Jump to ContentJump to Main Navigation
Manifest MadnessMental Incapacity in the Criminal Law$

Arlie Loughnan

Print publication date: 2012

Print ISBN-13: 9780199698592

Published to Oxford Scholarship Online: May 2012

DOI: 10.1093/acprof:oso/9780199698592.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (oxford.universitypressscholarship.com). (c) Copyright Oxford University Press, 2021. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use. date: 18 January 2022

Incapacity and Disability: the Exculpatory Doctrines of Insanity and Automatism

Incapacity and Disability: the Exculpatory Doctrines of Insanity and Automatism

(p.103) 5 Incapacity and Disability: the Exculpatory Doctrines of Insanity and Automatism
Manifest Madness

Arlie Loughnan

Oxford University Press

Abstract and Keywords

This chapter focuses on the substantive law of insanity and automatism. It suggests that, when a loose, broad, and partially moralized notion of incapacity — defined largely by extra-legal norms — pertained as a basis for exculpation, claims falling across the bounds of insanity and automatism were accommodated within an informal insanity doctrine and under a flexible criminal process. However, gradually, as mental incapacity came to be the subject of expert medical knowledge — a change that took place as much beyond as within criminal law — this broad notion of incapacity ossified into a narrower notion of disability, ushering in a more circumscribed approach to insanity. It was in this context that a discrete automatism doctrine appeared in the second half of the twentieth century.

Keywords:   insanity, automatism, incapacity, exculpation, criminal law, disability

This chapter and Chapter 6 cover a part of criminal law that is now traversed by the two exculpatory mental incapacity doctrines of insanity and automatism (also known as insane automatism and non‐insane automatism respectively). As is well known, the law on insanity is governed by the M'Naghten Rules, which were drafted in 1843.1 A discrete doctrine of automatism appeared only in the 1950s.2 Until this point, insanity and what would come to be called automatism coexisted in a way that was not sharply delineated, and, in this chapter (and the following), I examine insanity and automatism side by side, an approach which reveals the significance of legal concern with dangerousness. It was this concern that forced a cleavage between insanity and automatism, which eventually hardened into two distinct doctrines. Concern with dangerousness continues to inform the relationship between insanity and automatism, and appreciating its role here assists in understanding this part of the mental incapacity terrain.

In broad brush strokes, the first of the two main arguments advanced in this chapter is that when a loose, broad, and partially moralized notion of incapacity—defined largely by extra‐legal norms—pertained as a foundation for exculpation, claims now falling within the parameters of both insanity and automatism were accommodated within an informal insanity doctrine and under a flexible criminal process. This obviated the need for a specific exculpatory doctrine of automatism. Even once a formal doctrine of insanity developed in the first half of the nineteenth century, some claims to exculpation that now fall within the bounds of automatism were accommodated by the breadth of insanity. Gradually, however, as mental incapacity came to be the subject of expert medical knowledge—a change that took place as much beyond as within criminal law—this broad notion of incapacity ossified into a narrower notion of disability, fostering a more circumscribed approach to insanity. It was in this context that a discrete automatism doctrine (p.104) appeared in the second half of the twentieth century. By way of contrast with other mental incapacity doctrines, but reflecting the persistence of a broad, moralized notion of incapacity in this part of the mental incapacity terrain, automatism is delimited via a tripartite construction which tracks the lines of non‐culpability, catching a miscellaneous collection of cases in which individuals share little more than an absence of blameworthiness.

The second main argument of this chapter relates to the meanings given to those individuals seeking to rely on exculpatory insanity in and through legal processes. In brief, individuals pleading insanity have been constructed as dangerous, a construction which has been a driving force for most of the developments regarding the insanity doctrine. Reflecting the rise of expert psychiatric and psychological knowledge of mental incapacity, the notion of dangerousness has become a less moralized and more medicalized one in the recent history of insanity—interpolating with the more technical notion of risk in recent decades. The construction of insane defendants as dangerous forged an intimate and durable connection between the law of insanity and the issue of disposal. As this reference to disposal suggests, the close corollary of the argument made here about the meanings given to exculpatory ‘madness’ in criminal law is the evidentiary and procedural aspects of the way such ‘madness’ becomes known and is proved for evaluation and adjudication purposes. The evidentiary and procedural dimensions of exculpatory ‘madness’ form the focus of Chapter 6.

Of Unsound Minds and Wild Beasts: Insanity before M'Naghten

In his seminal work on mental incapacity, Nigel Walker traces the earliest recorded acquittal on the basis of insanity (‘the felon was of unsound mind’) to 1505.3 At some point during the early modern period, for reasons that are unclear, it became regular practice to acquit the insane defendant rather than leave him or her to be pardoned by the King.4 Although the absence of sources renders the picture of exculpatory insanity at this point somewhat unclear, it is generally accepted that an informal practice of excusing an insane defendant from trial long preceded the appearance of a formal insanity doctrine.5 At the time of the earliest recorded acquittal based on insanity, there seems to have been no substantial elaboration of (p.105) the meaning of a phrase like ‘unsound mind’ or any particular procedural structure for adjudicating claims to insanity. Although insanity was a somewhat disaggregated notion in the early modern era, with, for example, a recognized distinction between intellectual disability and insanity,6 overall, it was a broad and loose concept, and, under these conditions, some states—including those that might now form the basis of an automatism plea, such as physical injury to the head prior to an offence of violence—fell within the bounds of exculpatory insanity.

The capacity of informal criminal processes to accommodate insane individuals remained large, even as criminal process underwent significant changes associated with the rise of adversarial criminal procedure in the period over the eighteenth century. As mentioned in the previous chapter, together with changes in sentencing practices, developments in criminal procedure profoundly affected legal practices. Over time, the accelerating involvement of lawyers meant that judges came to perform the more limited role of ‘umpire and trial manager’, while juries came to play an ‘increasingly constructive’ role.7 If an individual was convicted, a range of sentencing options meant that a capital offence did not necessarily mean execution: some defendants raising informal insanity pleas were given partial verdicts, a ‘largely jury administered scheme of mitigation’ that enabled jurors to temper the harshness of the law.8 Some features of criminal process remained largely unchanged over the 1700s. Prosecutions continued to be brought by victims and what would now be called pre‐trial process remained ‘chancy’ and ‘largely informal’, in Keith Smith's words, and the way in which insane defendants were dealt with varied widely.9 At the start of the century, as at the end, if an insane individual was acquitted, no particular disposal was mandated, and what happened to the defendant varied according to his or her personal circumstances.10

(p.106) It was in this dynamic procedural and punishment context that the first famous insanity case appeared—that of Edward Arnold in 1724.11 Arnold was charged with maliciously shooting at a prominent local member of the aristocracy, Lord Onslow, under the recently enacted Black Act.12 Arnold pleaded that he did not know what he was doing and did not intend any harm. Evidence adduced at trial by Arnold's family and the local community indicated that Arnold was given to ‘irrational antics and minor acts of violence and damage’, but evidence led by the prosecution about the preparation of the offence suggested that Arnold could ‘form a steady and resolute design’.13 In his directions to the jury, Justice Tracy stated:

When a man is guilty of a great offence, it must be very plain and clear before a man is allowed such an exemption…it must be a man that is totally deprived of his understanding and memory, and doth not know what he is doing, no more than an infant, than a brute, or a wild beast [in order to avoid punishment].14

Although sometimes taken to indicate the strictness of informal insanity at this time, what has come to be known as the ‘wild beast’ insanity test was more of an informal standard than a ‘precise formula’ for assessing lack of intent.15 For this reason, the ‘wild beast’ insanity test does not conclusively indicate that, by this point in time, ‘partial insanity’—the kind of insanity that affected an individual in some respects but not others—was fatal to any claim to exculpation or that lack of reason (as opposed to lack of control) was the preferred basis for exculpatory insanity.

Over and above the issue of the strictness of the informal insanity law at this point, the ‘wild beast’ test is significant in another way. Viewed in light of the informal criminal processes then prevailing, the ‘wild beast’ insanity test appears significant for what it suggests about the relevance of the formal qualities of exculpatory ‘madness’, and the ways in which it was proved, for legal purposes. Paralleling my reading of Matthew Hale's reference to the ‘absolute’ character of the ‘madness’ associated with unfitness to plea (which I discuss in Chapter 4), the (p.107) ‘wild beast’ insanity test is more of a requirement of form than of degree or extent. Given the norms of proof then prevailing, Justice Tracy's direction that insanity must be ‘very plain and clear’ is as significant as his direction on the deprivation of understanding and memory. The conduct of an insane defendant had a particular significance in the criminal courtroom: exculpation of a defendant claiming insanity was made not so much via a deduction of his or her internal mental processes from his or her behaviour but on the basis that that behaviour constituted a ‘mad’ condition.16 Further, Arnold's insanity was regarded as evident in his conduct beyond the act comprising the offence.17 I take up the issue of evidence and proof of insanity in the next chapter.

