UK Armed Forces Personnel and the Legal Framework For Future Operations

Written submission from Dr Jonathan Morgan, Director of Studies in Law,

Corpus Christi College, University of Cambridge

Military Negligence:

Reforming Tort Liability after Smith v. Ministry of Defence

Executive Summary

The Supreme Court’s decision in Smith v. Ministry of Defence (failing to strike out negligence and human rights claims over combat injuries) is deplorable. It will have damaging consequences for military effectiveness. It also leads the judiciary beyond the limits of their proper constitutional role. Accordingly Smith should be reformed. Although the legal basis for the Smith decision is fairly weak, it would be naive to expect any radical change of approach from the courts in the near future. Therefore legislation will be necessary to restate the proper limits of tort liability for military negligence. A power exists under s.2 of the Crown Proceedings (Armed Forces) Act 1987 which can largely achieve this goal. But a consideration of the desirable functions of tort liability shows that simply introducing an immunity would be undesirable (to say nothing of its political acceptability). It is therefore argued that a new Government commitment to compensating combat injuries fully, on a no-fault basis, be coupled with the revival of Crown immunity using the 1987 Act.

About the Author

Dr. Jonathan Morgan is Fellow, Tutor and Director of Studies in Law at Corpus Christi College and University Lecturer in Tort Law, University of Cambridge. He is active as a teacher and researcher in the fields of tort and public law. He is one of the editors of Hepple and Matthews’ Cases and Materials on Tort. Publications in the specific field of public authority liability include "Policy reasoning in tort law: The courts, the Law Commission and the critics" (2009) 125 Law Quarterly Review 215, and a case-note on the Court of Appeal’s decision in Smith v. Ministry of Defence, "Negligence: Into Battle" [2013] Cambridge Law Journal 14. Articles on human rights law include "Amateur Operatics: The Realization of Parliamentary Protection of Civil Liberties" in Campbell, Ewing and Tomkins (eds), The Legal Protection of Human Rights: Sceptical Essays (Oxford 2011).

The Structure of this Paper

At the heart of the paper is a consideration of the Supreme Court’s decision in Smith v. MoD and how that approach should be reformed. But the paper opens with two sections that lay the ground by considering the purpose and social function of tort claims more generally, and the basic forms of liability (fault-based, strict, and immunity from liability) available to satisfy those purposes. The third section critically analyses Smith v. MoD. The legal basis of the decision (both common law and European Convention on Human Rights) is explored and found to be questionable. More importantly, the likely damaging effects of the decision are outlined-the threat of litigation’s impact on military effectiveness, and the constitutional danger of courts intruding into non-justiciable political territory. The fourth section explores how to reform Smith. The damaging effect of the decision can best be avoided by a combination of exclusion of negligence claims (a statutory power that the Secretary of State already possesses) and, in place of negligence, a commitment to pay full compensation on a no-fault basis. It is suggested that a direct commitment to this effect by the Government, rather than a radical reformulation of the Armed Forces Compensation Scheme, is the best way to achieve this. A brief recapitulation of these conclusions and recommendations concludes the paper. (In an appendix the author’s 2013 case-note on the Court of Appeal decision in Smith v. MoD follows.)

I. The Purpose of Tort Claims

While an action in negligence is a claim to be compensated for an injury, tort claims can play other social functions too, such as deterring culpable behaviour and holding to account governmental or other powerful institutional defendants. These are explored below as the basis for later discussion of claims for injuries sustained during military operations.


In a common law negligence claim "damage is the gist of the action", so that the action will lie only if somebody has actually been injured by the negligence of another. This may be contrasted with tort claims for (e.g.) battery or false imprisonment, or claims under ss. 6-8 of the Human Rights Act 1998, where all that need be shown is invasion of the necessary right. These torts are "actionable per se"-and nominal damages (or a declaratory judgment) may be granted to vindicate the right in question where such an invasion is shown. But where actual losses are proved these may also be recovered in such a "vindicatory" claim.

Tort compensation aims to put the claimant back where he or she would have been had it not been for the tort-restitutio in integrum. This takes full account of the personal situation of the claimant, for example his or her particularised "loss of the amenities of life", medical needs and earnings forgone, in a catastrophic injury case.


The prospect of being sued in tort may clearly act as an incentive to avoid the negligent behaviour that gives rise to the liability. However, it seems that tort liability is but an imperfect way of deterring negligence in practice.

On one hand, the deterrent edge of tort liability is blunted by the widespread presence of insuring against such liabilities (through "liability" or "third party" insurance-which is often a legal requirement as for e.g. motorists or employers). The criminal law avoids this drawback since it is (obviously) against public policy to insure oneself against the risk of being fined or imprisoned after criminal conviction. Moreover, criminal penalties are tailored to the degree of culpability of the defendant, whereas damages for negligence are calculated according to the severity of the claimant’s injuries rather than the gravity of the defendant’s fault. (And a highly negligent defendant who, through luck rather than judgment, injures nobody will escape tort liability altogether.)

Conversely, it is possible that tort liability may sometimes over-deter. Instead of inducing an activity to be carried on carefully the spectre of liability may cause that activity to cease. Such "overkill" is obviously a matter of concern where the activity forgone is something that benefits the public, such as a discretionary public service. [1] An example would be a local authority removing a children’s playground from a public park because of the fear of litigation were a child injured using it-and/or because of the (high) level of insurance premiums charged to indemnify the authority against such claims (which may often be the direct source of deterrence in practice). Judges have often been alive to this possibility, stressing (in a slightly different context) that: "Of course there is some risk of accidents arising out of the joie-de-vivre of the young. But that is no reason for imposing a grey and dull safety regime on everyone." [2]

Whether negligence liability induces optimum levels of carefulness, or under-deters such behaviour, or over-deters it, is a complex question that in the end requires real-world empirical evidence. Whether the legendary national "compensation culture" really exists (inducing widespread timidity), or is largely the result of media presentation of anecdote, is a case in point. [3]


In his famous account of the Rule of Law in Victorian England, AV Dicey placed tort law at the centre of government accountability. [4] Dicey argued that a French style droit administratif was at best unnecessary (and perhaps positively undesirable). In England, public officials were subject to the same law as everyone else. If they commit torts they are liable in the same way as anyone else, unless they can rely on positive legal authority to act in the way they have done (a mere plea of "state interest" not being sufficient authority): Entick v. Carrington (1765) 19 State Trials 1029. This remains a salutary principle today. Tort claims (for battery or false imprisonment) are still a straightforward way of requiring (for example) the police to show that they correctly used their powers of arrest and detention. But when public authorities now have such extensive statutory powers, it is not surprising that a system of administrative law has grown up (through the application for judicial review) to police the limits and exercise of those powers.

But albeit the "accountability" function of tort is diminished from its Diceyan heights, it has not disappeared altogether. Even negligence claims may perform a similar function. In alleging negligence against a public body, citizens can (providing sufficient prima facie evidence exists) oblige the authority to explain and justify the conduct that led to the injury of which the citizen complains. In one well-known case Doreen Hill, the mother of one of the victims of the "Yorkshire Ripper" serial killer, sued the West Yorkshire police for their negligence in having failed to apprehend him before he murdered her daughter. The plaintiff stated that any damages awarded to her would be devoted to an appropriate charity; her motive in bringing the claim was not financial but rather "with the object of obtaining an investigation into the conduct of the West Yorkshire police force so that lives shall not be lost in the future by avoidable delay in the identification and arrest of a murderer". [5] The House of Lords held that no duty of care had been owed to the deceased and so the claim failed. But Lord Templeman commented that Mrs. Hill’s motive for bringing the claim had anyway been misconceived: "The efficiency of a police force can only be investigated by an inquiry instituted by the national or local authorities which are responsible to the electorate for that efficiency." [6] So while it is widely thought important that government should be held accountable for its fault, there are doubts over a tort claim’s suitability to function as a full public investigation into wider public policy failings.

