1990:207–19). In other words, from this perspective expansion of tort liability both creates moral hazards by providing incentives for people to take less care of themselves and can operate to remove the choice of those people who wish to take responsibility over themselves.

This is what one would have expected to see in judicial opinions from the mid-1980s if the judges at the time had been influenced by Thatcherite ideas. I argue in what follows that we do not see such views in the decisions; in fact, we find in them quite different views.

Old Tories in the House of Lords: The State Can Do No Wrong

The New Right ideas were by no means unanimously favored by members of Thatcher's party. Many prominent members of the Conservative party's old guard believed Thatcher's views had little to do with the Tory tradition with which the party had been associated for a long time. Harold Macmillan, by then styled Earl of Stockton, spent the last years of his life bitterly criticizing his party's government for getting “nothing right” (quoted in Evans 1998:27). Another former Conservative prime minister, Edward Heath, was similarly critical of the Thatcher government for losing its way (Finlayson 1994:366–67; Evans 1997:608–10).

My argument is that the retrenchment of liability in the 1980s is closer to the views of these critics of Thatcher rather than to her own views. At least in the decade in which the change in scope of tort liability began, it is not easy to find court decisions that express the same suspicious attitude towards public authorities and the same veneration of personal responsibility;²⁹ what one does find in judicial opinions from this period are the sort of considerations that have been described as the “traditional Conservative ideals” of “strong government, patriotism, and authority” (Crewe & Searing 1988:365).

To proponents of this approach the state should retain under its control certain inherently “public” services, not just by declining to privatize them, but also by refusing to conceive of them in contractual terms. The “paternalism” between the state and the individual is the relation that exists between subjects and the patria, that is, one that reflects the “despotism of parenthood” (Scruton 1984:110–11).³⁰ These ideas were part of a “Tory culture” that emphasized “deferential attitudes towards authority … [an] anti-egalitarian ethos and … status hierarchy” (Gamble 1994:170; Johnson 1985:226–27, 234–35, 248). Where Burke spoke of the “generous loyalty to rank and sex, that proud submission, that dignified obedience, that subordination of the heart” (Burke 1968:170), latter-day Tories explained that “[i]t is the absolute duty of the state to have power over its subject … [and] therefore [the state must] withdraw from every economic arrangement which puts it at the absolute mercy of individual citizens” (Scruton 1984:111). This patrician Tory tradition thus rejected both the idea that state institutions are dangerous entities whose domain should be curtailed as much as possible and the one that saw citizens as customers who are in a position to make demands from it.

It is almost inevitable that this approach assumes public authorities to be populated by competent, altruistically-motivated, public servants.³¹ And it is this attitude that one finds in the central cases on liability of public authorities from the 1980s. Perhaps the clearest encapsulation is found in Hill v. Chief Constable of West Yorkshire (1989), one of the leading cases in the trend towards limiting liability of public authorities. This case involved a claim by the mother of the last victim of a serial murderer and rapist. There were indications that the police forces ignored relevant evidence that could have led to an earlier capture of the perpetrator. The mother alleged that had the police acted with greater vigilance, the murderer would have been caught earlier, and her daughter would have been spared. Such a case could have been dismissed on failure to prove causation or carelessness on part of the police. Instead, the case was dismissed for lack of duty of care. In the course of explaining why, Lord Keith said (1989:63):

Potential existence of [tort] liability [of public authorities] may in many instances be in the general public interest, as tending towards the observance of a higher standard of care in the carrying on of various different types of activity. I do not, however, consider that this can be said of police activities. The general sense of public duty which motivates police forces is unlikely to be appreciably reinforced by the imposition of such liability so far as concerns their function in the investigation and suppression of crime. From time to time they make mistakes in the exercise of that function, but it is not to be doubted that they apply their best endeavours to the performance of it.

