(a) Dunnage v Randall & UK Insurance Ltd

Before going further, it is important to consider the decision made in Dunnage v Randall, as this case promises to be the leading authority on the question of involuntariness and the objective expectations of the duty of care.Footnote ⁵ On 24 October 2007, Vince Randall covered himself in petrol whilst standing in his nephew's kitchen. His nephew, Terry Dunnage, tried to stop him. Vince resisted and during the struggle both Vince and Terry caught fire. In flames, Terry threw himself out of the window – saving his own life but suffering serious injuries. Vince did not – he died from his injuries. Vince suffered from florid paranoid schizophrenia. Terry sued the estate of Vince – his uncle had household insurance which provided cover in the event of accidental damage or injury but not for intentional injury or losses. The insurer argued Vince was not acting voluntarily therefore there was no intention or negligence. At trial, the judge (HHJ Saggerson) formulated what he saw as the relevant legal question: ‘Did Vince act as a conscious agent deliberately purposefully or recklessly in setting the fire albeit driven by delusions, or was his freedom of thought and action so subverted by illness that his capacity to think and act freely was eliminated so that he was not the causative agent of events leading to the damage?’Footnote ⁶ He was not free, conscious or liable according to the HHJ Saggerson. However, the Court of Appeal disagreed with the trial judge's decision and his formulation of the relevant question. On 2 July 2015, they held that Vince acted voluntarily but that he did not mean to injure his nephew – in other words, he was negligent. Lady Justice Arden, Lady Justice Rafferty, and Lord Justice Vos each took a different approach but generally stressed that a defendant's ability to reason or act needs to be completely overpowered or disabled before you can say an agent acted involuntarily. In a sense, the decision confirmed the status quo: mental disorders very rarely negate legal responsibility or render an action ‘involuntary’ in the law of tort.

(b) Examining the traditional approach

In coming to its decision, the Court of Appeal was informed by numerous authorities from several jurisdictions, including the early seventeenth century case of Weaver v Ward Footnote ⁷ which was cited in Donaghy v Brennan.Footnote ⁸ It is no surprise that Weaver is often cited; it is the traditional starting point for an analysis of pleas of insanity and, importantly, it is here that an articulation of involuntariness is found. An action of assault and battery was raised after the defendant's musket accidentally discharged during a skirmish with the plaintiff. It was reported that ‘the defendant casualiter & per infortunium & contra voluntatem suam, in discharging of his piece did hurt and wound the plaintiff, which is the same, &c. absque hoc, that he was guilty aliter sive alio modo’.Footnote ⁹ It is here that you find the now classic rejection of ‘insanity’: the report says ‘if a lunatic hurt a man, he shall be answerable in trespass; and therefore no man shall be excused of a trespass …’ But critically, it is remarked that such a condition may be applicable when ‘it may be judged utterly without his fault’. Such a situation is described by the court using examples rather than offering a definition:Footnote ¹⁰

At first blush, the situations mentioned in the quotation appear different from each other. One can be said to be an action of the defendant, ie running across the piece, whilst the other is not, ie a hand is forced to move. Even more challenging is that modern neuroscience tells us that incapacity can arise when a person loses their sense of control and not simply when their motor functions fail them. Such insights cut across the notion that incapacity relates to physical impediments alone. Moreover, there remain concerns that to acknowledge involuntariness within a negligence action comes dangerously close to undermining the objective expectations of the duty of care. Despite this uncertainty the case law is full of examples of involuntariness but bare on explanation or definition. It is useful therefore to gather what has been said.

(c) Recognising involuntariness

Both courts and academic commentary appear willing, albeit hesitantly, to recognise involuntariness as appropriate in some instances. For instance, in negligence actions courts have accepted that liability does not attach in situations such as unanticipated blackouts,Footnote ¹¹ sudden unconsciousness,Footnote ¹² a delusional belief that someone else is controlling your actions,Footnote ¹³ or an unmistakable breakdown between what the internal world directs, believing that you are God,Footnote ¹⁴ and the external world of action, what you can control.Footnote ¹⁵ In describing the outcome of Dunnage, Steele has said ‘it seems then that to avoid liability in negligence, there must be more than lost capacity to make meaningful choices: there must be a loss of capacity to act’.Footnote ¹⁶ Stevens too remarks, ‘there are extreme examples of mental incapacity which prevent liability from arising’.Footnote ¹⁷ In Markesinis & Deakin's Tort Law, the authors stress the importance of an act to the constitution of a tort claim. They say that, by and large, the primary elements of tort include conduct, causation, fault, and damage. Explaining what they mean by conduct, they remark that it can apply to both omissions and acts:

In Winfield & Jolowicz on Tort, the authors also acknowledge the importance of volition:

These decisions as well as the commentary give good cause for a more extensive investigation of what involuntariness is and, vitally, what it is not.