II. The Extent of Liability

For a given type of claim (in respect of a particular kind of loss, or against a particular class of defendant) there are logically three possible forms of liability. Most favourable to claimants, a defendant might be liable for the loss in question irrespective of fault-strict liability. Secondly, the defendant may be liable only when the loss was caused by his or her fault, as in the tort of negligence. Thirdly, the defendant may not be liable for a certain kind of injury at all, irrespective of fault-an immunity (absence of liability).

All three of these positions have obtained in cases of injuries to armed forces personnel over the past decades. Most obviously, there was a sharp change away from the historical Crown immunity from tort claims to ordinary fault-based (negligence) liability, with the enactment of the Crown Proceedings (Armed Forces) Act 1987. The recent decision of the Supreme Court in Smith v. Ministry of Defence shows how far-reaching that change has been. [7] But these changes in tort liability have taken place against the constant backdrop of the Government’s compensation/pension arrangements for wounded personnel. These date back (at least) to the War Pensions Scheme of 1917, today being embodied in the Armed Forces Compensation Scheme (AFCS). [8] The import of this scheme, broadly speaking, is that service personnel are entitled to government compensation for injuries or death caused by service, irrespective of fault on the part of the armed forces / MoD.

These three basic positions on liability can be analysed using the functions of tort liability suggested above.


As noted, the armed forces were immune from liability until the historical privilege of the Crown in this respect was removed by Parliament’s 1987 amendment of the Crown Proceedings Act 1947. Furthermore, it was accepted by all the judges in Smith v. MoD that there remains today a common law principle of "combat immunity". [9] Its precise ambit was, however (and remains) controversial.

It is obvious that any immunity from tort liability, whether general or specific, obstructs all of the social functions identified above. Those injured cannot claim compensation, and any deterrence or accountability effects of liability are lost. Two further points should be noted. First, that such immunity may nonetheless be justified, where the effects of liability for society as a whole would be so serious that removing an injured individual’s claim is the lesser evil. But we must be clear that individual hardship is being inflicted for the wider public good. [10] Secondly, that the removal of tort liability’s valuable social effects concentrates attention on the other means by which the same social functions can be performed. In the present context, the AFCS provides an alternative route to compensation for deaths and injuries in military service. [11] Accountability for (and so the deterrence of) military errors is expressed through Parliament generally and the Defence Committee specifically; through coroner’s inquests into deaths and through other (ad hoc) public inquiries; through the investigation of complaints made to the Service Complaints Commissioner for the Armed Forces; and even (potentially) through courts-martial. Also, in the ordinary courts, immunity from tort actions would not exclude judicial review of MoD decisions, including review under the Human Rights Act 1998.

Liability for Fault

This is the general law of negligence, given considerable impetus for service personnel’s claims by Smith v. MoD. But despite their preliminary victory in the Supreme Court in Smith, the various claimants still face the considerable hurdle of showing that there was a negligent breach of duty on the facts of the cases. By definition, this kind of liability provides compensation only for victims whose injuries are the result of demonstrable fault. But while the fault principle limits the number of victims who can claim compensation, it can be considered to have a salutary effect on the conduct of defendants. Sanctioning negligence holds defendants to account by investigating whether they have been at fault. Obliging the payment of compensation when they have been seems likely to deter negligent behaviour, on some level. [12]

Strict Liability

Payment of compensation without fault is the salient feature for present purposes of the AFCS. The obvious advantage of such strict liability for victims of injury is that compensation does not depend on their proving fault; so more victims will receive payments. But where the level of payments is lower than at common law there will still be an incentive to bring negligence claims, hoping to prove fault to obtain higher ("full") compensation. Distributive concerns are raised by the generosity of compensation: is full compensation on a no-fault basis affordable? Why, in justice, should it be available for only one set of employees?

It might seem that strict liability cannot deter mistakes nor hold bodies accountable for them, since fault is not investigated (indeed for the defendant to prove the absence of fault is no defence). Such an assumption would, however, be erroneous. Strict liability provides an inherent incentive to take all reasonable (i.e. cost-justified) precautions, through the logic of cost-internalisation. Where risky activities cause injuries to others, making the activity bear ("internalise") the cost of the injuries provides a financial incentive to reduce their occurrence and severity, on the person or body carrying such activities out. The actor will realise that investing to reduce injuries to others is justified by rational self-interest under such a regime, until the point is reached when the cost of taking additional precautions outweighs the financial savings from the reduction in the injuries for which they are liable. In other words, making an activity internalise the cost of injuries to others means that it also internalises the benefits of accident prevention.

Any alternative approach "externalises" costs from the risk-generating activity onto the injured parties, subsidising the activity at their expense. This is clearly the position with tort immunities but also, in part, with negligence liability. In adjudicating on fault, the courts are in practice limited to deciding whether a particular activity was carried out negligently (e.g. negligent driving). They do not adjudicate on whether carrying out the activity in the first place was "negligent", even though it creates a risk of injuries (driving in general, or taking the particular journey which caused the accident). The result is that the level of the activity (and thus the number of injuries it causes) is higher under a negligence regime than under strict liability.

So, arguably, strict liability offers a superior form of deterrence (and avoids difficult questions about "fault"). Also, with public authority defendants whose expenditure tends to receive very close scrutiny (e.g. of the Ministry of Defence by the House of Commons Defence Committee), strict liability heightens the financial effects of injuries for which the defendant must pay. This makes injury prevention a more important aspect of overall financial accountability.

III. Smith v. Ministry of Defence

The Supreme Court Decision in Smith

A number of claims were heard together in Smith, one group relating to the use of allegedly inadequate Snatch Land Rovers and the other group arising from a "friendly fire" incident on (and by) a British Challenger tank. Some of the claims relied on common law negligence, other claims relied on the right to life under Article 2 of the European Convention on Human Rights (ECHR) as incorporated into English law by the Human Rights Act 1998 (HRA), and some relied on both negligence and Article 2. The Ministry of Defence (as defendant) had applied to have the claims struck out as disclosing no cause of action, or for summary judgment in its favour on the basis that the claims were bound to fail. The approach of Lord Hope’s majority Supreme Court judgment to these various issues was broadly similar. The various claims were allowed to proceed to trial (and the Ministry of Defence’s applications refused). Lord Mance (with the agreement of Lord Hughes) and Lord Carnwath delivered separate dissenting judgments.

Departing from an earlier decision, [13] the Supreme Court unanimously held that the ECHR applied in principle to the military operation of British armed forces outside the UK. On the substance of Article 2 ECHR, the majority stated that they accepted that battlefield decision-making and military procurement were "a field of human activity which the law should enter into with great caution", [14] and that the existence of "issues relating to the conduct of armed hostilities [which] are non-justiciable is not really in doubt". [15] But on the other hand:

a finding that in all circumstances deaths or injuries in combat that result from the conduct of operations by the armed forces are outside the scope of Article 2 would not be sustainable. It would amount, in effect, to a derogation from the state’s substantive obligations under that article. Such a fundamental departure from the broad reach of the Convention should not be undertaken without clear guidance from Strasbourg [i.e. the European Court of Human Rights]. [16]

No such clear guidance existed. Thus, upon striking the balance between individual rights (i.e. the claimants’ right to life) and the public interest (that the courts should not adjudicate on non-justiciable issues), "No hard and fast rules can be laid down. It will require the exercise of judgment. This can only be done in the light of the facts of each case." [17] It followed that the claims must proceed to trial and the issue of whether the MoD had breached the claimants’ Article 2 rights decided only after hearing all of the evidence. [18]