The ‘wild beast’ test was not the only formulation of insanity in this era. The Old Bailey Proceedings indicate that a myriad of references were made to describe states that fell within the bounds of insanity. The brief references to a defendant's ‘distracted Gestures’, ‘the Oddness of his Behaviour’, or ‘violent phrensies of mind’18 contained in the trial records of the OBPs were designed to tap into ordinary people's understanding of ‘madness’. An example of the role of ordinary people's testimony about incapacity, is provided by the trial of Benjamin Allen on a charge of the theft of books and paper in 1768. After stating that he knew ‘nothing at all of it’ and explaining that he was asked to carry a bundle by a gentleman, four of Allen's acquaintances spoke on his behalf, one of whom said he had looked ‘upon him to be insane’ and another said that Allen was subject to fits which ‘had disordered him in his head very much’.19 Allen was given a partial verdict and sentenced to be whipped. As these extracts suggest, and as Roy Porter argues, in this era, ‘madness was an extremely broad sociocultural category, with many manifestations and meanings’ and ‘meanings of madness multiplied within lay culture, describing individuals, acts and situations out of the ordinary’.20 As it was part of common knowledge, ordinary people without specialist knowledge were regarded as competent to detect and evaluate ‘madness’. I discuss the relevance of social meanings of ‘madness’ in Chapter 6.

Each of the expressions employed to capture insanity was partly prescriptive and partly descriptive of an abnormal mental state: at this juncture, exculpatory insanity had not yet undergone any sustained conceptual elaboration in criminal law. Individuals articulated claims to exculpation on the basis of insanity alongside a range of other claims. References to defendants ‘acting like a crazed Person, for a (p.108) Week of 10 Days before the Time when the Murther was committed’21 and evaluative statements to the effect that the defendant was a ‘Lunatick’,22 or ‘out of his senses’23 or a ‘Distracted Person’24 intermingled with references to character, age, and other factors. As Dana Rabin argues, like other ‘pleas of mental distress’, insanity was raised by individuals in order to persuade the judge and the jury that their crime was ‘committed without criminal intent’.25 During this period, the ‘thin doctrine of capacity as a condition for criminal responsibility’26 that was a feature of the exculpatory criminal trial (which I discuss in Chapter 4) was only gradually being replaced by a more robust subjective concept of criminal fault.27 In this context, references to potentially exculpatory mental states (such as ‘unable to tell good from evil’) were designed to challenge the authenticity of the manifest meaning of a defendant's acts as criminal.

Facilitated by the fact that, at this time, there was little ‘refinement’ of matters such as ‘levels of mental culpability and recognition of defences involving incapacity or the actor's freedom and choice of action’,28 claims of mental abnormality and volitional incapacity—now divided by the boundary between insanity and automatism—coexisted alongside each other, and, indeed, were bound up together in some cases. Given the breadth of ‘madness’ as a sociocultural category, it is likely that individuals such as those who could point to both external and internal causes of incapacitous conduct fell within the bounds of socially defined ‘madness’.29 Mindful of the need for caution in searching for prototypical automatism cases,30 those OBP trial records which refer to an external cause may be tentatively regarded as informal claims to exculpation on the basis of automatism. The record of William Walker's trial for the murder of his wife in 1784 provides a good example of what may be thought of as an informal automatism claim.31 Walker stabbed his wife with a knife. There was no evidence of discord between the defendant and his wife, and several witnesses testified that they did not think (p.109) Walker was ‘mad’. The judge seemed confounded by the facts, and commented that it was ‘one of the most extraordinary cases I ever met with’. By way of defence, William Walker simply stated ‘I am not sensible as I did kill my wife, and please you my Lord’. In his summing up, the judge stated that something ‘singular and extraordinary’ must have happened to Walker in the days before the killing to disturb him from his ‘orderly state’, effectively imputing an external cause to the defendant, and Walker was found not guilty on the basis of insanity.

The OBPs also contain records of cases in which an external cause operated together with an internal cause. A good example of this type of case is the trial of Richard Montgomery in 1727. In that case, an apothecary deposed that he had administered medicines to the defendant ‘to remove his Indisposition’ the day before the alleged offences (theft and robbery) took place.32 The defendant's ‘extravagant Expressions’, evidence that ‘Lunacy ran in the Blood of the Family’ and the fact that one of the offences took place in plain view of people coming from church, were possibly behind the jury verdict that the defendant was non compos mentis. As this verdict indicates, the informal insanity law was sufficiently capacious to capture both internal and external causes of incapacity.

The Cleaving Apart of Insanity and Automatism

Concern with dangerousness forced the cleaving apart of insanity and what would come to be called automatism. This cleaving apart was initially prompted by another famous insanity trial, that of James Hadfield for high treason in 1800. Hadfield had attempted to shoot King George III, believing that this act would ensure that he himself would be killed but the world would be saved.33 As special privileges accompanied treason trials, Hadfield was entitled to assistance from counsel in the preparation of evidence and the examination and cross‐examination of witnesses.34 His counsel, Thomas Erskine, argued that, rather than ‘total deprivation of memory and understanding’, ‘delusion was the inseparable companion of real insanity’.35 A doctor from Bethlem examined Hadfield and stated in court that ‘when any question is put to him which relates to the subject of his lunacy, he answers irrationally’.36 The Justices interrupted Erskine's defence and the Attorney‐General confirmed that he did not want to challenge the evidence. Hadfield was acquitted and, in accordance with the practice that had prevailed since the early modern era, the jury gave both their verdict and its factual basis: ‘[w]e find the prisoner Not Guilty; he being under the influence of insanity at the time the act was committed’.37

(p.110) The uncertainty attending Hadfield's disposal led Parliament to pass the Criminal Lunatics Act 1800, an Act which contained the first fissures of a crevice between insanity and automatism, and represented the first step in the formalization of insanity (and insanity on arraignment/unfitness to plead). Section 1 of this Act provided:

That in all cases…of any person charged with treason, murder, or felony, that such person was insane at the time of the commission of such offence, and such person shall be acquitted, the jury shall be required to find specially whether such person was insane at the time of the commission of such offence, and to declare whether such person was acquitted by them on account of such insanity; and if they shall find that such person was insane at the time of the committing such offence, the court before whom such trial shall be had, shall order such person to be kept in strict custody, in such place and in such manner as to the court shall seem fit, until his Majesty's pleasure shall be known.38

While the Act did not define insanity for legal purposes, it fundamentally altered the procedural context in which claims to exculpation on the basis of insanity were made. As a result of the 1800 Act, it was no longer open to the jury simply to acquit the insane defendant.39 The Act also brought the detention of insane defendants into the criminal law, providing the court with power to keep such individuals in custody.40

As the 1800 Act suggests, this first step in the cleaving apart of insanity and automatism—and along the path to the formalization of the law on insanity—was something of a backdoor development, as it concerned the disposal of individuals (p.111) acquitted on the basis of insanity, rather than the law itself. The link established in the Criminal Lunatics Act 1800 between a successful insanity plea, a particular trial verdict (the special verdict) and indefinite detention would prove to be an enduring feature of exculpatory insanity.41 Even after indefinite detention fell away, and other disposal options were made available to courts, the link between a successful insanity plea and the special verdict has remained, conjoining the issue of criminal non‐responsibility and disposal in a way that is unique in criminal law. The special verdict forms part of the distinctive procedural frame governing insanity claims, and I discuss this in the next chapter.

The concern with disposal that was evident in the passage of the Criminal Lunatics Act 1800 reflected a particular attitude to insane individuals charged with criminal offences: that they were dangerous. It was this construction of insane defendants which legitimated the indefinite detention that followed a successful insanity doctrine. This construction of insane defendants as dangerous has been remarkably durable: with the exception of the most recent developments relating to the disposal of insane defendants, it has been the driving force behind the insanity doctrine since 1800.42 In Sullivan, Lord Diplock stated that ‘the purpose of the legislation relating to the defence of insanity, ever since its origin in 1800, has been to protect society against recurrence of the dangerous conduct’.43 The construction of insane defendants as dangerous has concentrated the attention of judges and law reformers on issues relating to disposal, issues which have dominated discussions of the law of insanity. This concern with dangerousness has been rearticulated in recent decades as a concern with risk—a less overtly moralized and more technical concept—that I discuss below.

At the turn of the nineteenth century, with the contours of the fault element of criminal offences and its relationship to exculpatory insanity yet to be explicated, the insane defendant's conduct formed the basis on which he or she was constructed as dangerous. An insane accused's conduct—including the acts comprising the offence, but also the conduct surrounding the offence—had a thick significance, extending beyond that now commonly accorded to the actus reus of an (p.112) offence. As I discuss in relation to evidence and proof of insanity, and also as part of my ‘manifest madness’ analysis, evaluation of an individual's conduct was made not so much via deduction of his or her mental processes from his or her behaviour, but on the basis that the behaviour constituted the ‘mad’ condition.44 This emphasis on the manifest meaning of the individual's conduct would produce a strange side‐effect: the seriousness of the offence with which the defendant was charged affected both the law and process under which he or she was processed. I discuss this in the final section of this chapter.