(d) Choice complicated by policy

Although notions of choice and control may be embedded within doctrine of involuntariness, it can become obscured by a court's consideration of policy. In such circumstances, the court becomes detained by the question of who should bear the losses which is relevant to the question of duty but not to the question of whether the defendant has the primary capability to perform the duty in the first place. For example, in Williams v Hays,Footnote ²⁰ the Supreme Court of New York considered the situation where a sleep-deprived, quinine-intoxicated, malaria-induced-delusional captain refused all help during a storm, eventually wrecking his master's vessel. Arguing that he was unconscious at the relevant time, he unsuccessfully contended that this mental disorder caused him to act without due care and diligence. On appeal, the issue was whether the circumstances of the storm had provoked the captain's condition and so rendered his actions beyond civil liability. Helpfully, the court summarised the defendant's argument succinctly, the captain too should be considered an innocent victim along with the plaintiffs.Footnote ²¹ In order to determine the case, both the lower courts and the Supreme Court took an evaluative approach, asking essentially who should bear the loss in such circumstances. The Supreme Court concluded by explaining:Footnote ²²

Rather than engage in the question of whether the delusions, malaria or quinine-intoxication amounted to a state of involuntariness, the court evaluated the facts based on how tort law should allocate the losses in these circumstances. Of course, this is a valid consideration in any such case, but it does not provide an answer as to whether involuntariness was present and if so, why that too should be disregarded. In effect, this left the question of involuntariness untouched – something which is commonplace throughout the otherwise relevant authorities.Footnote ²³

Nonetheless, the case law gestures towards a more substantive understanding of involuntariness. Take the Court of Appeal of Ontario's decision in Buckley and The Toronto Transportation Commission v Smith Transport Ltd, where Roach JA concluded that a mental illness meant ‘that [the defendant] did not understand the duty which rested upon him to take care, and further that if it could be said that he did understand and appreciate that duty, the particular delusion prevented him from discharging it’.Footnote ²⁴ Or for instance, in Mansfield v Weetabix, the Court of Appeal of England and Wales said that loss of control should not be simply equated to a loss of consciousness, saying ‘loss of control may or may not be accompanied by loss of consciousness’.Footnote ²⁵ More recently, for example, in Fiala v MacDonald, the Court of Appeal of Alberta said that liability should not be imposed when as a result of mental illness, ‘the defendant was unable to discharge his duty of care as he had no meaningful control over his actions at the time the relevant conduct fell below the objective standard of care’.Footnote ²⁶ Each case suggests that involuntariness does not necessarily equate to a physical incapacity to control actions but a medically recognised condition which impairs the capability to effect a choice.Footnote ²⁷ In Dunnage the effect of involuntariness was described but its meaning left unexamined. Thus, according to the Court of Appeal, it is something which ‘entirely eliminates responsibility’, which means if control is only eroded or diminished then liability affixes. So, evidence given by experts which qualified the diagnosis of complete elimination through the introduction of percentages, such as 95% elimination of control, undercut the assertion that the defendant's actions were involuntary.Footnote ²⁸ Further, drawing on Coley v R,Footnote ²⁹ the court in Dunnage also endorsed the distinction between the irrational and involuntary: ‘it needs sharply to be distinguished from [irrationality] … his detachment from reality some might describe as an absence of conscious action, but that fell short of involuntary, as distinct from irrational, action’.Footnote ³⁰ These are useful directions from the court but they do not give the full picture with regard to the meaning of involuntariness.

(e) Basic indicators

It is possible to extrapolate from these situations, however, some basic features which point towards the importance of choice and control. First, the notion of force used in Weaver is suggestive of an investigation into whether the defendant could resist or exercise a physical power. Secondly, there is also a sense of unavoidability within the situations cited above, ie the defendant could not have done otherwise. Thirdly, the examples point towards a lack of connection between the agent and their bodily actions, ie the defendant did not have a sense of causal agency but was part of a causal chain. Such an approach, which is implicit in Hobbart's report, suggests that involuntariness is illustrated by the defendant's loss of control, the defendant's loss of agency within the causal chain, and the defendant being rendered, in some way, an inanimate object. Fourthly, it could therefore be said that involuntariness is the absence of the capability to do otherwise, ie exercise a choice. Fifthly, the case law makes clear that involuntariness is not concerned with the cognitive abilities of a defendant but only their ability to control their actions. Specifically, it could be said that for courts involuntariness is about the absence of control which can be demonstrated factually. Hence it can be framed in the negative: the lack of factors or characteristics, such as the ability to direct the movements of the bodyFootnote ³¹ or the lack of a sense of control.Footnote ³² By taking this approach, a court can avoid the more challenging task of defining what is voluntary, opting merely to say what is necessary, ie control, but avoid saying what is sufficient, ie some additional criteria, such as purpose, motive, or independence.

(a) Choice as capacity

These indicators of involuntariness illustrate an expectation of human agents which is embedded within tort law. Such expectations can be found within commonplace expectations of persons and within theories of moral responsibility. It is no surprise then that in tort law you also find implicit expectations about the capabilities of agents to whom a duty is applied. Indeed, MacCormick says, there can be no normative order, including law, without a conception of persons. He goes on to say that, ‘fundamental to the existence of a person are capability to have interests and to suffer harm, and capability for rational and intentional action’.Footnote ³³ A key part of which, for MacCormick, is the ability to control and execute your actions. If viewed from this perspective, the plea of involuntariness speaks of the understanding that for an agent to qualify as an appropriate legal subject from whom a duty is expected, they need to have the capacity to choose.Footnote ³⁴ In conjunction with other conditions, such an expectation is often referred to as being part of a choice theory of responsibility.