As for common law negligence, while accepting that there is a defence of "combat immunity", [19] the Court similarly held in the Snatch Land Rover cases that its applicability could not be determined on the bare statement of alleged facts in the pleaded claims. Since "the details that are needed to place the claims in context will only emerge if evidence is permitted to be led in support of them", the Snatch Land Rover claims had to proceed to trial before the combat immunity issue could be decided. [20] The Court again warned that while the "risk … of judicialising warfare" through negligence claims must "of course" be avoided, the facts of cases arising out of active military operations were very various and could not be "grouped under a single umbrella" as if that risk were equally serious in every case. [21] Therefore, in deciding whether a duty of care was owed in the first place (would liability be "fair, just and reasonable"?), as a separate question from combat immunity, the facts again had to be investigated first. Thus the court could weigh each claim in context against the consideration (of "paramount importance") that

the work that the armed services do in the national interest should not be impeded by having to prepare for or conduct active operations against the enemy under the threat of litigation if things should go wrong. [22]

The Challenger (friendly fire) cases were determined in a similar way, save that the Court actually decided that no defence of combat immunity was available on the facts as pleaded and so that aspect of the MoD’s defence should not be allowed at the trial. The Court held that the Challenger claims related to training and equipment decisions that had been "sufficiently far removed from the pressures and risks of active operations" that it was "not to be unreasonable to expect a duty of care to be exercised", by planning, exercising judgment and thinking things though-"so long as the standard of care that is imposed has regard to the nature of these activities and to their circumstances". [23] This suggests a greater emphasis on the fact-specific question of whether those for whom the MoD was being held responsible (as employer) were actually negligent in the particular case, rather than the abstract question of whether such claims should be allowed to proceed at all. [24]

The thrust of all of the decision on negligence and Article 2 ECHR alike was that weighing the claimants’ rights (to life and to compensation for negligent injury) against the public interest that military operations not be impeded by judicial scrutiny had to take place on a factual, case-by-case basis. Although the Court did not (and could not) decide that any of the claims would succeed, it could not hold that any of them were bound to fail at this preliminary stage before the trial of fact had taken place.

Legal Basis of Smith: Common Law

The basis for the Supreme Court majority’s decision on common law negligence is frankly weak. In similar fashion to the judgment below of the Court of Appeal, [25] Lord Hope made no serious attempt to engage with the formidable barrage of case-law denying that (e.g.) the police owe a duty of care to the public to catch criminals (on the basis that this would deter the police from the effective discharge of their duties). [26] Lord Hope’s judgment merely notes the names of the relevant authorities (to record their citation by counsel for the defendant). [27] This passing mention leaves one entirely unclear as to why those leading cases were not thought applicable to the not dissimilar issues in Smith. Lord Carnwath (dissenting) said that he found the majority’s approach "difficult to understand" (which it surely is, because it is unexplained). [28] After a detailed account of the police cases and their reasoning, [29] Lord Carnwath suggested that the considerations against liability in Smith ("issues of vital national security raised by the preparation for and conduct of war") were if anything weightier than the "purely domestic policy concerns arising from police powers of investigation". [30] Lord Mance (dissenting) similarly observed that "The claims that the Ministry failed to ensure that the army was better equipped and trained involve policy considerations of the same character as those which were decisive in Hill, Brooks and Van Colle." [31]

The most charitable thing that one could say about the majority’s decision is that the case-law on public authority negligence liability is notoriously confused. Aside from the police cases, there have conversely been some notable high-profile relaxations, with the House of Lords proving more receptive to arguments that authorities owe duties of care in the immediate wake of the HRA. [32] More widely, there has been judicial abolition of the historic "immunities" [33] from suit of advocates presenting cases in court, [34] and now of expert witnesses. [35] But on the other hand, the police cases certainly do not stand alone. There have been equally high-profile recent affirmations of a negative judicial stance to public authority liability. [36] With such contradictory lines of authority it is easy to cherry-pick useful cases to support any given conclusion. That is, unfortunately, the best way of describing the selective engagement (really, the absence of serious engagement) with the case-law in the majority judgment of Lord Hope. A clear example is the citation of Lord Bingham’s hope that common law negligence will always march in step with liability under the HRA in the Van Colle case, [37] without recording that Lord Bingham was in a dissenting minority of one in that case with the majority holding that there was no common law duty of care irrespective of any rights under Article 2 ECHR! [38]

Legal Basis of Smith: ECHR

As noted above, there is no positive authority from the European Court of Human Rights (the Strasbourg Court) on the application of Article 2 in a situation such as Smith. As seen, the majority treated this as a reason for caution, and refused to strike out the ECHR claims. Article 2 could only be held inapplicable tout court to soldiers killed during active operations if there were clear Strasbourg authority to that effect. There was not.

Other judges have taken different views. Dissenting in Smith, Lord Mance was not willing to hold that Article 2 went any further than the law of negligence (inapplicable in his view, on the grounds of combat immunity and the non-justiciability of military decision-making). [39] The dearth of authority left the path open to reach the same conclusion on the ECHR: "In my opinion it is not possible to conclude that the Strasbourg court would hold that such matters are justiciable under the [European] Convention, any more than they are at common law." [40] Lord Mance expressed scepticism about the European Court of Human Rights’ having evolved what amounts to an "independent substantive law of tort, overlapping with domestic tort law, but limited to cases involving death or the risk of death" in the first place. [41] He concluded:

The prospect of the Strasbourg court reviewing the conduct of combat operations [by making the state liable for the death in combat of one soldier due to alleged negligence of his commander or of another soldier] seems to me sufficiently striking, for it to be impossible to give this question a positive answer. If the European Court considers that the Convention requires it to undertake the retrospective review of armed conflicts to adjudicate upon the relations between a state and its own soldiers, without recognising any principle similar to combat immunity, then it seems to me that a domestic court should await clear guidance from Strasbourg to that effect. [42]

Lord Carnwath pointed out that sending the cases in Smith to trial would not resolve the difficulty: "if the problem is a lack of directly relevant guidance from Strasbourg, it is hard to see how, simply by hearing further evidence or finding further facts, [the trial judge] will be better able to fill that gap, still less to do so ‘with complete confidence’." [43]

Lord Mance also emphasised the continuing relevance of Lord Brown of Eaton-under-Heywood’s "evident scepticism" when, in an earlier case, Lord Brown had rhetorically asked:

Is it really to be suggested that … Strasbourg will scrutinise a contracting state’s planning, control and execution of military operations to decide whether the state’s own forces have been subjected to excessive risk (risk, that is, which is disproportionate to the objective sought)? May Strasbourg say that a different strategy or tactic should have been adopted-perhaps the use of airpower or longer-range weaponry to minimise the risk to ground troops notwithstanding that this might lead to higher civilian casualties? [44]

Lord Mance noted another point. If domestic UK courts were to construe the ECHR narrowly against an individual claimant’s right, he or she would still be able to go to the Strasbourg Court to argue for a wider construction of the Convention rights. But if domestic courts construe the Convention generously against their state, it has no mechanism for "appealing" beyond the UK Supreme Court against the (arguably over-wide) construction of the rights. Lord Mance said he was "not over-enamoured" of this "one-way street" argument. [45] But it is still a very proper reason cautioning against over-expansive readings of the Convention by domestic courts, ahead of anything that the Strasbourg Court has recognised.

Connexion between Negligence and Human Rights Reasoning

Orthodoxy prior to Smith was established by the House of Lords’ decision in Van Colle. The fact that there might be infringement of a Convention right does not of itself give rise to any claim in negligence at common law. [46] The claims may run in parallel channels (NB Lord Mance’s comments above about Article 2 jurisprudence having generated an independent law of quasi-tort). But they do not mix.