In the decades following the passage of the Criminal Lunatics Act 1800, changes in the broader social and political context in which insanity trials took place reinforced this abiding concern with the dangerousness of individuals raising insanity at trial. The final decades of the eighteenth century had been marked by a growing public awareness of the social problem posed by the insane, generated by a developing interest in the phenomenon of insanity and a sense of what Martin Wiener calls ‘moral unease’ in subjecting ‘madmen to criminal sanctions,’ as well as a heightened concern with social order.45 The rise of insanity as a social problem at this point drew public attention to the criminally insane who, in Wiener's words, ‘kept now bumping up against the more ubiquitous law and against similarly expanding standards of respectability’.46 The first decades of the 1800s featured significant parliamentary and public debate about the ineffectiveness of the criminal law and its ability to properly take into account different levels of moral culpability.47 Concurrently, as Wiener suggests, public interest in crime, ‘both popular and serious’, was growing.48 Wiener argues that, because crime was a metaphor of disorder and loss of control, criminal laws came to occupy ‘cultural high ground’, and to be a central plank of the Victorian discourse of moralization.49 Insane defendants were positioned at the confluence of these broad social currents relating to criminality, the effectiveness of the criminal justice system and morality, which pushed the law on insanity into prominence.

The first decades of the nineteenth century were also marked by the rise of the psychiatric profession, embodying an expert knowledge of ‘madness’. The appearance of an expert knowledge about ‘madness’ in the nineteenth century significantly altered the knowledge context and the evidentiary and procedural practices (p.113) governing exculpatory ‘madness’. In this period, specialist knowledge about ‘madness’ was considerably contested and conflicted. As exemplified by the new notion of ‘moral insanity’, developing psychiatric knowledge emphasized individual will and emotions, as opposed to intellect.50 The notion of ‘moral insanity’ denoted ‘primarily an alienation of feelings, of natural sentiments’, and ‘spoke to the impulsive nature of the will, which drove the afflicted person into motiveless, revolting activity’: according to Joel Eigen, it thus addressed the issue of criminal liability directly.51 ‘Moral insanity’ and other diagnostic entities would be significant in the development of the law of insanity and other mental incapacity doctrines such as infanticide in the subsequent decades. This changing knowledge context—discussed in detail in Chapter 6—contained the germ of the changing foundation of the law of insanity—from a broader and looser idea of incapacity, to a narrower and more technical, and eventually more medicalized, idea of disability.

A ‘fierce and fearful delusion’: Daniel M'Naghten and the Creation of the M'Naghten Rules

In 1843, Daniel M'Naghten, aiming to shoot the Prime Minister, shot and killed his private secretary, Edward Drummond. M'Naghten was charged with willful murder and pleaded not guilty. At trial, the Solicitor General, referring to Arnold's Case and Hadfield's Case, stated that M'Naghten could not be excused on the grounds of insanity if he had ‘that degree of intellect which enabled him to know and distinguish between right and wrong’.52 Alexander Cockburn, M'Naghten's counsel, argued that, although M'Naghten had done the act, he should not be held responsible for it because the ‘fierce and fearful delusion’ that he was being persecuted subsisted at the time of the killing and meant that he was unable to control his actions.53 Witnesses for the prosecution and defence included lay people who had known the prisoner in Glasgow and London, and several medical experts, including Dr Monro, who was the Superintendent of Bethlem. After hearing the medical witnesses, Chief Justice Tindal stopped the proceedings and remarked that ‘the whole of the medical evidence is on one side’.54 The jury found M'Naghten (p.114) ‘not guilty by reason of insanity’. This outcome was highly controversial, with Queen Victoria and others expressing concern that the verdict was unduly lenient.

The M'Naghten Rules were formulated in response to five questions put to 15 judges of the Queen's Bench by the House of Lords, when, subsequently, they were called to appear to defend the M'Naghten decision. The now famous judicial statement drafted in response to the questions from the Lords provided in part:

[T]he jurors ought to be told in all cases that every man is to be presumed sane, and to possess a sufficient degree of reason to be responsible for his crimes, until the contrary be proved to their satisfaction; and that to establish a defence on the ground of insanity it must be clearly proved that at the time of the committing of the act, the party accused was labouring under such a defect of reason, from a disease of the mind, as not to know the nature and quality of the act he was doing, or, if he did know it, that he did not know he was doing wrong.55

The M'Naghten Rules form the current law on insanity in England and Wales, and I provide a detailed analysis of them in the next section of this chapter. While the Rules themselves have been of primary interest to legal scholars, the way in which they were created has been at least as significant as their content. The unique character of the Rules—a judicial formulation, developed independently of a specific trial, and in a legislative context—has earned them the label ‘judicial legislation’,56 and served to entrench them in the criminal law of England and Wales.

The M'Naghten Rules triggered rather than settled debate about the insanity doctrine in the years after 1843. The scope of the insanity doctrine, its purpose (to exculpate those who were not morally accountable or those who could not be deterred) and its basis (in cognitive or volitional disorder) were the subjects of discussion by legal and political commentators. As Keith Smith suggests, the debate about M'Naghten insanity turned on two issues: the meaning of ‘wrong’ (whether according to ‘the laws of God or man’) and what came to be called ‘irresistible impulse’ (the extent to which the law accommodated or ought to accommodate volitional incapacity).57 While the insanity doctrine was the subject of discussion on the part of judges and commentators, there was no significant conceptual (p.115) development of ideas of criminal non‐responsibility at this time. In Smith's words, ‘fundamental structural questions’ concerning the relationship between defences and mens rea or voluntariness ‘remained quite beyond express judicial interest’.58 Debate over the meaning of wrongness and the status of ‘irresistible impulse’ continued into the twentieth century when, as I discuss below, the Court of Criminal Appeal adjudicated both issues.

In part because of what Smith labels judicial ‘failure to articulate the theoretical basis upon which the defence rested’,59 and as a result of the intervention of expert medical professionals in trials involving insanity, the operation of the law in the decades after M'Naghten was more flexible than its technical construction suggested.60 Thus, while the M'Naghten Rules have been interpreted as a limitation on the scope of exculpatory insanity,61 the conceptual indeterminacy and the operational flexibility of criminal process provided some scope to accommodate claims for exculpation that would now fall within the bounds of automatism. Some support for this argument about the capacity of the newly formalized insanity doctrine is provided by Eigen in his study of criminal trials from the Victorian era. Eigen argues that, in this period, a diverse population of ‘mentally wayward’ defendants sought exculpation on the basis of their abnormal mental states. Against the background of a burgeoning expert psychiatric knowledge, and popular interest in mesmerism, various defendants presented the court with an array of afflictions such as ‘unconsciousness’, mental ‘absence’, amnesia, and impulse that departed qualitatively from ‘delusion, delirium and mania’, the mainstays of insanity.62 Eigen suggests that ‘unconsciousness’ was ‘originally employed as a synonym for lack of awareness’ but gradually expanded to ‘a state of mental life inaccessible to the waking person’, covering ‘the truly unknown features of a prisoner's affect’.63 This period seems to mark the beginning of the significance of unconsciousness for what would later be known as automatism.64

(p.116) A narrower idea of disability as a basis for insanity was emerging during this era—invoked for instance in the use of terms such as ‘disease’ and ‘defect’ in the M'Naghten Rules—but a lingering, moralized notion of incapacity subsisted as a basis for the insanity plea. Conditions of the late Victorian era combined to emphasize a moralized idea of abnormality as a foundation for exculpation (even as this foundation was itself gradually becoming a more medicalized concept—as a result of the rise of expert medical knowledge of ‘madness’). In the wider cultural and political frame, the late Victorian era was characterized by an emphasis on personal responsibility, which Roger Smith refers to as an ‘individualized form of conformity with the social order’.65 The aim of engaging the law as an instrument of moralization pushed the notion of a subjective test for criminal liability to the fore as a defendant's ‘powers of reason and self‐government’ were subjected to new scrutiny.66 As a result, Wiener argues that defences that classed the defendant as abnormal were more palatable than those that positioned the defendant on a continuum with non‐criminal individuals.67 For this reason, Wiener claims that the insanity plea rose to prominence while other pleas, such as intoxication, declined. According to Wiener, ‘the gradually broadening recognition of mental unsoundness’ presented less of a challenge than intoxication or provocation because it meant that excused defendants were not likely to be returned to society and ‘it did not offer a competing vision of “normal” behaviour to that of the Victorian judiciary and Home Office’.68 The effect of these changed conditions was such that, by the end of the era, the prominence of the insanity doctrine had outstripped its practical role and raised it to a position of symbolic importance in the criminal law.

Insanity As We Know It: the M'Naghten Rules

The M'Naghten Rules represent the current law on insanity in England and Wales. Their importance in the context of insanity and, more broadly, across the mental (p.117) incapacity terrain, is such that they warrant a detailed discussion, which I offer in this section. There are three limbs to the current insanity doctrine, and I discuss each in turn.