Although it is common to find consideration of this notion within debates about criminal responsibility, when it is considered more closely it is evident that most theories of fundamental legal responsibility generally agree that the ability to choose is a necessary or common element of legal responsibility.Footnote ³⁵ It is useful to clarify its meaning. On this, Hart's description is instructive: Footnote ³⁶

It is immediately familiar: to be responsible we must be able to choose what we do. Yet if this notion of capability is to shed light on how the involuntariness plea operates within tort law, it needs to be appropriately situated within the structure of tort law. If not, it can be discounted too quickly as irrelevant given its otherwise close connection to morality and notions of punishment. First, however, it is essential to elucidate what is meant by choice and how it relates to control. In doing so, it soon becomes clear that such a notion is not necessarily tied to specific theories of punishment or criminal law but to conditions of human agency or personhood.

Helpfully, Moore has built upon Hart's definition. He has said that responsibility under the choice theory only attaches when the ability to choose was made impossible by conditions which are beyond the control of the agent. Or in an even weaker sense, there is a lack of choice when the ability to choose was made ‘very difficult’ by factors beyond the control of the agent. The first formulation by Moore chimes with the approach generally taken by courts when dealing with questions of involuntariness.Footnote ³⁷ For instance, it can be found in the dictum of Lord Justice Vos in Dunnage who said: ‘only defendants whose attack or medical incapacity has the effect of entirely eliminating any fault or responsibility for the injury can be excused’.Footnote ³⁸ The emphasis on ‘entirely eliminating’ speaks to the need for the loss of control to be complete; that is, Moore's stronger iteration: the ability to choose is made impossible. Further, in explaining fault, Lord Justice Vos invoked Lord Scott's description of an action taken by someone suffering from severe depression. Lord Scott concluded that such action was still liable to attract fault, saying: ‘But he was not an automaton. He remained an autonomous individual who retained the power of choice’.Footnote ³⁹ This underscores that for Vos and Scott involuntariness must be complete and, importantly, it relates to the ability to choose.

The challenge, Moore says, is to then determine the circumstances under which it is considered impossible or very difficult to exercise choice given factors that are beyond the agent's control. First, as already noted, in Dunnage the expectation was that the loss of control must be absolute and not partial. Secondly, it appears that courts are willing to accept elimination by causal factors, such as being forced, or agent-specific factors, such as heart attacks and blackouts. Lady Justice Arden said that Vince was negligent because: ‘he was able to choose to bring the petrol and lighter into the claimant's flat’.Footnote ⁴⁰ Further, Lady Justice RaffertyFootnote ⁴¹ quoted Neill J with approval when he said ‘automatism involves a complete loss of consciousness … The driver will … escape liability if his actions were wholly beyond his control’.Footnote ⁴² Later she said that the actions of Vince immediately prior to striking the match embodied the ‘choice he made … He made at that stage at least once decision, that is not to use it immediately. That evidence suggests that he had at that stage control which later he lost’.Footnote ⁴³ Therefore at least it can be said that here the question is about the ability to choose otherwise being made impossible. It might be argued, however, that this analysis wrongly elevates choice to a level which is beyond its importance; that is, tort law does not value choice in this way.

(b) Valuing choice and fault

Some might say that choice is rightly apposite for criminal law but that it does not fit with the underlying aims and goals of tort law. It could be supposed that in putting forward this understanding Hart, Moore and others are concerned with punishment, culpability and explaining why criminal law values choice as a condition of responsibility. Some go further. Cane, for example, argues that although many moral theories are based on a culpability model of responsibility, legal responsibility is not necessarily based on fault, culpability, or a singular focus on the agent.Footnote ⁴⁴ Cane argues that descriptively speaking some torts are not even based upon responsibility for past conduct, intention, or fault. In making this claim, Cane does not wish to say that law's structure of responsibility is necessarily different from morality, but he does wish to challenge the idea that the conduct of the agent is the primary driver of responsibility in tort. Hence, Cane objects to characterisations of responsibility which are focused upon the ‘quality of will’ rather than the overarching value of our obligations to others.

Although Cane does not examine the involuntariness plea specifically, on his analysis the existence of this doctrine could be explained by arguing that responsibility does not attach because the agent could not be said to be the cause of the harm. It is not because the agent lacked control but because, put simply, they were not the cause. On this approach, tort is only concerned with ensuring that those who cause harm will honour their obligation of reparation. Indeed, tort liability is strict and concerned ultimately with conduct not choice. Criminal law therefore can be discounted as it has different purposes and considerations.