There have been suggestions that HRA adjudication has had a more pervasive influence on tort liability, in particular whether courts should decline to adjudicate altogether on non-justiciable issues. In a case decided the year before the House of Lords’ decision in Van Colle, Lady Justice Arden suggested that the preliminary question of whether a given matter was justiciable at all had "assumed less importance" in tort cases because of the HRA: "following the 1998 Act courts have now to consider questions of social policy with which they were not previously concerned. From this, in my judgment, it is possible to conclude that courts will hold that fewer matters are now non-justiciable [in negligence] on the grounds that they involve policy issues." [47]

It seems that notwithstanding Van Colle, [48] such reasoning may have influenced the majority decision in Smith. Both Lord Mance and Lord Carnwath specifically complained in their dissenting judgments that Lord Hope and the majority had given priority to the ECHR claims (followed, as discussed above, by a relatively brief treatment of common law negligence). Lord Carnwath stated that negligence ought to have been considered first, since "our primary responsibility should be for the coherent and principled development of the common law, which is within our own control", whereas "We cannot determine the limits of Article 2". [49] His Lordship suggested that the focus of counsel’s submissions on the ECHR jurisdictional question had "distorted" "the balance of the relevant issues" (with the unspoken implication that the majority judgment was similarly unbalanced). [50] So Lords Mance and Carnwath, having held (after lengthy discussion) that there was no duty of care in common law negligence, used this conclusion as the basis for dismissing an alternative claim under Article 2. Conversely, in the majority judgment, the decision that Article 2 liability could not be ruled out seemed to drive the conclusion that a negligence claim could not be denied either, by declaring there to be no duty of care.

Consequences of Smith: Practical and Constitutional

All the judgments in Smith recognise the potentially damaging practical consequences of extending legal liability too far. But the dissenting judges allege with good reason that having identified this trap, Lord Hope and the majority blundered into it regardless.

As Lord Hope himself says:

A court should be very slow indeed to question operational decisions made on the ground by commanders, whatever their rank or level of seniority. [51]

The allocation of resources to the armed services and as between the different branches of the services, is also a question which is more appropriate for political resolution than it is by a court. Much of the equipment in use by the armed forces today is the product of advanced technology, is extremely sophisticated and comes at a very high price. Procurement depends ultimately on the allocation of resources. This may in turn be influenced as much by political judgment as by the judgment of senior commanders in Whitehall as to what they need for the operations they are asked to carry out. It does not follow from the fact that decisions about procurement are taken remote from the battlefield that they will always be appropriate for review by the courts. [52]


subjecting the operations of the military while on active service to the close scrutiny that may be practicable and appropriate in the interests of safety in the barrack block or in the training area is an entirely different matter. It risks undermining the ability of a state to defend itself, or its interests, at home or abroad. The world is a dangerous place, and states cannot disable themselves from meeting its challenges. Ultimately democracy itself may be at risk. [53]

it is of paramount importance that the work that the armed services do in the national interest should not be impeded by having to prepare for or conduct active operations against the enemy under the threat of litigation if things should go wrong. [54]

All of this is a welcome recognition of the dangers of over-judicialisation. But while the rhetoric sounds well, does the actual decision in Smith live up to the strictures? It does not. As Lord Mance stated, "I do not consider that the majority approach reflects or meets this imperative" (i.e. the last statement quoted above). [55]

As seen above, the majority decision largely refused to decide whether Article 2 was engaged, or whether there was a duty of care at common law (either whether it would be "fair just and reasonable" or ruled out by "combat immunity"). All of these were, it held, questions that required the case to go to trial and for all the evidence to be heard before they could be resolved. The single exception to this studied indecision was the ruling that combat immunity was not an applicable defence in the Challenger claims. [56] Lord Carnwath complained, correctly, that the majority had failed to give "adequate" guidance to the eventual trial judge, and future courts faced with such issues: it was not enough to exhort the eventual decider to be "cautious", etc-"Having heard full argument on all these issues, [the Supreme Court] should be able to rule whether the claims are in principle viable or not". [57]

Lord Mance commented that the majority’s decision would lead to more litigation on such matters in the future. This is obviously true of the Smith claims themselves, which were allowed to proceed to trial, but Lord Mance predicted a much wider effect:

the approach taken by the majority will in my view make extensive litigation almost inevitable after, as well as quite possibly during and even before, any active service operations undertaken by the British army. It is likely to lead to the judicialisation of war… [58]

This would indeed seem "inevitable". If the highest court in the country is unable to say whether or not there is liability in principle without a full trial of the action, then no future case alleging negligence in active military operation can safely be decided in favour of the defendant (MoD) without such a trial. But as Lord Hope himself accepted, it is "the threat of litigation if things should go wrong" that may have an adverse effect on military decision-making. The majority’s decision, however, seems to make such litigation unavoidable. This is particularly true of the majority’s comment that breach of duty (i.e. fault) is to be given more emphasis than deciding (as an abstract proposition) whether a duty of care is owed. Fault can never be decided without detailed examination of the facts of the case. As Lord Mance says, negligence claims must "involve courts in examining procurement and training policy and priorities over years, with senior officers, civil servants and ministers having to be called and to explain their decisions long after they were made". The same point holds, of course, for commanders in claims based on failures in decision-making in the theatre of combat.

In short, it is difficult to see how the Supreme Court’s decision that the Smith claims should go to trial could not lead to the judicialisation that the majority themselves feared (albeit that the trial judge was counselled to be cautious in actually deciding both Article 2 and common law claims). Lord Mance (perhaps satirically) considered whether various military disasters of the past (from Isandlwana to the Fall of Singapore) might not have sparked litigation for negligent decision-making from planners, trainers and commanders, had they occurred after the decision in Smith. [59] He commented that exhorting trial judges to be cautious (as the majority had done) offered "no real solution". [60] Only by declaring that there was no liability even in principle could the damaging threat of litigation be avoided.

The point need not be made at any length here that that threat of "over-deterrent liability" must be taken seriously. This argument has had a chequered history in public authority cases generally down the years. As seen, it has consistently been accepted in the police cases. [61] But on other occasions, judges have declared that public servants are made of "sterner stuff" and will not be deterred from their duties by the threat of liability. [62] It has frequently been observed that judges are speculating here about the likely effect of liability on the behaviour of potential defendants. The implication is that overkill or over-deterrence should be ignored as a factor against liability unless there is hard empirical evidence that it has been produced by, or will be the effect of, negligence liability. But when the effects of over-deterrence might be as serious as skewing procurement priorities away from maximum military effectiveness towards a "safety first" approach, or, in battle, the inculcation of (literally) over-defensive strategy or tactics, such hard evidence is (we suggest) unnecessary. The real possibility of disastrous outcomes should satisfy the probative requirement. There is ample material to draw such a conclusion here. Air Chief Marshall Lord Stirrup (a former Chief of the Defence Staff) has recently told a House of Lords Committee that military personnel are increasingly concerned about their personal legal liabilities in wartime situations, concluding: "One of the potential consequences of this is not that you have fewer casualties; it is actually that you have more". [63] The Committee agreed that the "negative effect on the morale and operational independence of the armed forces" was a justified concern about "courts scrutinising operational decisions". [64]

As well as practical effects on military decision-making, the embrace of liability may have undesirable constitutional implications. Courts traditionally do not review inherently political decisions, [65] recognising that it is proper for the political process (primarily the accountability of Government ministers and departments to Parliament) to play this role. As the House of Lords Constitution Committee has recognised, the fear that military deployment decisions might be rendered justiciable is a strong reason against the legal formalisation of Parliament’s role in approving the deployment of the armed forces. [66] It is precisely because there is consensus that Parliament is the "the appropriate forum for controlling and scrutinising deployment decisions" that any change increasing "a risk of the domestic courts being invited to rule on the lawfulness of a deployment decision" should be avoided. [67]

In conclusion, the refusal of the Supreme Court in Smith to strike out the claims brings a real risk of defensive decision-making among military planners and commanders. Moreover, it inevitably requires judicial examination of sensitive matters of high national policy. As Lord Carnwath observed, these were the opposing horns of a dilemma for the claimants in Smith: the stronger they urged that the decisions challenged were taken prior to active operations, the more they implicated issues of planning and procurement ("discretionary decisions about policy and resources [that] are not justiciable"). [68] It is strongly arguable that for these equally unpalatable outcomes to be avoided, the decision in Smith urgently needs to be reversed.