(i) ‘defect of reason’

The first limb of M'Naghten insanity, ‘defect of reason’, is both a cause and an effect: it must cause an individual not to know the ‘nature and quality’ of the act, or, alternatively, that it was ‘wrong’, and it must be an effect of a ‘disease of the mind’. The requirement that a defendant seeking to make an insanity plea must suffer from a ‘defect of reason’ has been narrowly interpreted such that ‘defect of reason’ denotes cognitive defects. In Kemp, Lord Devlin distinguished between defects of reason caused by ‘diseased’ minds and those caused by merely ‘untrained’ minds, stating that:

A defect of reason is by itself enough to make the act irrational and therefore normally to exclude responsibility in law. But the Rule was not intended to apply to defects of reason caused simply by brutish stupidity without rational power.69

Although a ‘defect of reason’ need not be permanent, a defendant must be ‘deprived of the power of reasoning’ at the time of his or her act—it is not enough that a defendant be ‘momentarily absentminded or confused’.70 The effect of interpreting ‘defect of reason’ in M'Naghten to denote cognitive defects has been to exclude defects of conation, the psychological processes of desire and volition. The narrow, cognitive component of M'Naghten has been the focus of much of the criticism directed at the insanity doctrine.71 However, some of the heat of these criticisms of M'Naghten dissipated with the introduction of diminished responsibility (which encompasses volitional incapacity).72

(ii) ‘disease of the mind’

The ‘disease of the mind’ limb of M'Naghten insanity forms the core of the doctrine, providing a discrete and ostensibly scientific basis for the exculpation that flows from a successful plea. Judicial interpretation of ‘disease of the mind’—around the idea of an internal as opposed to external cause—has not only set the scope of insanity, but also determined the dividing line between insanity and automatism. The meaning of the phrase ‘disease of the mind’ was first subject to (p.118) judicial discussion in the first instance decision of Kemp, in which Justice Devlin stated that the phrase refers to ‘the mental faculties of memory, reason and understanding’ and was intended to be a limitation on the scope of ‘defect of reason’.73 A ‘disease of the mind’ may be organic or functional, permanent or ‘transitory or intermittent’ as long as it pertains at the time the defendant acts.74 In Bratty v Attorney‐General for Northern Ireland, Lord Denning stated that ‘disease of the mind’ includes but is not limited to the major mental disorders as identified by psychiatrists and other medical professionals.75 What constitutes a ‘disease of the mind’ is a question of law for the judge and a range of conditions have been held to constitute diseases of the mind for the purposes of the insanity doctrine.76

The broad approach to ‘disease of the mind’—as a ‘disease which affects the proper functioning of the mind’77—has led to the odd situation where some physical disorders, such as hyperglycaemia, sleepwalking, and epilepsy, fall within the ambit of ‘disease of the mind’ and thus within the bounds of insanity for criminal law purposes.78 In Burgess, which concerned sleepwalking, the Court of Appeal favoured expert evidence to the effect that the defendant was in a ‘hysterical dissociative state’ when the offence occurred, which was defined as a state ‘in which, for psychological reasons, such as being overwhelmed by his emotions, the person's brain works in a different way’.79 Thus, as Burgess suffered from ‘an abnormality or disorder, albeit transitory, due to an internal factor’, the only option open to him was insanity.80 The inclusion of physical disorders within the scope of the insanity doctrine is the product of the organization of exculpatory mental incapacity doctrines according to the source of the defendant's incapacity (his or her disability).81

(p.119) In addition to consideration of the effect of a particular condition on an individual's ‘mental faculties of memory, reason and understanding’, whether mental disorder is prone to recur affects the categorization of a particular condition as a ‘disease of the mind’. Like disposal, this consequentialist consideration is underpinned by a concern with dangerousness. In a comment implying a causal relationship between mental incapacity and violence, Lord Denning stated in Bratty that ‘any mental disorder which has manifested itself in violence and is prone to recur is a disease of the mind’.82 In Burgess, the Court of Appeal stated that, although the low probability of recurrence does not mean a condition is not a ‘disease of the mind’, ‘if there is a danger of recurrence that may be an added reason for categorising the condition as a disease of the mind’.83 This approach relies on a perceived causal relationship between mental disorders and violence such that consideration of the likelihood of recurrence can be appended to more materialist concerns with the effect on ‘the mental faculties of memory, reason and understanding’ of organic or functional disorders. This naturalized or taken‐for‐granted relationship between violence and mental disorder legitimates the broad scope of ‘disease of the mind’ for the purposes of the insanity doctrine.

As this approach to ‘disease of the mind’ makes clear, concern with dangerous individuals is driving the way in which the boundary of the insanity doctrine is drawn.84 As Lord Denning stated in Bratty, an involuntary act by a person suffering from a ‘disease of the mind’ does not give rise to an unqualified acquittal because ‘that would mean that he would be let out to do it again’.85 Similarly, in Hill v Baxter Justice Devlin stated that ‘if there is some temporary loss of consciousness arising accidentally, it is reasonable to hope that it will not be repeated and that it is safe to let an acquitted man go entirely free’ but ‘if disease is present, the same thing may happen again, and therefore, since 1800, the law has provided that persons acquitted on this ground should be subject to restraint’.86 As I discuss in the next section of this chapter, the ‘disease of the mind’ requirement marks the boundary between insanity and automatism, and thus the boundary between the special verdict and an ordinary acquittal, and a broad approach to ‘disease of the mind’ has ensured a narrow scope for automatism. The effect of the judicial interpretation of ‘disease of the mind’ is to exclude those individuals whose incapacity (‘defect of reason’) is the result of an external as opposed to an internal cause. The presence of an external trigger, such as a diabetic's injection of insulin, means that an (p.120) individual can rely on automatism.87 As R D Mackay argues, the scope of the phrase ‘disease of the mind’ has ensured that most states of automatism (where that term is used descriptively) fall within the bounds of insanity and thus result in special verdicts as opposed to acquittals.88

(iii) Knowledge of the ‘nature and quality’ of the Act or that it was Wrong

The third limb of M'Naghten insanity specifies two ways in which a ‘defect of reason’ must affect an individual if he or she is seeking to rely on the insanity doctrine: it must affect either his or her knowledge of the ‘nature and quality’ of the act, or his or her knowledge that it was ‘wrong’. The third limb of M'Naghten insanity has been accorded a narrow scope and has been criticized on the basis that it restricts the exculpatory potential of the doctrine.89 In relation to the ‘nature and quality’ of the act requirement, it is likely that the M'Naghten judges regarded this phrase as ‘too clear to need explanation’,90 and it has only been subject to limited judicial attention in the twentieth century. The phrase has been interpreted to refer to the physical circumstances and consequences of the defendant's act.91 The effect of this interpretation of the ‘nature and quality’ requirement is to exclude consideration of the defendant's appreciation of the ‘moral or social nature of his act’.92 As a result of this approach, under M'Naghten strictly interpreted, the content of a defendant's delusions will be determinative of whether he or she fits within The Rules.93

By contrast with the reference to the ‘nature and quality’ of the act, the meaning of the reference to knowledge of wrongness in the M'Naghten Rules has been the subject of significant debate, beginning in the decades after M'Naghten. Over the twentieth century, wrongness for the purposes of M'Naghten has come to be given a narrow scope, and, in this respect, the law of insanity evidences the rise of a technical conception of criminal responsibility in the late modern era.94 In 1916, the Court of Criminal Appeal adjudicated the issue of the meaning of wrongness in the decision of Codère. In that decision, Lord Reading CJ stated that, for M'Naghten insanity, ‘the standard to be applied is whether according to the (p.121) ordinary standard adopted by reasonable men the act was right or wrong’.95 This interpretation meant M'Naghten wrongness was a moral standard. However, since this decision, the courts have moved to a narrower interpretation of ‘wrongness’ that equates it with ‘legal wrong’. In Windle in 1952, the House of Lords concluded that ‘wrongness’ had always meant contrary to law. Lord Goddard CJ stated that:

Courts of law can only distinguish between that which is in accordance with the law and that which is contrary to law…the law cannot embark on the question, and it would be an unfortunate thing if it were left to juries to consider whether some particular act was morally right or wrong.96

The decision in Windle means that wrongness in M'Naghten connotes legal wrongness.97 Mirroring other developments in the jurisprudence relating to mental incapacity doctrines, the reasoning in Windle indicates that a more formal and technical as opposed to moral‐evaluative standard for exculpatory insanity is considered preferable in the current era.