Yet the description of tort and its structure of responsibility need not be undermined by adopting a choice theory of responsibility to describe the conditions expected of an agent to whom we apply the law of tort. You can hold that tort law is informed by a balancing between the needs of the claimant and defendant but equally that the law expects some basic characteristics, capacities, and abilities. Or to put it otherwise, it has implied pre-conditions before someone can qualify as a legal subject to whom active capacity is applied. Therefore, being concerned about the ability to choose does not contradict the need for a legal obligation to be concerned with the outcome or a balance between defendant and claimant. In presenting matters this way, the suggestion is that tort law does not ascribe responsibility to someone incapable of even achieving the basic functions of legal personality, ie control and the ability to choose. In some circumstances, the duty itself can be insensitive to the fact that the agent was never going to be able to satisfy what was required of them, such as negligence. Such an outcome can be justified on various grounds, such as moral luck, distribution of risk, or that the duty of care is a legal mechanism for sharing the costs of living within a society. Characterising negligence in this manner has led some to conclude that tort is essentially a strict liability regime; meaning that it is unconcerned with the abilities of the agent only if they were, in fact, the cause.

However, this debate about strict liability and how to justify such a position has detained the attention of most scholars of tort to the neglect of an importantly different question: what circumstances of incapability are relevant and on what basis? Or the more fundamental question of what conditions does the law of tort use to determine whether liability should be imposed in the first place. Once this question is considered it becomes clear that no matter whether you characterise negligence as strict or not there remain invariably circumstances where the capacity of the agent is relevant. Indeed, upon deeper reflection it emerges that this is not a question about the nature of the duty but rather underlying assumptions about to whom tort law applies. For example, Gardner argued that rationality was a foundational condition which was expected by law and morality, and it was something that was analytically necessary for legal responsibility to operate.Footnote ⁴⁵ Hart, however, merely noted that expectations of capacity were frequently found within different regimes of legal responsibility even if they were not necessarily required to satisfy his definition of legal responsibility.Footnote ⁴⁶

Whatever the case, the interpretation offered here argues that the case law gestures towards a deeper conception of capacity that is embedded within several tort law regimes throughout common law jurisdictions. If you are to follow MacCormick or Gardner rather than Hart, then you may even conclude that it is essential for tort law regardless of the jurisdiction in question. Regardless of whether this can be demonstrated or not, it can be said that an instance of involuntariness, as discussed in the cases above, relates to an investigation of the conditions that an agent needs to satisfy before they can be considered as a pertinent object upon which to place an obligation. On that basis the choice theory of responsibility speaks to one of the fundamental criteria we have of the agent to whom we apply a duty: they must have the ability to choose between alternatives. It is not a question of whether tort values choice over duty. What is being established here, and what is important to the involuntariness plea, is the criteria for qualification as a legally liable agent to whom tort applies. In tort law, liability assumes that a degree of free action is possible for the agent to whom a duty applies.

(c) Choice theory as a guide

If a choice theory of responsibility and its related literature is relevant to tort law's recognition of involuntariness,Footnote ⁴⁷ then some important conditions must be applied if it is to operate effectively.Footnote ⁴⁸ For example, Hart utilised a distinction between the capacity to do otherwise and the opportunity to do otherwise. Such a distinction captures the situations described in Weaver, where someone is physically forced to do something, ie they did not have the opportunity to do otherwise. Or the lack of capacity may explain Buckley and Toronto Transportation Commission v Smith Transport Ltd,Footnote ⁴⁹ where a truck driver drove under the delusion that his employer was controlling the movement of the truck by remote control. In such an instance, the defendant did not think they had the ability to control their actions. For involuntariness, it does not matter if both occur or just one or if they are internal or external: ‘factors that impair choice, whether internal, such as brain damage, or external, such as a threat, seem capable of acting either on their own or in combination’.Footnote ⁵⁰ So the lack of choice can materialise either externally or internally.

Secondly, it is important that the lack of opportunity to do otherwise has not arisen due to the fault of the defendant. Such an approach chimes with cases such as Mansfield where the notion of prior fault precluded any attempt by the defendant to argue that they were incapacitated at the relevant time. Of course, the defendant in Mansfield may have lacked at that moment the opportunity to do otherwise because they were in the cab of a lorry experiencing a hypoglycaemic episode, yet the question was whether it was the defendant's fault that there was no opportunity to do otherwise.

Thirdly, there is the need to disambiguate between that the existence of conditions of apparent negation of choice and the reality of it. For example, in Dunnage the Court of Appeal underscored that although the defendant was suffering from an episode of florid delusion, he had not loss control of his actions. As Lady Justice Arden said, ‘I do not consider that Vince's capacity was removed. He was able to choose to bring the petrol and lighter into the claimant's flat’.Footnote ⁵¹ So the circumstances may suggest that he did not have control, but he did, according to the court. Despite his condition, he had a choice. It is therefore crucial to determine the effects of the condition at the relevant time, not overall.