IV. Reforming Smith

Is Legislation Necessary?

Of course the Supreme Court’s decision in Smith is not the final word on this question-not even in the Snatch Land Rover and Challenger cases themselves, where the questions of duty of care and breach of duty in negligence, and breach of Article 2 ECHR, await determination at trial. Naturally the MoD remains free to argue strenuously against liability on the facts of these and future cases (by denial of duty and breach, invocation of "combat immunity", and arguing for a narrow application of Article 2). But given the tenor of the majority decision in Smith, it is most unlikely that any trial judge (or the Court of Appeal) will lay down any general principles limiting liability under either heading. Indeed, it is most unlikely that the Supreme Court will revisit the question any time soon when it has decided the matter (to its own satisfaction at least) by a seven-judge panel, albeit by a bare majority. As noted above, the MoD will not be able to challenge what seems to be a remarkably broad interpretation of Article 2 before the European Court of Human Rights. Thus, irrespective of whether particular cases succeed on the facts, they will surely continue to be brought and thus the prospect of the military "having to prepare for or conduct active operations against the enemy under the threat of litigation if things should go wrong" remains. [69]

It seems highly unlikely therefore, in the absence of an unprecedented volte face by the Supreme Court, that the mischief of Smith will be removed by judicial action (even with a concerted strategy of contesting liability by the MoD). If liability in negligence is to be denied, a legislative solution is needed. In fact a ready-made procedure exists in s.2 of the Crown Proceedings (Armed Forces) Act 1987 for the revival, by ministerial order, of Crown immunity in cases involving the armed forces. This will be considered below.

While such legislation could be attacked on the grounds of constitutional principle (as a breach of the Rule of Law), such criticisms would be weak. Every time the police exercise their powers of arrest and detention they are being immunised by legislation (or by their remaining common law powers) from what would otherwise be liability in the torts of battery and false imprisonment. So legislative immunity for public authorities from what would otherwise be common law torts is widespread and uncontroversial.

It may be worth noting one argument that positively favours legislation. It is sometimes suggested that judges should concern themselves only with (rights-based) "principle" and eschew (consequentialist) "policy" in tort adjudication, because they are not well equipped to consider the latter. Although a minority position, it has occasionally attracted judicial support, [70] and rather more often that of academic purists. [71] If this argument is correct it fortifies the conclusion above that tort claims arising out of military action will not (indeed cannot, properly) be limited by judicial denial on the grounds of damaging constitutional and operational consequences. The proper way to place limits on the claims and rights that service personnel would otherwise have at common law is through legislation.

The legislative removal of Article 2 claims poses more formidable challenges. States have the power to derogate from the ECHR but only "In time of war or other public emergency threatening the life of the nation" and then only "to the extent strictly required by the exigencies of the situation" (Article 15(1), ECHR). It is obvious that such a derogation could not properly be made in terms wide enough to deal with the general problem of claims against the military posed by Smith v. MoD. The courts will, of course, review the legality of such a derogation and although deference will be shown to the judgment of the political arms of government in making it, [72] the argument that Smith-type claims against the military under Article 2 represent an emergency threatening the UK would surely not survive scrutiny. Of course Parliament might repeal the HRA altogether, although this would not prevent individuals claiming breach of their Article 2 rights in the European Court of Human Rights. To preclude that would necessitate repudiation of the entire Convention by the United Kingdom. There are respectable arguments for both of those courses of action. But even the most ardent opponent of human rights legislation would have to admit (in best Jim Hacker style) that to repeal the HRA just to reverse Smith v. MoD would use a sledgehammer to crack a nut by allowing the tail to wag the dog. At any rate, repeal of the HRA is a much larger question that cannot be adequately addressed in the present paper. Whether a generous enough no-fault compensation scheme (in place of negligence liability) would forestall HRA claims is considered below.

Section 2, Crown Proceedings (Armed Forces) Act 1987

This provision allows the Secretary of State to revive the effect of s.10 Crown Proceedings Act 1947 (which had preserved the common law immunity of the Crown in respect of armed forces claims, but which has now otherwise been repealed by s.1 of the 1987 Act). Such an order is made by statutory instrument subject to negative annulment procedure in both Houses (s.2(5)). Any revival takes effect "either for all purposes or for such purposes as may be described in the order" (s.2(1)(a)). The purposes may be described "by reference to any matter whatever and may make different provision for different cases, circumstances or persons" (s.2(3)). Any order cannot have retrospective effect (s.2(4)). The condition for making such an order is that it "appears" to the Secretary of State "necessary or expedient" either: "by reason of any imminent national danger or of any great emergency that has arisen" (s.2(2)(a)) or: "for the purposes of any warlike operations in any part of the world outside the United Kingdom or of any other operations which are or are to be carried out in connection with the warlike activity of any persons in any such part of the world" (s.2(2)(b)).

It appears that no order has been made to date under s.2 of the 1987 Act. But it could clearly be used to reverse the effect of Smith as far as common law negligence, insofar as the conditions in s.2(2)(b) are met. That is to say, that it appeared to the Secretary of State necessary (or expedient) for the effective conduct of warlike operations overseas (and connected activities) that claims by personnel on active service alleging negligence in the training, equipment and/or command of personnel should be barred. The exact terms of the statutory instrument would of course require very careful consideration so that the mischief of Smith was reversed without precluding legitimate claims that pose none of the same difficulties. The resulting order might face a challenge as being ultra vires s.2 or in some other way violating the principles of judicial review, but considerable deference would have to be shown by the reviewing court to the Secretary of State’s judgment of what "appears … expedient" on a matter of national security.

An order under the 1987 Act could not entirely remove the vice of Smith-for example it would not affect claims for injuries suffered inside the UK that alleged negligence in high-level procurement decisions. Such claims still involve courts determining non-justiciable issues, albeit outside the theatre of war. But an order would provide a clear legislative restatement of the supposedly uncontroversial general principle of combat immunity, to which the majority decision in Smith has given a narrow and context-specific definition.

Immunities Again

It seems that the Secretary of State could largely reverse Smith v. MoD by an order under the Crown Proceedings (Armed Forces) Act 1987, and obviously Parliament could achieve the same goal even more comprehensively by new primary legislation. But should this be done?

It was argued above that immunities prevent tort law from performing its functions of compensation, deterrence and accountability. It may be that the public interest in ensuring the effectiveness of the armed forces is so great that individual rights (their claims to tort compensation) can justifiably be sacrificed to protect that public interest. But we should still ensure that individuals’ interest in compensation for their injuries, and the wider social functions of tort law, are still promoted so far as possible.

It was argued above that strict liability offers certain advantages as a deterrent, through the mechanism of cost-internalisation. It has been argued that fault-based liability is unwise in the Smith situation because adjudicating upon negligence would over-deter important public functions (effective defence of the realm) and draw the courts into non-justiciable questions. Therefore, the answer seems to be to replace Smith not with a complete absence of liability (an immunity, with nothing in the place of negligence) but rather with a form of strict liability. This should ensure that the Ministry of Defence still has a keen financial incentive to take all cost-justified precautions to prevent deaths and injuries of forces personnel, while ensuring compensation for the latter without the hurdle of proving fault.

It is then suggested that Smith v. MoD be reversed by an order under the Crown Proceedings (Armed Forces) Act 1987, but only if the negligence claims in Smith are to be replaced by compensation payable to the dead and wounded in battlefield situation on a no-fault basis.