This third limb of M'Naghten provides the clue to understanding the different ways in which insanity may exculpate an individual. Although the usual way in which the insanity doctrine is assumed to function is to negative the mens rea element required for an offence,98 this is only a partial account of M'Naghten insanity. The relationship between insanity and mens rea is not so straightforward: insanity may negative mens rea, but it does not reduce to a denial of mens rea.99 As several commentators have pointed out, knowledge of the wrongness of an act is not an application of the ordinary rules of mens rea.100 An individual may have the mens rea for an offence, and yet, as a result of a defect of reason resulting from a disease of the mind, he or she may not know that the act was wrong. Thus, the insanity doctrine actually operates in two ways: it either negates an element of the offence (mens rea) or it has a more global impact, exculpating an individual although he or she performed the actus reus with the requisite mens rea.101

(p.122) The Appearance of a Discrete Automatism Doctrine and the Rise of Disability as a Basis for Insanity

In this section, I pick up the historical story again, to chart the appearance of a discrete automatism doctrine, and the rise of disability as a basis for insanity. There were three developments in the law of insanity in the first half of the twentieth century that prepared the ground for the appearance of a discrete automatism doctrine (and indirectly contributed to the rise to prominence of a narrow idea of disability, as a basis for exculpatory insanity). The first development related to the scope of the insanity doctrine, which was hollowed out by two distinct changes to mental incapacity more broadly. These changes reflected the ongoing formalization of the doctrines across the terrain of mental incapacity. With the passage of the Infanticide Acts 1922 and 1938, all killings of newborn children by their mothers were taken outside the reach of the law of murder and the law of insanity.102 In addition, the House of Lords decision in Beard's Case, considering the effect of intoxication on an individual charged with a criminal offence, dismissed the M'Naghten test as irrelevant to such an issue. The Court concluded that insanity and intoxication are distinct, and that it is ‘inconvenient to use the same language in charging juries in relation to different defences’.103 The reasoning of the House of Lords here hints at the rise of a technical conception of insanity, according to which it was a more circumscribed notion with precise meaning, rather than a more nebulous notion connoting incapacity in a general way.

The second relevant development in the first half of the twentieth century related to the enhanced profile of the actus reus component of criminal offences. In the House of Lords decision in Woolmington, in addition to their famous comments on the burden of proof, the judges made reference to a requirement that an offence consist of a voluntary act on the part of the defendant. The House of Lords stated that, when dealing with a murder charge, the Crown must prove ‘death as a result of the voluntary act of the accused’ as well as malice on the part of the defendant.104 The decision in Woolmington does not elaborate on what has come to be called the voluntary act requirement. However, in approving of the earlier decision of Davies, which held that there is no onus on the defendant to prove that the act alleged was accidental where intent was an element of the offence,105 it seems that the reference to a voluntary act in Woolmington connoted non‐accidental or deliberate. It was on top of these ideas about voluntariness that a discrete automatism doctrine would later develop.

(p.123) Around the same time as the decision in Woolmington was handed down, judicial and legislative discussion of M'Naghten insanity combined to definitively exclude what was termed ‘irresistible impulse’ or ‘uncontrollable impulse’ from the reach of insanity. The definitive exclusion of ‘irresistible impulse’ was the third of the three developments preparing the ground for the appearance of automatism. As a disorder of volition, ‘irresistible impulse’ had haunted the development and operation of the law of insanity throughout the nineteenth century (in ideas of ‘moral insanity’, for example). In 1923, the Committee on Insanity and Crime (the Atkin Committee), which had been convened to advise on reform of the law on insanity, proposed a new defence of ‘irresistible impulse’, which Tony Ward argues was designed to ‘provide a legal justification for verdicts which were reached by juries under the existing [M'Naghten] Rules’.106 At the same time, in Kopsch, the Court of Criminal Appeal decisively rejected the proposition that irresistible impulse fell within the bounds of the insanity doctrine. The Court referred to ‘uncontrollable impulse’ as a ‘fantastic theory’, which, if it were to become part of the criminal law, would be ‘merely subversive’.107

Taken together, the hollowing out of insanity, the Woolmington decision, and the definitive exclusion of ‘irresistible impulse’ from the insanity doctrine amplified the importance of voluntariness, will, or consciousness for criminal responsibility and shrank the official space provided for exculpation where these elements were absent. This effect came to be felt in the second half of the twentieth century. By the 1950s, when the first reference to a discrete automatism doctrine appeared in the case law, the M'Naghten Rules had been conclusively drawn to exclude volitional impairment, and diminished responsibility, which encompasses volitional impairment, had not yet been introduced into the English and Welsh law.108 The development of a discrete doctrine of automatism at this point represented a judicial response to the restricted scope of insanity and the formalization of other mental incapacity doctrines, onto which was overlaid a growing sense of the inappropriateness of the label insanity, with its increasingly technical connotations, for all claims of exculpatory ‘madness’.

The appearance of a discrete doctrine of automatism marked a reconfiguration of the merciful space around exculpatory mental incapacity. This reconfiguration has gone largely unremarked. In part because the appearance of a discrete doctrine of (p.124) automatism was depicted as merely a matter of making explicit what had always been implicit in this area of criminal law, and because the significance of the formal legal development was obscured by the descriptive use of the label automatism (evident in terms such as ‘insane automatism’ in place of insanity and ‘self‐induced automatism’ in place of intoxication), the novelty of the new contours of the terrain of mental incapacity seems to have been regarded as more a matter of nomenclature than substance. The general understanding, more implicit than explicit, seems to have been that the formal change marked more faithful adherence to the organization of mental incapacity doctrines along the lines of the cause of incapacitated conduct.109 But, what is as significant is the way in which this change quarantined the broad, moralized notion of incapacity that underpins the doctrine of automatism. I return to this point about the basis of automatism in the next section of the chapter.

At the same time as the terrain of mental incapacity was undergoing reconfiguration to produce a discrete automatism doctrine, broader social developments, extending well beyond the bounds of the criminal law, were coming to have a profound although largely taken‐for‐granted effect on the law of insanity. This effect—the rise of a notion of disability as a basis for insanity, definitively displacing the broader and more moralized notion of incapacity—was generated by the rising social profile of psychiatry and psychology as forms of expert knowledge. The post‐war era was something of a high‐water mark of faith in the transformative potential of this type of knowledge and, relatedly, in the rehabilitative ideal that suffused the criminal justice field at the time.110 It was in this era that such knowledges took on mass influence. It was in the 1950s and 1960s, when economic, social, and cultural developments combined to forge a focus on individual authenticity, self‐determination, and self‐expression—in short, a psychological way of thinking—to spread among the populations of societies such as those of England and Wales.111 Under these changed social conditions, insanity for criminal law purposes, which had already come to be more narrowly drawn, attracted a more medicalized meaning, a change which seems to have had progressively greater impact up to the current era.

In the absence of change to the M'Naghten Rules, it is necessary to look beyond the ‘law on the books’ for support for my suggestion that these extra‐legal developments had a profound impact on exculpatory insanity in the second half of the twentieth century. Some evidence can be gleaned from the difference between the first and second generations of reform recommendations regarding M'Naghten insanity, issued in the 1950s and 1970s, respectively. The first generation of reform (p.125) proposals is represented by the Royal Commission on Capital Punishment. In its 1953 Report, the Commission recommended that the jury should be directed to determine whether ‘at the time of the act the accused was suffering from a disease of the mind (or mental deficiency) to such a degree that he ought not to be held responsible’.112 As Smith argues, this proposal rested on the belief that ‘responsibility was essentially a subjective, ethical question, making the relationship between insanity and criminal responsibility inherently incapable of precise definition’.113 Alongside such a belief, this proposal also exudes a confidence about the broad acceptability of an overtly moralized insanity law. By the time of the second generation of reform recommendations, the overtly‐moralized basis for exculpation via insanity had become less palatable. In 1975, the Butler Committee proposed a new special verdict, ‘not guilty on evidence of mental disorder’, which was to be more closely linked to medical concepts and indeed to medical evidence.114 The contrast between these proposals indicates that, even in the absence of change to the M'Naghten Rules, insanity came to be thought of as grounded in a more medicalized notion of disability. This change in the grounding of exculpation on the basis of insanity is also evident in changes to practices of evidence and procedure governing insanity, which are canvassed in Chapter 6.

The Persistence of Incapacity: the Requirements of the Doctrine of Automatism

As a discrete doctrine, automatism now occupies an area that had been taken up by a capacious informal insanity law operating under a flexible criminal process. When compared with insanity, automatism rests on a broader and more overtly moralized notion of incapacity, as opposed to a narrower, more technical, and medicalized notion of disability. The doctrine of automatism has been defined according to three components, and I structure my discussion of the law around them.