(d) Beyond dualistic approaches

The last point to stress is that if a choice theory informs the involuntariness plea, then questions will ultimately be informed by factual enquiries into what the defendant could or could not do at the material time. Therefore, it is important to emphasise that medical science generally does not view human action in a dualist manner, ie that action is divided into a mental and physical competence. This means that a court, when considering involuntary actions, should not be confined to physical movement alone. For although scientific study of human volition may adopt varied notions of control,Footnote ⁵² it is common for neuroscientists to make a distinction between self-directed and sensory-initiated actions to identify those which are voluntary from involuntary. Or you might describe this as contrasting the subjective experience of control and immediate responsive reflex actions. They do not speak of internal and external control. It is more sophisticated, which must be understood when dealing with pleas of involuntariness and soliciting evidence from expert witnesses. To take this approach is, of course, to take seriously the common perception that some actions are voluntary rather than predetermined, which for lawyers is unproblematic. It starts from an analysis of our conscious experience of volition. Yet it should be noted that to take seriously the perception of control, volition or self-directed movement does not entail a commitment to its empirical causality, only that it is a key aspect of human experience that can be explained, in part, through exploration of the brain's neurological activity.

Hence, some studies try to identify the neural and physiological mechanisms involved in movements reported to be voluntary without necessarily seeking out the neural mechanisms of voluntary actions, only the correlation between the sense of volition and the action reported to be self-initiated. On this basis, it is possible from the perspective of neuroanatomy and neurophysiology to identify the mechanisms of self-initiated action and those which are mere reflex responses:Footnote ⁵³

Following this definition, it is possible to identify the processes of the brain that are related to psychogenic movement disorders, alien hand syndrome and other conditions which affect control, including schizophrenia. It is also noticeable that this definition rests on the notion that a sense of control is key to understanding volition scientifically. It is therefore not possible for a court to side-step difficult questions about human action by using the concept of medically recognised condition, because as far as contemporary neuroscience is concerned, the experience of control remains one of the central conditions of any definition of volition. Yet what we do gain from examining the approach of contemporary neuroscience is some assurance that the concept of volition and involuntariness is not necessarily without a concrete scientific basis. Of course, it should be noted that neuroscience does not generally approach this question in a black or white manner. It is assessed from the point of view of a scale of control or sense of volition. Thus, if the use of the involuntariness plea is to be informed by a choice theory of responsibility, which appears to be the most viable interpretation, then it is key that future decisions of a court are informed by a more nuanced understanding of human conduct, which includes the insight that it is a sense of control which is central to how mainstream medical science analyses human action.

(a) Disentangling intentional and negligent harm

It is vital to understand how involuntariness and intentional harms relate to each other. For in a case where a mental element is key to establishing a claim, the presence of a mental disorder may be important – because it affects not the ability to control actions but the ability to form the required intention. These are not cases which touch upon the capacity of the agent but implicate the lack of a key competent of the tort, ie a specific motive or intention.Footnote ⁵⁴ So, although a simple plea of insanity may be technically irrelevant in a negligence action, that does not mean the nature of the mental condition is not pertinent to claims formulated on the grounds of, for example, assault, trespass, or conversion.Footnote ⁵⁵

The significance of mental disorders here, however, is different from instances of involuntariness. Unlike a negligently caused harm, an intentional tort that requires the demonstration of a mental element directly engages with the mental state of the defendant at the time the potential tort occurred either by presumptions, ascriptions, or direct evidence.Footnote ⁵⁶ Here the question is not about the fundamental lack of capacity which renders the agent completely beyond the effective operation of an obligation to avoid the prohibition imposed upon them. The concern is whether you have the sufficient intention to carry out an otherwise prohibited action, which you have an obligation not to breach intentionally. The question is when you have a mental disorder which negates the proper formation of an intention, has the obligation to avoid such action been breached? When considering this question, the court is fixed at the prohibition stage and not with primary questions of whether the obligation to avoid specified actions can be imposed due to a basic lack of capacity, ie control. The central point here is that questions of intention do not necessarily relate to questions of primary responsibility – the defendant's ability to choose to do otherwise – but to the grounds of liability, ie the elements of the tort explicitly require an intention.

First, if a defendant can establish that the condition negated the presence of malice or intent, then the claim will fail.Footnote ⁵⁷ This is not about capacity but about the criteria of the duty. Hence, such cases inevitably invoke a discussion about the scientific and clinical understanding of the defendant's cognitive abilities, including the ability to understand the nature of the act in question. The assessment is not about whether the duty can be imposed on the agent in the first place. It is about whether the mental disorder or capacity negates a key condition of liability. These issues about the cognitive abilities can be just as troublesome during evidence as questions of involuntariness. A failure to keep them separate from the question of involuntariness can create an additional complexity when trying to ascertain the nature of the claim, how it should be resolved and how to interpret the evidence. Involuntariness is about foundational capacity to choose. Yet, in general terms, if a defendant can demonstrate that their mental disorder impacted upon their ability to form an intention or motive, then this will defeat the claimant's assertion that an intentional tort has taken place. On the other hand, with negligence, there is, of course, no need to demonstrate an intention or motive but only that objectively speaking the conduct amounted to negligence. On that basis, it can appear that considerations of mental disorders are not strictly relevant nor part of the grounds of liability a claimant needs to establish.