Lord Keith was not alone. Calveley v. Chief Constable of Merseyside (1989) dealt with a claim by police officers who were suspended on allegations of misconduct that were later found to have been unfounded. Claiming negligence on part of the investigating officers, they sued for the wages they lost during their suspension. Again, the claim was dismissed not on its merits, but rather on the basis of lack of duty of care. In his decision Lord Bridge expressed (1989:1238) a view very similar to Keith's:

it would plainly be contrary to public policy, in my opinion, to prejudice the fearless and efficient discharge by police officers of their vitally important public duty of investigating crime by requiring them to act under the shadow of a potential action for damages for negligence by the suspect.

It is tempting to think that this attitude was confined only to the police, against whom British courts have a long history of timidity (Stevens 2005:75), but similar statements are found in cases dealing with other public authorities. In another opinion by Lord Keith, in Rowling v. Takaro Properties Ltd. (1988), this time in a Privy Council decision on an appeal from New Zealand, he wrote (1988:502) with regard to a claim against a minister who made an administrative decision that caused harm to an individual that

in the nature of things, it is likely to be very rare indeed that an error of law of this kind by a minister or other public authority can properly be categorised as negligent. As is well known, anybody, even a judge, can be capable of misconstruing a statute; and such misconstruction, when it occurs, can be severely criticised without attracting the epithet “negligent.” Obviously, this simple fact points rather to the extreme unlikelihood of a breach of duty being established in these cases. …

Though this discussion looks as though it deals with the question of breach of duty, it appears in the judgment under the heading of duty of care and it led him to the very restrictive view that in general “it would … be in the public interest that citizens should be confined to their remedy … in those cases where the minister or public authority has acted in bad faith” (Rowling 1988:502). Maintaining the rule of law, the importance of subjecting executive action to legal oversight, and the retrospective application of legal interpretation, could all have swayed the decision in a different direction, but in the end it was the traditional Tory attitude toward state institutions that carried the day.³²

A final example, this time dealing with a local authority, comes from a somewhat later decision from the Court of Appeal. M. v. Newham London Borough Council (1995) consolidated several claims. One was a claim by a mother and child who were separated for almost a year as a result of a mistaken diagnosis that the mother's partner was sexually abusing the child. The other claims involved the opposite situation, of children who sued the public authorities for their failure to separate them from abusive families. All claims were dismissed. In the Court of Appeal Lord Justice Staughton stated (1995:675–76):

One advantage that is claimed for imposing a duty of care is that it encourages people not to be negligent. … [But] even if psychiatrists and social workers were likely to have to pay damages personally, I do not suppose that they would be any less caring for children in need than they are already; they might, as I have said, adopt defensive practices; but I doubt if their general level of care would change.³³

Even when not stated explicitly, this attitude helps explain the widely-discussed “defensiveness” argument, found in many decisions restricting liability against public authority. (In fact, in Hill it appears immediately after the passage quoted above from the decision.³⁴) This argument alleges that the imposition of liability on public services is likely to lead to an overly defensive attitude of the public authority. The defensiveness argument has been (rightly) criticized by commentators for lacking in empirical support and recently some judges expressed unease about it (Brooks v. Commissioner of Police of the Metropolis 2005:[6]). But this argument makes much more sense if we assume that public services are populated by officials who are altruistically-motivated and (the occasional mistake aside) are doing a good job. If this is the case, it follows that any imposition of liability on public officials will inevitably lead to overdeterrence and to the feared defensive frame of mind. Without such an assumption any general claim on the likely effect of the imposition of more extensive liability on public services is unwarranted.

In none of the cases discussed above was this line of argument the only one mentioned. Perhaps next to the other arguments the words quoted above have appeared like rhetorical flourishes; and it may be that for this reason this particular argument against the imposition of tort liability has received relatively little academic attention. But there is no reason to ignore an argument that appears in many of the central decisions in this area, especially as essentially the same argument is found in East Suffolk Rivers Catchment Board v. Kent, the case that provided the doctrinal and intellectual foundation for Hill and the cases that followed it. In East Suffolk Lord Thankerton said that “there are special circumstances in the case of statutory bodies … which should lead to the application of a less exacting standard than ordinarily prevails,” because in the circumstances in which they operate “much may be condoned as well-meant error of judgment” (East Suffolk 1941:95–96).