(p.126) (i) ‘Total destruction of voluntary control’

To make out a plea of automatism, a defendant must suffer a ‘total destruction of voluntary control’.115 This must have the effect that the defendant does not understand the nature and quality of his or her act.116 In a line of cases concerning driving offences, the courts have concluded that anything short of ‘total destruction of voluntary control’, such as impaired or reduced control, is insufficient for exculpation on the basis of automatism. Because nothing short of involuntariness or unconsciousness is sufficient for automatism, it does not exculpate individuals who are only partially incapacitated. Although it has been recognized that consciousness and control over action are matters of degree,117 automatism in England and Wales does not (at least formally) countenance anything short of a ‘total destruction of voluntary control’. The requirement of total loss of control has a profound effect, significantly restricting the availability of automatism.118 In addition, over and above the practical effect of this requirement, because the automatism doctrine is available only to those defendants who are unconscious or acting involuntarily, automatistic defendants are able to be constructed as abnormal (a difference of kind, as opposed to a difference of degree), even if their abnormality is only transient.119

There is ambiguity about whether automatistic incapacity centres on involuntariness or unconsciousness or whether it may be either. The former approach emphasizes the effect of the relevant external factor on action or actus reus, and the latter emphasizes its effect on mental state or mens rea. The case law on automatism furnishes evidence of each approach to the meaning of automatistic incapacity. On the one hand, some of the judicial and academic discussions of automatism suggest that it is defined by the presence of involuntary conduct.120 On the other hand, (p.127) some discussions about automatism centre on unconsciousness.121 It is possible that ambiguity about the meaning of acting in a state of automatism is unavoidable because, as Glanville Williams suggests in relation to epilepsy, the distinction between impairment of consciousness and impairment of self‐control is a fine one.122 Indeed, alongside discussions of automatism as either involuntariness or unconsciousness, there are a number of comments encompassing both ideas. In perhaps the most famous definition of automatism, Viscount Kilmuir defined automatism as:

connoting the state of a person who, though capable of action, is not conscious of what he is doing.…It means unconscious, involuntary action and it is a defense because the mind does not go with what is being done.123

Overall, the ambiguity about the kind of incapacity that forms the basis of automatism indicates that the doctrine operates primarily through the principle of non‐culpability. If, in Paul Robinson's words, nearly any disability may give rise to the ‘involuntariness defence’ because the resulting ‘dysfunction is apparently sufficiently gross that it establishes its own abnormality’,124 the question of whether it entails involuntariness or unconsciousness or either is not crucial to its coherence.

The significance of the ambiguity about whether automatistic incapacity centres on involuntariness or unconsciousness lies in the way in which automatism relates to the elements of the offence. Contra arguments that automatism goes to either the actus reus or the mens rea, the most precise way to conceptualize the relationship between the automatism doctrine and the elements of a criminal offence is that it relates to both actus reus and mens rea. Automatism is a claim of incapacity that has both a physical and a mental dimension. As Norval Morris writes, when pleading automatism, the defendant is claiming that the mental element of the actus reus (voluntariness) is lacking or, alternatively, that the physical element of the mens rea (consciousness) is lacking.125 Thus, as Andrew Ashworth suggests, although ‘one way of rationalizing’ automatism is to maintain that it ‘negates the actus reus, since (p.128) it shows that the conduct or omission was not the result of the defendant acting but of something happening to the defendant’, automatism is actually a broader claim, more a denial of authorship.126 On this basis, it is clear that there is something of a disjunction between the voluntary act requirement of an offence and the exculpatory doctrine of automatism. The act requirement (voluntariness) is an offence doctrine (which excludes mere thoughts from the reach of the criminal law and triggers the special rules that ground liability for omissions), while involuntariness is more accurately conceptualized as an exculpatory doctrine, functioning to ensure that certain actors who have caused harm are not liable on the basis that they cannot control their conduct.127 The idea that automatism entails a broad claim along the lines of a ‘denial of authorship’ hints at the more global way in which incapacity affects particular individuals.128

(ii) Internal/External Factor Distinction

If an individual suffers a ‘total destruction of voluntary control’, it must also be the result of an external factor if he or she is to come within the bounds of the automatism doctrine. A range of external factors has grounded claims to automatism.129 As Lord Denning cautioned in Bratty v Attorney‐General for Northern Ireland, ‘it is not every involuntary act which leads to a complete acquittal’.130 If an individual's behaviour is the result of an internal factor, that is, it arises from a ‘disease of the mind’, he or she will only be able to rely on the insanity doctrine. As mentioned above, the principle that the existence of an internal cause precludes reliance on the automatism doctrine was developed in the first instance decision of Kemp, and confirmed by the House of Lords in Bratty.131 In Bratty, the House of Lords stated that because, in that case, the only basis for the allegedly unconscious act was a ‘disease of the mind’ (in that case, arising from psychomotor epilepsy), (p.129) ‘there can be no room for the alternative defence of automatism’.132 If, however, an individual can legitimately rely on both an internal and an external cause for his or her automatistic behaviour, he or she is entitled to raise both automatism and insanity.133

Although the internal/external distinction, built on the M'Naghten requirement of a ‘disease of the mind’, is an ‘intellectually tidy arrangement’,134 it is an artificial approach to exculpatory mental incapacity producing arbitrary results. The use of the internal/external factor distinction to mark the boundary between automatism and insanity has led to the odd situation where a diabetic individual can fall into one or other category, depending on the level of insulin in his or her blood. If a diabetic commits an offence when he or she has too little insulin (high blood sugar or hyperglycaemia), he or she will only be able to rely on insanity, not automatism, because diabetes is regarded as an internal factor.135 On the other hand, if a diabetic defendant commits an offence when he or she has an excess of insulin in their bloodstream (low blood‐sugar or hypoglycaemia), he or she will be able to rely on the defence of automatism because the injection of insulin will amount to an external factor.136

In addition to the arbitrary results produced by the internal/external factor distinction, and despite its apparent technicality, there is no clear or necessary line between internal and external causes.137 This is demonstrated by the change in the status of sleepwalking defendants. In England and Wales, sleepwalking, or somnambulism, was regarded as a condition that gave rise to automatism until the (p.130) decision of Burgess in 1991, when it was held to be a ‘disease of the mind’, thus giving rise to insanity.138 Burgess appealed his conviction for wounding with intent to cause grievous bodily harm (inflicted on a friend) on the basis that he was sleepwalking during the events that comprised the offence and should have been able to rely on automatism. The Court of Appeal dismissed the appeal, concluding that, on the evidence, the defendant's condition was properly characterized by the trial judge as a ‘disease of the mind’.139 Beyond the issue of its spurious technicality, and its arbitrary results, the internal/external factor distinction has also been criticized because it has no basis in expert medical knowledge. As the Butler Committee stated, the ‘sharp divide’ between insanity and automatism is ‘unknown to medical science’.140

For the purposes of automatism, external factors are limited to physical or tangible causes. As mentioned above, in Burgess, the defendant argued that he was in an automatistic state when the events comprising the offence took place. In concluding that there were no external factors operating on the defendant at the time of the offence, the Court of Appeal commented that ‘the possible disappointment or frustration caused by unrequited love is not to be equated with something such as concussion’.141 Similarly, in Hennessy, the defendant was unsuccessful in arguing that stress, anxiety, and depression were external factors for the purposes of automatism. There, the Court of Appeal concluded that stress, anxiety, and depression may be ‘the result of the operation of external factors’ but are not ‘in themselves separately or together external factors of the kind capable in law of causing or contributing to a state of automatism’.142 A justification for the restriction of external factors to physical factors is proffered by the Court in Hennessy which concluded that factors such as stress and anxiety were prone to recur and lacked ‘the feature of novelty or accident which is the basis of the distinction drawn by Lord Diplock in Reg. v Sullivan’.143

As this use of the ‘feature of novelty or accident’ to prop up the restriction of the category of external factors suggests, concern with dangerousness has driven the process of drawing a line between internal and external factors. A number of commentators make this point.144 The plea of automatism, arising from an external (p.131) factor, marks out an individual who is less dangerous (or not dangerous at all) when compared with an insane defendant. An external cause, such as an excess of insulin caused by an injection, or a concussion caused by a head injury, is ‘externally demonstrable’145 and constructed as non‐replicable or one‐off and can therefore safely fall within the scope of automatism. The relevance of a concern with dangerousness to the courts is neatly captured by Justice Devlin who stated that ‘if there is some temporary loss of consciousness arising accidentally, it is reasonable to hope that it will not be repeated and that it is safe to let an acquitted man go entirely free’.146

(iii) Prior Fault

Even if an individual suffers a ‘total destruction of voluntary control’ and it is the result of an external factor, he or she will only be able to rely on automatism if he or she is not in any way to blame for the automatistic state. This third and final component of automatism circumscribes exculpatory involuntariness along moral culpability lines (and ensures the scope of the doctrine lies in judicial hands).147 The effect of this component of automatism has been felt chiefly in relation those cases of automatistic behaviour caused by intoxication (via alcohol or drugs), which have been hived off and dealt with via the law of intoxication (which is also known as ‘self‐induced automatism’), an area of law marked by a palpable moral condemnation of intoxicated offending.148 Beyond quarantining automatistic behaviour caused by alcohol or drugs from that caused by other factors, the requirement that an automatistic individual evidence no prior fault has the potential to disqualify other individuals who might seek to claim exculpation on the basis of automatism. Some suggestion to this effect was given by the Court in Bailey, where the defendant had taken insulin but neglected to eat as his doctor had directed. The Court of Appeal held that an automatism plea was open to the defendant, (p.132) had he raised sufficient evidence of it.149 According to the Court, the relevant question is ‘whether the prosecution has proved the necessary element of recklessness’.150 This comment suggests that the courts might engage in a general inquiry into the defendant's prior fault in determining whether he or she may make use of the law of automatism.