However, what must be emphasised is that questions of involuntariness are distinct from both issues (ie whether the act was negligent or intentional); they are more fundamental. To make this clear, it is necessary to explain what is meant by intention.Footnote ⁵⁸ Generally it can be said that there are two very broad senses in which intention is used within tort law: one sense of intention concerns the conduct of the agent, whereas the other relates to the consequences of that conduct. In both law and philosophy conduct is concerned with the notion of choice – it expresses a choice theory – while consequences coalesce around ideas of purpose, objective, or goal. On this understanding, there is no need to enquire as to what was intended or whether the defendant knew their actions would cause or bring about a certain consequence. Rather, what the court should ask is whether the defendant retained the ability to choose what to do and had the ability to control their actions. Courts ask this because they are, in effect, enquiring into the agent's connection to their conduct. The distinction between a conduct sense of intention and a consequence notion of intention demonstrates that involuntariness can – and should – be held apart from questions of malice, motive, or the purpose-orientated sense of intention. It also illustrates that a duty can incorporate a consequence-sense of intention whereas a capacity assessment of an agent entails a different type of assessment of intention, ie conduct intention.

(b) Not simply a breach of duty

The foregoing brings us to consider how involuntariness is about the capabilities of the agent rather than a condition of the duty. There is a common misconception that to invoke involuntariness is to challenge the otherwise orthodox formulation of a duty as being objective. It does not. Rightly, the Court of Appeal in Dunnage stressed at several points that the standard of care is an objective test, which does not account for any subjective factor, idiosyncrasy or abnormality which would affect the defendant's ability to meet the expectations of a reasonable person.Footnote ⁵⁹ Although this point is well-established within the case law from Glasgow Corporation v Muir Footnote ⁶⁰ to Nettleship v Weston,Footnote ⁶¹ and accepted by HHJ Saggerson at the first instance, it was unhelpfully reintroduced by counsel for the claimant in the Court of Appeal. For the claimant, Mr Spearman QC sought to push the existing law further by arguing that ‘the objective standard applies to all the ordinary person's actions, voluntary or involuntary…’.Footnote ⁶² In a bid to reformulate the conventional understanding of how involuntariness operates within tort, the claimant argued ‘that where injury is by an individual disabled from controlling his actions, liability depends on whether the ordinary reasonable man would be so disabled, not on whether acts are involuntary, nor on mental capacity’.Footnote ⁶³ Although it was argued by the claimant that this was the orthodox approach, it is hard to find such an articulation within the case law and it was rejected by the Court of Appeal. It does, however, merit some consideration to explain that structurally it is the wrong approach to take to such questions.Footnote ⁶⁴

The claimant's approach should have been to stress that the lack of control went to the heart of basic responsibility, meaning the duty of care could not be applied to the agent. Yet by taking the line of argument outlined above and supposing that the duty of care could accommodate a subjective condition for those suffering a mental health disorder, the claimant sought to evade the elemental question of capacity by collapsing it into a question of whether there had been a breach of the duty of care (or more generally, prima facie liability). Thus, for example, a person who suffers a stroke, losing consciousness, slumping at the wheel and causing injury to others does not fall below the standard expected of a reasonable person – the standard of care would not expect such a person to withstand the effects of the stroke and remain conscious. Likewise, people are expected to resist, withstand and control their actions if they suffer from insane delusions according to the standard of care. It does not take into consideration that this specific defendant did not have the capability to withstand such delusions due to an illness; it asks what a reasonable person would do in such circumstances. On this argument, it was contended for the claimants that the question was simply about whether the actions of the defendant amounted to a breach of the duty of care. As much as the law did not consider personal characteristics, it should not consider physical or mental disabilities either. The Court of Appeal, however, was not prepared to entertain this kind of extension of the meaning of the standard of care. As Lord Justice Vos remarked, the claimant's argument ‘somewhat glosses over the real question which required the examination of what is really required before a defendant can be said to have acted in breach of a duty of care undoubtedly owed’.Footnote ⁶⁵ Or, to put it more systematically, there is – analytically – a preceding question of whether a duty of care can be applied to the defendant in the first place.Footnote ⁶⁶

Again, like questions of intention discussed above, this point re-affirms that involuntariness is a different sort of legal question from the conditions of a duty. For one, when it is accepted by the court, the defendant will escape civil liability entirely.Footnote ⁶⁷ It does not alter the expected standard of care or the conditions of liability but rather demonstrates that the facts of the case do not engage questions of prima facie liability. Or, to put it another way, although often overlooked, or taken for granted in tort actions, there is an assumption or implied foundational requirement that liability will not be applied to someone who does not have the ability or capacity to satisfy the requirements of their duty.Footnote ⁶⁸ So, the focus is not upon what the defendant should have done or what the reasonable person would have done, but on the abilities, condition and situation of the defendant. It functions at a more fundamental level of capacity than abnormalities, subjectivities or personal characteristics, posing basic questions about whether the defendant satisfies the conditions of personhood or fundamental legal responsibility. On this account involuntariness is relevant to what Hart called ‘capacity responsibility’, which asserts ‘that a person has certain normal capacities’.Footnote ⁶⁹ Operating at a prior stage to questions of reasonableness, it situates the legal debate around questions of capacity, control and ability.Footnote ⁷⁰ If the standard of care covers different questions from that of involuntariness, this has implications for how such cases should be pleaded, ie on whom does the onus of proof lie, what evidence is required, ie whether there is an assumption of capacity which should be rebutted, and how such cases should be structured procedurally, ie whether it is a defence and what kind of defence.Footnote ⁷¹ It also raises the question of how involuntariness should it be structurally and procedurally understood: is it a denial of liability completely or does it operate like an excuse?