To highlight the uniqueness of this ideological perspective it is worth comparing it to its two main competitors considered above. First, consider the attitude found in a tort law textbook (McBride & Bagshaw 2005:203 n.55, quoting McKinstry 2002), whose explanations of the law are often informed by New Right ideology:

While we would agree with one commentator that “fraud, laziness, ineptitude and money-grabbing are the hallmark of Britain's public sector” … it is not clear that finding that a duty of care was owed in this kind of case would bring about any improvement in the dismal performance of public bodies in the UK. … [E]xpanding the scope of public bodies' liability in negligence only serves to starve them of the money they need to perform their services, as more and more of their funding is diverted into paying for litigation and making compensation payments to claimants.³⁵

Though this book aims to explain existing doctrine, the position it adopts is clearly at odds with what one finds in East Suffolk and its 1980s progeny. For one, the negative view of public bodies it takes is very different from what we find in the cases. Furthermore, in line with New Right ideology, the book argues that public bodies should be treated in the same way as private bodies (McBride & Bagshaw 2005:206; discussed further in Priel 2011:20–22), whereas the cases favored a more restrictive attitude to state liability.

Even more interestingly, the attitude found in the 1980s cases is also fundamentally different from the view one finds during the period of political consensus on the welfare state. In Dorset Yacht the imposition of tort liability was challenged by the suggestion that it would lead to an overly defensive attitude. After quoting from a New York Court of Appeal decision that relied on the defensiveness argument to limit liability on public authorities, Lord Reid, one of the leading architects of the reformulation of negligence liability along welfarist lines, famously said:

It may be that public servants in the State of New York are so apprehensive, easily dissuaded from doing their duty, and intent on preserving public funds from costly claims, that they could be influenced in that way. But my experience leads me to believe that Her Majesty's servants are made of sterner stuff. … I can see no good ground in public policy for giving immunity to a Government Department (Dorset Yacht 1970:1033).

Like the 1980s cases (and unlike the New Right view), Lord Reid's words show a positive attitude towards public authorities. Where he differed from them is in coupling this position with a view regarding the changing role of the state. Consequently, he refused to take the deferential attitude towards public authorities that played such a central role in the limitation on liability in subsequent years. As we have seen, the decisions from the consensus years agreed with the earlier (and later) position that public authorities should be treated differently from private defendants, but maintained that the difference favored broader liability.

What Explains the Change in Doctrine?

What I have sought to show so far is a missing piece in the explanations of the doctrine on negligence liability of public authority in English law. I claimed that arguments that most commentators in the field consider to be irrelevant appear in the most important decisions and have influenced the outcome of these decisions. In this section I try to show how this perspective can help us understand the broad doctrinal trends described in the beginning of this article.

As judges do not usually discuss such matters openly, this section is somewhat more speculative, but I think we can offer a plausible account of the influence of politics on law in this context. To do that we must distinguish between two different ways in which politics can influence the law. One is the more familiar direct influence model, according to which the judges' (or the general public's) political opinion influences the law. On this view judicial opinions are “smokescreens” (Markesinis et al. 1999:39) for what really drives the decision, which implies that if judges claim to be basing their judgments exclusively on legal arguments, they are either lying or deluded. This is the model of political influence presupposed both by those who think English law can be explained in political terms (e.g., Griffith 1997) and those who reject this view (Stevens 2009:651–52).

It is impossible to rule out that English judges are influenced by political opinions. Like other human beings they see the world from a certain perspective, and in a country in which “[t]he national culture is Tory culture” (Johnson 1985:234; and Devlin 1978:505–06, admitting that this outlook is true of the senior English judges), they may have hardly noticed the political undertones of the view they took in the 1980s: the traditional Tory ideas that made their way into their decisions may have just seemed natural to them.³⁶ But I take seriously the various institutional constraints created in English law to keep law and politics apart (these include a strong doctrine of Parliamentary supremacy, stricter adherence to precedent, limited use of non-legal materials in litigation, non-political process of judicial appointments, more formal legal education), and believe that English courts have had some success in limiting the direct impact of politics (Priel 2012:325–27).³⁷ Yet even in such an environment, I believe there is a different way in which politics can influence the law. Unlike the direct influence view, according to which judges are engaged in politics even when claiming to base their decision on the law, the model of indirect influence seeks to show that political change can shift the boundaries of what counts as political and by implication what counts as legal, thereby shifting the boundaries of “permissible” legal argument. (Of course, the indirect model does not rule out the possibility of direct influence of politics of law.)