The no prior fault requirement for automatism is typically hidden in the definition of voluntariness. As only certain types of conduct (those for which the defendant is not at fault) will be regarded as ‘involuntary’, the descriptive aspect of automatism (was a defendant in a state of automatism?) obscures its moral‐evaluative aspect (does he or she deserve to be held liable for the offence?). But it is this broad moral‐evaluative aspect of the doctrine that forms its core, and marks automatism out on the mental incapacity terrain. As I discuss in Chapter 2, unlike other mental incapacity doctrines, automatism does not prescribe a particular disability as a baseline for exculpation.151 This moral‐evaluative aspect of automatism accounts for what Robinson calls the ‘special role’ of the ‘involuntariness defence’ in the system of ‘excuse defences’.152 Robinson's conceptualization of automatism suggests that the fact that the doctrine tracks lines of moral culpability is justified on the basis of the shortcomings of other ‘excuse defences’. In recognizing the ‘catch‐all’ character of automatism, Robinson's account of the doctrine as a companion to other doctrines pinpoints the valuable flexibility encoded in exculpation which is independent of a particular disability. Robinson suggests that, in practice, ‘the absence of a disability requirement has the collateral effect of broadening the defence to cases beyond instances of complete lack of volition’; but, for Robinson, if something less than a ‘total lack of volition’ forms the basis for a successful plea, this may be justified because it represents ‘a proper result in cases of blamelessness that otherwise would be denied a defence’.153

(p.133) Automatism's grounding in a broad moralized notion of incapacity injects flexibility into the mental incapacity domain. Some suggestion of the value of this flexibility is provided by the 1990 Court of Appeal decision of R v T. T was charged with robbery and assault occasioning actual bodily harm. On arrest, T stated that she had been raped three days before the offence. At trial, she adduced medical evidence that she was suffering post‐traumatic stress disorder and had been in a dissociative state at the time of her offences. The Court held that rape could constitute an external factor and thus that automatism was open to T.154 The Court of Appeal decision is most accurately explained as a response to T's sympathetic situation as a survivor of rape. In other circumstances, it might not be sufficient for a defendant seeking to rely on automatism to be in a ‘dream’, as was found to be the case in this instance.155 Thus, even though, as Lord Justice Lawton put it in Quick, it is only ‘those in desperate need of some kind of a defence’ that enter the ‘quagmire of law’ that is automatism,156 the more moralized automatism doctrine provides welcome flexibility on the mental incapacity terrain.

On the Eve of Reform?

As discussed above, a concern with dangerousness has been a driving force for most of the developments regarding the insanity doctrine, including the cleaving apart of insanity and automatism from within a capacious informal insanity law. The durable legacy of a concern with dangerousness also accounts for a further unusual aspect of the law of insanity: it was not clear until recently that the insanity doctrine is available to defendants tried summarily. The Criminal Lunatics Act 1800 had restricted the special verdict to offences of treason, murder, and felony, and, despite its extension to misdemeanours via the 1840 Act, it was generally assumed that the insanity doctrine was unavailable in magistrates’ courts.157 However, in the 1997 decision of R v Horseferry Road Magistrates ex parte K, the Court concluded that the (p.134) 1800 Act had not removed the common law doctrine of insanity from summary trials and thus, that a defendant could raise insanity in a summary case when charged with an ‘appropriate charge’.158 The uncertainty over exactly what constituted an ‘appropriate charge’ was resolved when the Queen's Bench Division concluded that insanity is only available in magistrates’ courts where an individual has been charged with an offence containing a mens rea element.159

Like the problem itself, the extant solution to the question of the reach of the insanity doctrine betrays an abiding concern with dangerousness. With trials on indictment, and now trials on indictment plus summary offences with a mens rea element, broadly corresponding to serious criminal offences, a concern with dangerousness seems to account for the extension of the insanity doctrine only so far. But, with both a burgeoning number of summary offences and the advent of the Human Rights Act influencing criminal evidence and procedure in recent years, this two‐track system (which also applies to unfitness to plead) is under some pressure. In addition, greater social and cultural awareness of mental disorder, and enhanced lay familiarity with the psychiatric lexicon, raises questions about the ongoing utility of the label insanity for a plea based on a broad range of conditions (including physical conditions such as epilepsy). Indeed, the uncertainty about the availability of insanity across the spectrum of criminal offences, compliance with the European Convention on Human Rights, and the use of the ‘outdated’ label insanity are each specifically identified as factors warranting a review of the law of insanity, which is part of the Tenth Programme of Law Reform.160 Given the trajectory travelled by insanity to this point in time, it seems reasonable to expect that any reformulation of the law would closely tether the doctrine to its now solid grounding in a medicalized notion of disability.

Beyond the issue of the specific formulation of an insanity doctrine, over recent decades, dangerousness has been rearticulated as a concern with risk, a less overtly moralized and more technocratic concept. But this concern with risk has nonetheless reinscribed the significance of the backdoor concern with the disposal of insane individuals that has driven the insanity doctrine for so long. Described in various ways—as a set of technologies, or a way of thinking about interactions between individuals and collectivities, for instance—risk captures a broad social development affecting the myriad of ways in which responsibility is understood and (p.135) assigned to states, agencies, families, and individuals in the era of late modernity.161 By contrast with dangerousness, risk is more amenable to quantification, and seems less speculative. This new concern with risk now pervades criminal justice practices, impacting on the development of criminal offences,162 for instance, and is reflected in the rise of preventative detention, the eponymous practice of ‘risk assessment’ and testing, and the advent of a discrete label for certain offenders—‘dangerous and severe personality disorder’—which is an administrative rather than medical label.163 Psychiatric knowledge, practices, and professionals have been harnessed to the task of addressing risk, with, according to Nikolas Rose, a ‘logic of prediction’ supplanting a ‘logic of diagnosis’.164 Under these conditions, the consequentialist dimension of the law of insanity takes on a new lease of life, effectively revitalizing a concern with disposal that had been sustained for so long by concern with dangerousness.


(1) M’Naghten (1843) 10 Cl & Fin 200. The spelling of M'Naghten is disputed (see B L Diamond ‘On the Spelling of Daniel M'Naghten's Name’ in D J West and A Walk (eds) Daniel McNaughton: His Trial and the Aftermath (Ashford: Gaskell Books, 1977) 86–90 and R Moran Knowing Right from Wrong: The Insanity Defense of Daniel McNaughtan (New York: The Free Press, 1981) xi–xiii for discussion). I adopt a common variant of the spelling of M'Naghten.

(2) A reference to a discrete doctrine of automatism first appeared in the 1951 decision of Harrison‐Owen: R v Harrison‐Owen [1951] 2 All ER 726.

(3) N Walker Crime and Insanity in England (Vol 1: The Historical Perspective) (Edinburgh: Edinburgh University Press, 1968) 25–6. Walker suggests that, as trial by ordeal was replaced with trial by jury in the medieval era, insane individuals charged with serious offences (such as homicide) became likely to be tried and, if convicted, left to the royal prerogative of mercy.

(4) See Walker Crime and Insanity in England (Vol 1) 25.

(5) See, eg, Walker Crime and Insanity in England (Vol 1) 19. According to this informal practice, and in the context of private investigation and prosecution, the insane individual's family would provide compensation to the victim or his or her family and look after the insane person (Walker Crime and Insanity in England (Vol 1) 26). On the system of private investigation and prosecution, see P King ‘Decision‐Makers and Decision‐Making in the English Criminal Law 1750–1800’ (1984) 27(1) Historical Journal 25, 27; J H Langbein The Origins of the Adversary Criminal Trial (Oxford: OUP, 2003) 10–13.

(6) See Walker Crime and Insanity in England (Vol 1) 27–8, 36.

(7) K J M Smith Lawyers, Legislators and Theorists: Developments in English Criminal Jurisprudence 1800–1957 (Oxford: Clarendon Press, 1998) 44–5.

(8) M J Wiener Reconstructing the Criminal: Culture, Law and Policy in England, 1830–1914 (Cambridge: CUP, 1990) 59; see also King ‘Decision‐Makers and Decision‐Making’ 37; D Rabin Identity, Crime and Legal Responsibility in Eighteenth Century England (Basingstoke: Palgrave Macmillan, 2004) 31; Langbein The Origins of the Adversary Criminal Trial 58; J M Beattie Crime and the Courts in England 1660–1800 (Oxford: OUP, 1986) 530; D Hay ‘Property, Authority and the Criminal Law’ in D Hay, P Linebaugh, J G Rule, E P Thompson and C Winslow (eds) Albion's Fatal Tress: Crime and Society in Eighteenth Century England (London: Allen Lane, 1975) 17–63. If an insane defendant was convicted after the passage of the Transportation Act 1718 (4 Geo. I c.11), he or she faced the possibility of imprisonment or transportation to America or Australia from 1719, a development that ‘widened the discretionary powers of the judge and jury in the face of the increasing number of capital statutes’ that were passed in this era: see Rabin Identity, Crime and Legal Responsibility 35; see also P King Crime, Justice and Discretion in England 1740–1820 (Oxford: OUP, 2000) 355.