(c) Denial or justification

To answer this question, Goudkamp's analysis of the use of insanity pleas in tort actions is instructive and insightful but also, for present purposes, somewhat misleading in the way it formulates individual torts. Using the concepts of absent and affirmative defences,Footnote ⁷² or in later work, denials and defences,Footnote ⁷³ his scholarship helpfully establishes that it is vital to consider how mental disorders fit within the arrangement of a legal argument. Persuasively contending that a neglect of these distinctions can distort legal analysis, he suggests that drawing upon criminal law's extensive consideration of these issues can provide insights about the organisation of claims. Through this, Goudkamp has extensively demonstrated that some responses to a claim in tort are forms of denial whereas others are types of defences. A denial is what he describes as an absent element response in which one or more of the conditions of liability is simply denied. Such a defence goes to the very cause of the action, disputing the presence of a necessary element of liability – such as that the act was reasonable. An affirmative defence, however, does not refute the existence of conditions of liability but contends that there is a justification or some sort of exemption which renders the otherwise unlawful actions lawful. As he explains, such a defendant ‘seeks to avoid liability not by denying the claimant's allegations but by going around them’.Footnote ⁷⁴ Although this analytical approach has its merits, making evident the importance of structure, it also has limitations which, if followed rigidly, unnecessarily constrain how involuntariness would be applied by a court. Whether it is in how he characterises a denial or, how torts are formulated, there is something troubling about denying a general capacity requirement within an account of liability.

Applying his framework to circumstances where mental disorders or physical incapacity may be considered relevant, Goudkamp argues that an absent element defence will generally be indirectly applicable if the tort specifies an act element. That is, one of the major premises of his argument is that each individual tort can be broken down into specific conditions, which if not satisfied are open to a denial on behalf of the defendant. On this basis, he asserts that that in order for involuntariness to be relevant as a denial, it would need to be explicit within the definition or express conditions of the tort that a certain act is required.Footnote ⁷⁵ Yet it would be unusual for such an element to be overtly stipulated in a claim, ie for a claimant to assert, before anything else, that the defendant at the time of the injury had sufficient capacity or was acting voluntarily. Nor do you normally find it expressly mentioned within juristic writing about specific torts. He argues that to expect a claimant to establish this is impractical and, at least prima facie, it is not customary when discussing a tort to start with a primary condition of voluntariness. Hence, he argues that voluntariness is only relevant if the tort specifies this within the established definition, such as intention. So, it is evident that there will be some cases where an intentional tort will be met with evidence amounting to a denial or absent element defence because, for example, it is not possible to say that the defendant intended to deceive the claimant, or to harm them, or their business interests. Furthermore, his examination of the case law leads him to conclude that overall, there are very few examples where a common law court has accepted pleas of insanity or involuntariness as affirmative defences or some sort of justification for the otherwise wrongful act.

However, this approach truncates matters in a similar manner to how the claimant in Dunnage attempted to collapse questions of involuntariness into the standard of care. That being so, Goudkamp does not identify questions of involuntariness, and whether there was an act to begin with, as necessarily prior to questions about liability. Rather he envisages the conditions of each tort to be express, including the requirement of an act, without acknowledging the background yet fundamental assumptions, that any express conditions of tort liability are made against; that is, the understanding that the defendant had at the very least a minimal level of control, capacity or competence, ie has acted.Footnote ⁷⁶ Certainly, Goudkamp believes that insanity should be a defence, but his position is based on the argument that it should be recognised as an affirmative defence based on both public policy and moral reasons. Yet what this fails to account for is that under the present law of most common law countries insanity or pleas related to voluntariness are denied generally because ‘insanity’ has no nominate status within the legal system. This does not mean to say that the effects or circumstances of a mental disorder or incapacity event are not relevant to the court nor that there is a foundational expectation that liability only arises when there has in fact been some sort of act. You just cannot try to deny liability with a bare plea of medically recognised mental illness, yet you can try to show why that undermines the ascription of responsibility.