No-Fault: The Armed Forces Compensation Scheme

The Armed Forces Compensation Scheme is the obvious existing source for such no-fault compensation. But clearly the Scheme at least sometimes continues (despite recent reforms) to be less generous than tort damages assessed to provide restitutio in integrum. Otherwise negligence claims (with their expense, delay and stress) would never be brought by service personnel. Indeed, the Government’s rationale for welcoming what became s.1 of the Crown Proceedings (Armed Forces) Act 1987 (allowing tort claims against the armed forces for the first time) was that damages awarded by the courts in personal injuries cases had risen considerably above the compensation provided by (the forerunner of) the AFCS. In fairness therefore, military personnel ought be allowed to bring tort claims as any civilian might against his or her employer. [73] But logically, the other answer to this problem would be to raise the level of compensation under the AFCS to parity with tort damages. The need for tort claims by personnel against the Ministry of Defence would then disappear. AFCS claimants would be no worse off in terms of compensation levels, and in fact significantly better off overall since they do not have to show that their injuries resulted from anyone’s negligence. (Thus, as noted above, averting the over-deterrence and non-justiciability problems of Smith-style negligence claims.)

The feasibility of raising AFCS benefits to exact parity with tort damages is, however, doubtful. First there is the obvious problem of cost to the Ministry of Defence, especially if the common law approach were adopted across the board (i.e. for all deaths and injuries attributable to service in the armed forces). NB that the Government should absorb the full cost of military injuries caused to provide the correct incentive to take all cost-justified precautions, so an increase in the total cost of AFCS would have deterrence as well as (obviously) compensation benefits. But if it proved politically impossible to fund a general uprating of all AFCS payments, a special exception could be considered for injuries and deaths sustained on active service (i.e. such claims only would be entitled to full, tort-level compensation under AFCS).

The current scheme makes no differential in levels of payments between injuries sustained on active service and otherwise. The recent Boyce Review of the AFCS decided against any change to this principle, on the basis of equity of treatment. It is the fact of agreeing to serve and (thereby) signalling willingness to sacrifice that makes service injuries different from others but similar to each other; and since personnel often have no choice about whether they are posted on active operations or not, an "all of one company" approach (not differentiating the circumstances of the injury) remains appropriate. [74] The recommendations of the Boyce Review (which were all accepted by the Government) are, no doubt, quite correct under the existing regime of tort liability. But if the claims of those on active service were to be ruled out by an order under s.2, Crown Proceedings (Armed Forces) Act 1987 as recommended above, there would be a strong reason to give such injuries higher compensation under the AFCS. The Scheme would become the substitute for and not merely an alternative to tort compensation, for such claimants.

A more fundamental obstacle exists to using the AFCS as a means of compensating injuries on the full (tort) measure, even in a subset of active service claims. The AFCS is, like all such schemes, essentially based on generalised tariffs and tables, in sharp contrast to the personalised approach of tort law. Tort’s restitutio in integrum principle tailors compensation to the precise losses and needs of each individual claimant. The AFCS does not provide such a bespoke service but rather offers a range of off-the-peg compensation amounts. The AFCS has two basic elements. First, a lump sum from a tariff graduated according to severity of injury; this equates to the pain and suffering (loss of amenity) element of tort damages, or the non-pecuniary loss. The equivalent of pecuniary loss under the AFCS is a regular "guaranteed income payment" based on the salary of the claimant at their point of his or her leaving the services (when the injury is severe enough to have caused this). It is tax-free and index-linked. Notably, however, there is no award for medical, nursing or other care needs under the AFCS: these needs are supplied through the usual state services (NHS, etc). That major exclusion from AFCS compensation has recently been approved by the Boyce Review, rationalised as a "cross-Government approach" to injury support. [75] But it may well be that personnel left with severe disabilities feel compelled to bring tort claims to get compensation to pay for (e.g.) accessibility modifications to their houses, when such compensation is excluded from the AFCS and may not readily be available under the general social security system.

Leaving aside the latter issue, it would be impossible to adopt the restitutio in integrum approach without utterly transforming the AFCS: its very nature (as a tariff-based scheme) and therefore the mode of its administration. The Boyce Review recommended against the introduction of any significant "personal element" into the AFCS. It would make consistent decision-making difficult, and it would raise transparency and fairness concerns (it might not be obvious why different claimants received different payments without the disclosure of personal factors-but confidentiality considerations would prevent such disclosure). [76] Certainly, to expect the Whitehall staff who currently administer the tariff-based AFCS to move seamlessly to quantifying compensation on the tort approach would ask much of them. It would also be a reinvention of the wheel. The tort system (including the judges, lawyers and others who already administer it) possesses all the necessary knowledge.

No-fault Compensation on the Full Tort Quantum

Instead of attempting to transform the AFCS into a para-tortious compensation system, it would instead be neater for the Government simply to agree to pay full, tort-quantum compensation on a no-fault basis (for injuries in service that cannot proceed as tort claims because of an order under s.2 of the Crown Proceedings (Armed Forces) Act 1987). This would operate outside AFCS for the reasons just given.

The commitment to pay compensation without proof of negligence could and ideally should be embodied in legislation. Most conveniently, it could be included in the statutory instrument implementing the order under s.2 of the 1987 Act. However, s.2 does not seem to contemplate such extensive conditions being including in orders made under it, and there would be a serious danger that any no-fault damages provisions included in the statutory instrument would be ultra vires s.2. In the absence of primary legislation to authorise the payment of no-fault, tort-level compensation, the Government could then simply declare itself bound to make such payments "under the prerogative", as the Criminal Injuries Compensation Scheme was originally promulgated in 1964. [77] This could be stigmatised as merely an ex gratia concession, and one that the Government might withdraw at any time. But in political reality, any Government with a Parliamentary majority is not bound even by primary legislation. [78] The original CICS scheme was no more, but certainly no less, robust than its statutory successor. There is no reason to believe that a Government commitment to full compensation for wounded personnel, as the explicit quid pro quo for restriction of their tort law rights using s.2 of the Crown Proceedings (Armed Forces) Act 1987, would be any less enduring. Political pressure would help ensure this.

Compensation and Human Rights

If the suggestion above were taken up, and full compensation were to be paid on a no-fault basis for all injuries sustained in combat, then a major impetus encouraging the continued initiation of Article 2, ECHR claims would fade. This is important because, as seen, s.2 of the Crown Proceedings (Armed Forces) Act 1987 only applies to common law negligence (pre-1947 Crown immunity is simply irrelevant under the HRA). Injured personnel, or the dependents of those killed in service, would no longer need to bring Article 2 claims (which at the moment are usually brought in tandem with negligence claims) as a means of getting compensation, under the proposal above. It is true that some might still wish to bring claims under the HRA to ensure vindication of Article 2 rights and to hold the Government to account for breach of them. But the Government could plausibly argue that it would be unnecessary for a court to hear the claim (requiring adjudication upon military/procurement decision-making discussed above) when the Government had already made full financial satisfaction. It seems unlikely that damages awarded under s.8 HRA will exceed the level of tort damages; if anything the practice to date has been to award more modest awards in HRA cases. [79]

In short, payment of full compensation on a no-fault basis would in practice preclude most Article 2 claims from being brought in cases like Smith v. MoD. Such compensation might even provide a basis for the courts to decline to hear any such cases on the basis that full satisfaction had already been provided by the state.

V. Recommendation

In conclusion it is recommended both:

(1) That an order be made under s.2 of the Crown Proceedings (Armed Forces) Act 1987 to clarify the scope of combat immunity in tort (which has been left doubtful, and attenuated, by the Supreme Court’s decision in Smith v. MoD).

(2) That in place of the tort claims thus removed, and instead of attempting to weave the tort principle of restitutio in integrum into an enhanced Armed Forces Compensation Scheme, the Secretary of State should declare that compensation will be paid on the full tort quantum to all service personnel who suffer death or injury but have no claim pursuant to the s.2 order, on a no-fault basis.