To understand the indirect model, it is necessary to distinguish first between two close but different senses of politics. A view may be considered “political” in the sense that it touches on questions relating to the organization of community; and it can also be “political” in the sense that it is subject to political controversy. It is not difficult to show that there are certain “political” assumptions underlying the private law in the first sense. To give one obvious example, private law presupposes the legitimacy of private property, and that is not a neutral political assumption. This political assumption, however, is one that is uncontroversial within the political mainstream of most Western countries. It is thus not political in the second sense, because basing their decisions on this assumption does not require judges to justify their decisions by appeal to arguments that are external to the law. In such cases politics is embedded into the legal doctrine in a way that allows one to justify a decision by appeal only to “internal” legal materials. Politics becomes problematic for (English) lawyers when it is not so embedded into the law, and this most commonly happens when political issues in question are publicly controversial.

American courts have a long history of deciding on matters that were controversial in exactly this sense (slavery, school segregation, abortion, health care, and so on). Much of the theoretical work in this field is therefore dedicated to articulating that role and to devising institutional mechanisms that will allow courts to consider politically controversial matters without deciding them on their political merits (Priel 2013). Competing theories of interpretation, the search for “neutral principles,” the distinction between principles and policies—these are all means used by American courts and academic lawyers in order to allow law to have a say in some political controversies while remaining outside of it. (Whether these mechanisms succeed is another matter.)

English courts have taken a different approach: rather than develop mechanisms for distinguishing between proper and improper engagement in politics, they have traditionally simply left political matters to politicians.³⁸ But even in a jurisdiction in which courts are concerned to keep out of politics, political change can still have indirect influence when it affects what is politically controversial. As mentioned before, for years after World War II the welfare state was a mainstream political position in Britain, as commitment to its protection and even expansion was shared by the two major political parties. As such, support for it was no longer seen as political in the second sense mentioned above. (Indeed, given that contestation belonged to the political fringes, opposition to the welfare state would have been seen as political in the forbidden sense.) When no longer politically controversial, the welfare state became one more fact about the world that judges had to take into account in maintaining their historical role of shaping the path of the common law to fit changing circumstances. The courts could thus explain their doctrinal innovations as part of what uncontroversially is part of their job (cf. Devlin 1979:2).

Once the consensus on the welfare state evaporated, once Thatcher's views had become so prominent in the discourse, the scope of the welfare state moved to the center of political contestation. In this new environment the following earlier decisions might have seemed part of a broader agenda of promoting “social justice,” something that by then was a partisan (and for this reason off-limits) concern for a judge (cf. Devlin 1979:8). For the very same reason adopting the New Right's approach to the problem would have been equally unappealing for the judges, for it too could not be presented in non-political terms, it too was the product of a political ideology. Against this background, resurrecting the majority opinion of East Suffolk (and dismissing much of the development that came after it as having no “basis of established principle”) may have seemed the most politically neutral route.

Conclusion

James Callaghan, Margaret Thatcher's predecessor, famously said (quoted in Morgan 1997:697):

There are times, perhaps once every thirty years, when there is a sea-change in politics. It then does not matter what you say or what you do. There is a shift in what the public wants and what it approves of. I suspect there is now such a sea-change—and it is for Mrs. Thatcher.

When such a shift occurs, more than anything else it changes the frame of reference of old debates: some views that until then would have been considered beyond the pale become part of acceptable discourse; and, if only for some time, other views that had been part of the debate become obsolete. What I have argued in this article is that it can also change the boundaries of political argument, and thereby, at least in some contexts, the boundaries of acceptable legal argument.