Stevens underscores this point. Speaking of ‘extreme examples of mental incapacity’ he notes, ‘these are not examples of incapacity operating as an excuse. Rather the incapacity is so extreme as to prevent the attribution of a physical action to the person moving’.Footnote ⁷⁷ This is an argument which Gardner makes strongly when he argues that where incapacity goes ‘very deep’ then defences, whether they be excuses or exemptions, are an inappropriate way to address this question or liability because you are denying responsibility tout court.Footnote ⁷⁸ MacCormick also pursues this point when he says, ‘any conception of law that envisages it as in any way guiding or regulating the conduct of conscious and rational agents … has to have at least an implicit concept of active capacity’.Footnote ⁷⁹ The defences and denial structure as posed here does not necessarily accommodate this insight. Arguably, to use a denial is itself to acknowledge that you are responsible and capable of explaining your actions rationally or at least in a manner that is explicable. Denials therefore do not fully capture the nature of involuntariness. Of course, this opens more questions, which cannot be fully answered here but for now the point to stress is that involuntariness operates at a deeper level than how Goudkamp defines defences and certainly at a prior stage to consideration of the prima facie conditions of liability. It may still be a defence, but it should not be collapsed into prima facie questions of liability as it is more fundamental.

(d) Conflating fault and action

As a legal analysis, the argument put forward by the claimant in Dunnage, supposing that everything could be addressed at the duty stage of the debate, is arguably symptomatic of another confusion that often occurs when the question of capacity and negligence arise, something which both Lord Justice Vos and Lady Justice Arden elucidated in their judgments. That is, questions related to involuntariness are not, directly, related to questions about the standard of care or conditions of prima facie liability. To be clear, one issue is whether the standard of care or normative expectations should be altered for someone suffering from a mental health disorder.Footnote ⁸⁰ Rarely, however, as noted, does a court or commentator depart from the principle that the standard of care is objective and should be applied without variation due to subjective circumstances, abilities, or skill.Footnote ⁸¹ This is not the same issue, however, as that which arises when the court seeks to determine whether the actions of the defendant were voluntary or not, which relates to conduct and control. In Winfield & Jolowicz on Tort, the authors underline this point, highlighting that often cases dealing with mental disorders or instances of sudden incapacitation are, in fact, about ‘prior fault’ rather than whether liability attaches during a period of incapacity or mental disorder. Footnote ⁸² So, this doctrine, as they explain it, affixes liability at an earlier point within the chain of causation but certainly not at the point of incapacitation or the loss of control.Footnote ⁸³

A good example of a prior fault case is Waugh v James K Allan Ltd. Footnote ⁸⁴ Although modern writers often cite Waugh to support the proposition that a voluntary action is required to determine liability, the case itself says very little about voluntariness.Footnote ⁸⁵ Indeed a great deal of the court's consideration is about the decision of the driver to step into the cab of a lorry whilst feeling unwell.Footnote ⁸⁶ As the writers of Winfield & Jolowicz on Tort note, this case, like others, is about the standard of care and not the effect of insanity or incapacitation itself. As the authors suggest, to plead your case in this manner helps avoid more difficult questions about whether, at the time of the injury, you can hold the defendant liable. Further they note that shaping your claim as relating to prior fault addresses concerns that the rejection of an insanity defence is in some way unjust to those suffering from a mental disorder or incapacitation. Yet the question of involuntariness is exactly that type of difficult question, which in some cases – such as in Dunnage – cannot be side-stepped.

In Fiala v MacDonald, the Court of Appeal in Alberta grasped this point. Although it did not adopt the language of involuntariness, it did examine both the policy and structural issues relating to incapacity and tort liability. It made the distinction between situations where ‘an inability to meet the reasonable person test is no defence’ to an action of negligence and ‘exceptions granted to children, and the physically disabled…’. It concluded, therefore, that it is ‘not unreasonable’ in some situations to recognise in some very limited situations that mental illness may render the defendant incapable and beyond the reach of tort liability.Footnote ⁸⁷ Indeed, it noted that the literature and case law reveals that there has been judicial acceptance that some circumstances of mental illness warrant an exception to the general rejection of pleas of insanity within tort actions. In coming to this decision, it said ‘To find negligence, the act causing damage must have been voluntary and the defendant must have possessed the capacity to commit the tort’.Footnote ⁸⁸ Importantly, it remarked that to acknowledge this does not undermine the fundamental tenets of tort, but rather reinforce them. It does not ‘erode the objective reasonable person standard’. In fact, it ‘will preserve the notion that a defendant must have acted voluntarily and must have had the capacity to be liable’. To recognise this does not mean that fault is no longer necessary; only that the notion of choice is important to tort law's interpretation of what amounts to ‘fault’.Footnote ⁸⁹

Involuntariness is therefore a different sort of legal question from what the standard of care should be.Footnote ⁹⁰ When it is applied the defendant will escape civil liability entirely,Footnote ⁹¹ as if operating like a denial rather than a justification. Or it is framed in terms of consideration as to whether a duty of care is owed by the defendant in the first place, because the defendant was suffering from a mental disorder at the time of the claimant's injury. Additionally, the focus is not upon what the defendant should have done or what the reasonable person would have done; it addresses the primary question of the abilities and situation of the defendant. It appears to function at a more fundamental level, posing the basic question as to whether the standard of care should be applied in the first place to an individual in the circumstances of the defendant. Operating at a prior stage to questions of reasonableness, it situates the legal debate around questions of capability, control, opportunity, and choice. If the standard of care covers different questions from that of voluntariness, this suggests that each question needs to be addressed separately to ensure consistency and coherence.