Jonathan Morgan, "Negligence: Into Battle" [2013] Cambridge Law Journal 14-17

CAN soldiers killed or injured during combat sue the Ministry of Defence for failing to protect them? At first blush this sounds like the latest in that series of questions to which the terse answer is "no". Such claims, in negligence, have previously been given short shrift: Mulcahy v Ministry of Defence [1996] Q.B. 732 (which P.S. Atiyah said was "surely entitled to the prize for the most undeserving claim of the decade (which is saying something)": The Damages Lottery (Hart, 1997), p. 90). On the other hand, the Ministry clearly owes duties to its employees both at common law and under the Health and Safety at Work Act 1974, as confirmed in cases concerning injuries during military training exercises (e.g., Chalk v MoD [2002] EWHC 422 (QB) and Fawdry v MoD [2003] EWHC 322 (QB)). Furthermore, since the claim in Mulcahy was dismissed, the Human Rights Act 1998 has imposed new duties upon the Government. Might the line in the sand now be crossed?

Smith v MoD [2012] EWCA Civ 1365 concerned soldiers wounded or killed during the Iraq war. There were two groups of incidents and claims. In the first, numerous "Snatch" Land Rover vehicles (which were notoriously lightly armoured) had been attacked using "improvised explosive devices". In the second incident, a tank from a different regiment of the British army had shelled the claimant soldiers (mistaking their identity). The claimants sought to rely upon the MoD's obligation to safeguard their right to life under Article 2 of the European Convention on Human Rights, or upon common law negligence, or both. The gist of the alleged breaches was a failure to provide suitable equipment (properly armoured Land Rovers; automatic recognition systems to guard against "friendly fire") or adequate training in vehicle recognition.

*C.L.J. 15 Our focus here will be upon the common law claims. The European Convention was held inapplicable in accordance with the decision in Regina (Smith) v Oxfordshire Assistant Deputy Coroner [2010] UKSC 29, [2011] 1 A.C. 1. (It was noted that a Strasbourg challenge was pending against Smith v Oxfordshire in Pritchard v U.K., but that any reconsideration would have to await the European Court's judgment.) The starting point for Moses L.J. (with whom Rimer L.J. and Lord Neuberger M.R. agreed) was the Ministry's duty qua employer to provide a safe system of work for the claimant soldiers. This was too well established to be disputed. The question was whether that duty yielded to the "combat immunity" relied upon in Mulcahy. In the end, the Court of Appeal held that that was a question of fact which would have to be determined at the trial of the action. Accordingly, the judge below (Owen J.) had been wrong to strike the claims out on the basis of "combat immunity" (cf. [2011] EWHC 1676 (QB)).

This sounds like an un-illuminating classification of the central issue as one of "fact". But Moses L.J. gave some guidance on the proper scope of "combat immunity". It was not sufficient that the injuries in question were sustained during battle (otherwise all of these claims would, necessarily, have failed). The question was whether the supposedly negligent decisions were taken during "active operations". Decisions about training and equipment taken some time before the conflict in question could not enjoy "combat immunity". Otherwise, it would be "difficult to see how anything done by the Ministry of Defence" would fall beyond it (at [62]). So decisions "away from the theatre of war" would not enjoy the immunity, which was to be narrowly construed. Only if the court would be required to sit in judgment on decisions made in the course of active operations would "combat immunity" bar claims.

There are good constitutional grounds for this narrow approach. As Elias J. pointed out in Bici v MoD [2004] EWHC 786 (QB) the successful invocation of "combat immunity" hinders the court's "historic and jealously guarded role of determining [when] rights have been unlawfully infringed by an act of the executive". His Lordship cited the great case of Entick v Carrington (1765) 19 Howell's State Trials 1029 to show that the Government may not "simply assert interests of state or the public interest and rely upon that as a justification for the commission of wrongs". This is stirring stuff, and important. There has been public disquiet about the alleged underfunding of Mr Blair's wars by his Chancellor of the Exchequer. The Ministry of Defence should not be permitted to hide failures to fund vital protective equipment under a cloak designed to protect battlefield decisions against judicial questioning.

*C.L.J. 16 But even assuming that the decisions did not fall within "combat immunity", is it proper for them to be scrutinised by the courts in an action for damages? In Smith the Ministry argued not, although unsuccessfully. Military procurement (involving decisions about the allocation of scarce resources) was said to be a political matter for which ministers were answerable exclusively to Parliament. The courts should not trespass into such matters: they were non-justiciable. Lord Rodger had said as much in Smith v Oxfordshire (at [127]). Moreover, arguments that resource allocation is non-justiciable had prevailed in the past, in judicial review cases (e.g., Regina v Cambridge Health Authority, Ex parte B [1995] 1 W.L.R. 898). Also the courts had consistently protected the autonomy of, for example, the police to decide how best to fight crime by denying a duty of care, ever since Hill v Chief Constable of West Yorkshire [1989] A.C. 53). So why did the argument fail?

Moses L.J. held that while justiciability would be relevant if considering a "novel" duty of care (as in Hill), it simply did not arise in the case before him when the duty qua employer was so well established. But, with respect, this seems a slender ground for distinguishing Hill. Earlier cases on the MoD's employer liability may well have been decided without even considering justiciability. The historical accidents of legal development provide no sure reason to ignore the justiciability argument when it does arise and is clearly relevant, especially given its support in the authorities.

Secondly, however, Moses L.J. relied on quite a different line of authority. He cited Barrett v Enfield L.B.C. [2001] 2 A.C. 550 and Phelps v Hillingdon L.B.C. [2001] 2 A.C. 616 to show that "the mere fact that questions might arise as to policy, and as to the allocation of scarce resources, did not preclude the existence of a duty to take care" (at [48]). Rather, these matters should be taken into account in tailoring the standard of care to be applied (cf. Bolam v Friern Hospital Management Trust [1957] 1 W.L.R. 582).

This is rather surprising. Barrett and Phelps were both decided in the febrile months following Osman v UK (2000) 29 E.H.R.R. 245, when the courts became most reluctant to strike out any claim on duty of care grounds lest they be held to have breached Article 6 of the ECHR. But once the Osman heresy had been corrected in Z v UK (2001) 34 E.H.R.R. 97, the House of Lords reverted to its previous approach, routinely denying duties of care. The non-justiciability argument in Smith v MoD (which involved funding national defence procurement) was anyway much stronger than that in Barrett or Phelps (which involved decisions by social workers and educational psychologists). Moreover, Moses L.J.'s favoured strategy of controlling liability by means of a variable standard of care (i.e., *C.L.J. 17 breach rather than duty) was more recently championed entirely unsuccessfully by the late Lord Bingham, dissenting in JD v East Berkshire Community Health NHS Trust [2005] UKHL 23, [2005] 2 A.C. 373 and Smith v Chief Constable of Sussex; Van Colle v Chief Constable of Hertfordshire [2008] UKHL 50, [2009] 1 A.C. 225. What Lord Bingham tried in vain the Court of Appeal in Smith v MoD has now been accomplished, by declining to engage with those authorities at all.

Public authority tort liability is notorious for complexity. For it is a tricky business to weigh up the competing constitutional concerns: the state should not claim sweeping immunities for its (otherwise tortious) actions (e.g., Entick v Carrington); but the courts should not second-guess matters of high policy for which politicians should properly be accountable to Parliament. Yet if ministerial responsibility is seen to be "falling short", this should be addressed directly; it would be unwise for the judiciary to fill the "vacuum" (cf. Regina v Home Secretary, Ex parte Fire Brigades Union [1995] 2 A.C. 513, 567 per Lord Mustill). By contrast with such inherent problems, needless complication arises from incompatible lines of case-law. One might have believed that Barrett and Phelps had joined Junior Books v Veichi [1983] 1 A.C. 520 in "the slumber of the uniquely distinguished" (cf. The Orjula [1995] 2 Lloyd's Rep. 395 per Mance J.). But Smith v MoD has awoken them once more. The Supreme Court may yet restore order (an appeal is to be heard in February 2013).

November 2013

[1] Cf. Rowling v Takaro Properties Ltd [1988] AC 473, 504 per Lord Keith of Kinkel (cited in Smith v. MoD [2013] UKSC 41, [127] per Lord Mance).

[2] Tomlinson v. Congleton Borough Council [2004] 1 AC 46, [94] per Lord Scott of Foscote. See also last footnote.

[3] See e.g. Lewis R, Morris A and Oliphant K, "Tort Personal Injury Claims Statistics: Is There A Compensation Culture In The United Kingdom?" [2006] Journal of Personal Injury Law 87.

[4] AV Dicey, An Introduction to the Study of the Law of the Constitution (1885).

[5] Hill v. Chief Constable of West Yorkshire [1989] A.C. 53, 64 per Lord Templeman.

[6] Cf. ibid, 64-65.

[7] [2013] UKSC 41; [2013] 3 W.L.R. 69.

[8] Armed Forces (Pensions and Compensation) Act 2004; Armed Forces and Reserve Forces (Compensation Scheme) Order 2011.

[9] NB [2013] UKSC 41, [89] per Lord Hope (the doctrine’s existence is "not in doubt") (judgment of majority).

[10] Cf. Van Colle v. Chief Constable of the Hertfordshire Police [2008] UKHL 50, [75] per Lord Hope of Craighead; [139] per Lord Brown.

[11] The argument that providing no-fault compensation also has a valuable deterrent effect is developed below.

[12] Cf. discussion of Deterrence above, and of the specific effects of Smith v. MoD, below.

[13] R (Smith) v Oxfordshire Assistant Deputy Coroner [2011] 1 AC 1.

[14] [2013] UKSC 41, [64]-[66].

[15] Ibid, [58].

[16] Ibid.

[17] Ibid, [76].

[18] Ibid, [78]-[80].

[19] [2013] UKSC 41, [89] per Lord Hope (the doctrine’s existence is "not in doubt")

[20] Ibid, [96].

[21] Ibid, [98].

[22] Ibid, [100].

[23] Ibid, [95].

[24] Ibid, [98]-"the question in the case of the Challenger claims is not whether a duty was owed but whether, on the facts, it was breached".

[25] [2012] EWCA Civ 1365, criticised by the current author: J Morgan, "Negligence: Into Battle" [2013] Cambridge Law Journal 14-17. (This case-note was cited by Lord Carnwath [2013] UKSC 41, [170]. It is included as an appendix to the current paper.)

[26] E.g. Hill v. Chief Constable of West Yorkshire [1989] AC 53 (HL), Brooks v. Commissioner of Police of the Metropolis [2005] UKHL 24, Van Colle v. Chief Constable of Hertfordshire Police [2008] UKHL 50.

[27] [2013] UKSC 41, [97].

[28] [2013] UKSC 41, [170].

[29] [2013] UKSC 41, [167]-[169].

[30] [2013] UKSC 41, [170].

[31] [2013] UKSC 41, [128].

[32] Cf. Barrett v. Enfield London Borough Council [2001] 2 AC 550 and Phelps v. Hillingdon London Borough Council [2001] 2 AC 619, discussed and relied upon in the decision below in Smith [2012] EWCA Civ 1365.

[33] As they have been misdescribed: cf. Barrett v Enfield London Borough Council [2001] 2 AC 550, 559 per Lord Browne-Wilkinson, cited by Lord Carnwath in Smith [2013] UKSC 41, [163].

[34] Arthur JS Hall & Co v. Simons [2002] 1 AC 615.

[35] Jones v Kaney [2011] UKSC 13, cited by Lord Hope in Smith [2013] UKSC 41, [94].

[36] E.g. Gorringe v. Calderdale Metropolitan Borough Council [2004] UKHL 15, JD v. East Berkshire Community NHS Trust [2005] 2 AC 373, Jain v. Trent Strategic Health Authority [2009] 1 AC 853, Mitchell v. Glasgow City Council [2009] 1 AC 87 (all decisions of the House of Lords).

[37] [2013] UKSC 41, [98].

[38] Van Colle v. Chief Constable of Hertfordshire Police [2008] UKHL 50.

[39] [2013] UKSC 41, [151].

[40] [2013] UKSC 41, [143].

[41] [2013] UKSC 41, [142].

[42] Ibid.

[43] [2013] UKSC 41, [156].

[44] R (Smith) v Oxfordshire Assistant Deputy Coroner [2011] 1 AC 1, [146], cited in Smith v. MoD [2013] UKSC 41, [130].

[45] [2013] UKSC 41, [142].

[46] Van Colle v. Chief Constable of Hertfordshire Police [2008] UKHL 50: contrast the judgment of the Court of Appeal in the same case, sub nom. Smith v. Chief Constable of Sussex [2008] EWCA Civ 39.

[47] Jain v. Trent Strategic Health Authority [2008] Q.B. 246, [62].

[48] NB criticisms above of the side-lining of the whole line of police case-law.

[49] [2013] UKSC 41, [156].

[50] [2013] UKSC 41, [155].

[51] [2013] UKSC 41, [64].

[52] [2013] UKSC 41, [65].

[53] [2013] UKSC 41, [66].

[54] [2013] UKSC 41, [100].

[55] [2013] UKSC 41, [147].

[56] Cf. on this point Lord Mance [2013] UKSC 41, [125].

[57] [2013] UKSC 41, [154].

[58] [2013] UKSC 41, [150].

[59] [2013] UKSC 41, [133].

[60] [2013] UKSC 41, [134].

[61] Hill v. Chief Constable of West Yorkshire [1989] AC 53 (HL), Brooks v. Commissioner of Police of the Metropolis [2005] UKHL 24, Van Colle v. Chief Constable of Hertfordshire Police [2008] UKHL 50.

[62] Dorset Yacht Co v. Home Office [1970] A.C. 1004, 1033 per Lord Reid (prison officers). See also Capital and Counties Plc v. Hampshire CC [1997] Q.B. 1004 (fire brigade).

[63] House of Lords Constitution Committee, Second Report, 2013-4 Session, "Constitutional arrangements for the use of armed force", para 55.

[64] Ibid.

[65] See e.g. A v. Secretary of State for the Home Department [2005] 2 AC 68, [29] per Lord Bingham.

[66] House of Lords Constitution Committee, Second Report, 2013-4 Session, "Constitutional arrangements for the use of armed force", para 50-61.

[67] Ibid para 54. Smith v. MoD was cited to illustrate this danger, ibid para 56.

[68] [2013] UKSC 41, [161].

[69] [2013] UKSC 41, [100].

[70] E.g. Lord Scarman in McLoughlin v O’Brian [1983] 1 AC 410.

[71] E.g. R Stevens, Torts and Rights (Oxford 2007) ch 14.

[72] See e.g. A v. Secretary of State for the Home Department [2005] 2 AC 68.

[73] Hansard HC Deb 8 December 1986, vol 107 cc85-86W (Secretary of State for Defence). Cited by Lord Carnwath in Smith v. MoD [2013] UKSC 41, [178].

[74] Ministry of Defence, Review of the Armed Forces Compensation Scheme, Cm 7798 (2010), 2.83-2.88.

[75] Ibid 2.26.

[76] Ibid 2.23.

[77] For differing views on whether this was truly an exercise of the royal prerogative compare Regina v. Criminal Injuries Compensation Board, Ex parte Lain [1967] 2 Q.B. 864 with (e.g.) HWR Wade, "New vistas of judicial review" (1987) 103 LQR 323. The Scheme is now on a statutory basis, cf. Criminal Injuries Compensation Act 1995.

[78] Compare Criminal Justice Act 1988 ss.108-117 and Regina v. Secretary of State for the Home Department, Ex parte Fire Brigades Union [1995] 2 A.C. 513.

[79] See e.g. Regina (Faulkner) v Secretary of State for Justice [2013] UKSC 23. For criticism cf. J Varuhas, "A tort-based approach to damages under the Human Rights Act 1998" (2009) 72 MLR 750.

Prepared 13th January 2014