(9) Smith Lawyers, Legislators and Theorists 42. As Dana Rabin writes, ‘constant negotiation shaped the decision to prosecute, the gathering of evidence, the testimony given, the judge and jury reception of the information at trial, and the judge's decision to recommend for or against a royal pardon’ (Identity, Crime and Legal Responsibility 25). In this context, as Rabin plausibly suggests, some matters involving insane defendants—particularly those related to minor offences—would have been among those not to come to trial (Identity, Crime and Legal Responsibility 24).

(10) Historical studies indicate that some individuals were discharged into the care of relatives, or private asylums, while those who were poor or who seemed to constitute a continuing danger were detained in gaol: see Beattie Crime and the Courts in England 84; Walker Crime and Insanity in England (Vol 1) 42–3.

(11) Edward Arnold (1724) 16 St Tr 695.

(12) (1723) 9 Geo. 1 c.22. See generally Hay ‘Property, Authority and the Criminal Law’ 17–63.

(13) See Walker Crime and Insanity in England (Vol 1) 55 and R Moran ‘The Origin of Insanity as a Special Verdict: The Trial for Treason of James Hadfield’ (1985) 19(3) Law and Society Review 487, 502 respectively.

(14) Extracted in Walker Crime and Insanity in England (Vol 1) 56. Arnold was convicted but, as a result of Lord Onslow's intercession, he was imprisoned rather than executed. See T Maeder Crime and Madness: The Origins and Evolution of the Insanity Defense (New York: Harper and Row, 1985) 11; Walker Crime and Insanity in England (Vol 1) 57.

(15) J P Eigen ‘Delusion's Odyssey: Charting the Course of Victorian Forensic Psychiatry’ (2004) 27(5) International Journal of Law and Psychiatry 395, 398. Several scholars have interpreted this test to indicate that only total madness or a complete lack of memory and understanding would suffice for an insanity acquittal at this time (see for example Moran ‘The Origin of Insanity as a Special Verdict’ 488). Yet, the variability of insanity cases in this era militates against such generalizations. As a result, the judge's directions to the jury should be interpreted in their context: as Beattie has argued, because of the status of the victim, the court clearly pressed hard to defeat the insanity doctrine in Arnold's Case (Beattie Crime and the Courts in England 85).

(16) See Chapter 3 for a discussion of the formal qualities of ‘madness’ per my ‘manifest madness’ analysis.

(17) As Walker points out, at trial, evidence was adduced about Arnold's conduct when he was in prison after the offence. Walker argues that ‘thus the way was paved’ for the subsequent introduction of medical witnesses ‘who had examined the accused after his crime’: see Crime and Insanity in England (Vol 1) 55–6.

(18) See OBP, Thomas Draper, 17 May 1727 (t17270517‐12); OBP, William Barnelly, 9 July 1729 (t17290709‐64); and OBP, Richard Greenwood, 29 April 1767 (t17670429‐50) respectively.

(19) OBP, Benjamin Allen, 24 February 1768 (t17680224–73).

(20) R Porter Mind‐Forg’d Manacles: A History of Madness in England from the Restoration to the Regency (London: Athlone Press, 1987) x, 29.

(21) OBP, Coustantine Mac‐Yennis, 26 February 1724 (t17240226–78).

(22) OBP, Edward Bromfield, 9 September 1696 (t16960909–27).

(23) OBP, Francis David Stirn, 10 September 1760 (t17600910–19).

(24) OBP, Henry Clifford, 13 January 1688 (t16880113–30).

(25) Rabin Identity, Crime and Legal Responsibility 1–2.

(26) N Lacey ‘Responsibility and Modernity in Criminal Law’ (2001) 9(3) Journal of Political Philosophy 249, 261.

(27) In George Fletcher's terminology, this was the era of ‘manifest criminality’, whereby liability for an offence rested on the presumption that the defendant's act manifested his or her ‘criminal purpose’. See G P Fletcher Rethinking Criminal Law (New York: OUP, 2000) 232.

(28) Smith Lawyers, Legislators and Theorists 43.

(29) As discussed in Chapter 8 in relation to infanticide, these social meanings had particular gender inflections, which impacted on the ways in which women's claims to exculpatory ‘madness’ were articulated and received by courts.

(30) The brevity of the OBPs records and the informality of claims to exculpation in this era meant that factors which would later be important, such as the cause of incapacity (internal or external), were sometimes left unidentified. It is also likely that, as Nigel Walker argues in relation to the nineteenth century, some charges brought against defendants who engaged in what would now be called automatistic conduct were rejected by the grand jury and did not come to trial: see Crime and Insanity in England (Vol 1) 167.

(31) OBP, William Walker, 21 April 1784 (t17840421–13).

(32) OBP, Richard Montgomery, 30 August 1727 (t17270830–29).

(33) R v Hadfield (1800) 27 St Tr 1281 (Hadfield's Case).

(34) See Langbein The Origins of the Adversary Criminal Trial 84–5, 97–102.

(35) Extracted in Walker Crime and Insanity in England (Vol 1) 77.

(36) Extracted in Walker Crime and Insanity in England (Vol 1) 76.

(37) Extracted in Moran ‘The Origin of Insanity as a Special Verdict’ 510. Hadfield was sent to Newgate prison.

(38) Criminal Lunatics Act 1800, s 1, which was subtitled ‘An Act for the Safe Custody of Insane Persons Charged with Offences’ (39 & 40 Geo. III c. 94). The provision in the Act that a defendant found ‘not guilty on account of insanity’ would be detained ‘in strict custody…until His Majesty's pleasure be known’ maintained the tradition that the King embodied the justice and mercy of the realm: see Moran ‘The Origin of Insanity as a Special Verdict’ 515. The 1800 Act was to apply retrospectively, so as to include Hadfield: Walker Crime and Insanity in England (Vol 1) 78. Unlike the provision relating to insanity on arraignment, also contained in the 1800 Act, the insanity provision applied only to offences of treason, murder, or felony, not to misdemeanours, perhaps as the result of an assumption that the insanity doctrine applied only to capital offences: Walker Crime and Insanity in England (Vol 1) 80. The Insane Prisoners Act 1840 (3 & 4 Vict. c. 54) introduced the special verdict of ‘not guilty by reason of insanity’ for misdemeanours.

(39) Walker Crime and Insanity in England (Vol 1) 78. Rupert Cross argues that a distinction between insanity and automatism has been implicit in the criminal law since this Act, with the distinction between insanity and automatism marking the difference between indefinite detention and a complete acquittal (R Cross, ‘Reflections on Bratty's Case’ [1962] Law Quarterly Review 236, 238). But it is important to note that the parameters of as yet emergent categories of insanity and automatism were by no means fixed and that exculpation on the basis of what might now be called automatism remained informal and variable, achieved in the absence of a discrete doctrine until the second half of the twentieth century.

(40) Although it had been possible to detain insane defendants before 1800, the Act introduced a ‘more systematic means of containing them within a voluntarist legal system’ (Wiener Reconstructing the Criminal 85). As Wiener argues, this procedure offered a ‘middle path’ between humanity and security: it was an alternative to conviction and punishment, and also provided ‘new legal means to incarcerate an offender’ (Reconstructing the Criminal 85). While in theory, defendants such as Hadfield could be released if they were no longer a danger to themselves or others, in practice, the period of confinement was life (Moran ‘The Origin of Insanity as a Special Verdict’ 515). The effect of the passage of the 1800 Act was that an individual did not have to be convicted of a crime in order to be confined under the criminal law (Moran ‘The Origin of Insanity as a Special Verdict’ 517).

(41) In the Trial of Lunatics Act 1883, the special verdict was repackaged as ‘guilty but insane’, thus technically altering the form of the special verdict from an acquittal to a conviction (see Moran ‘The Origin of Insanity as a Special Verdict’ 519). However, although the special verdict was briefly regarded as a conviction—from the decision in Ireland (R v Ireland (1910) 9 Cr App R. 139) until that decision was overturned in Felstead (R v Felstead [1914] AC 534)—the special verdict has always effectively been an acquittal, evidenced by the fact that the defendant could not appeal against it (Moran ‘The Origin of Insanity as a Special Verdict’ 519; S White What Queen Victoria Saw: Roderick Maclean and the Trial of Lunatics Act, 1883 (Chichester: Barry Rose Law, 2000) 68). It was only after the passage of the Criminal Procedure (Insanity) Act 1964 that a special verdict accrued a right of appeal (s 2). The Criminal Procedure (Insanity) Act 1964 also returned the form of the special verdict to ‘not guilty by reason of insanity’ (s 1), a change which had been recommended by the inter‐war Committee on Insanity and Crime (the Atkin Committee) (Report of the Committee on Insanity and Crime (Cmd 2005, 1924), 11–12) and United Kingdom Royal Commission on Capital Punishment 1949–1953 Report (Cmnd 8932, 1953) para 456).

(42) See R Moran ‘The Punitive Uses of the Insanity Defence: The Trial for Treason of Edward Oxford (1840)’ (1986) (9) International Journal of Law and Psychiatry 171, 189; see also E Colvin ‘Exculpatory Defences in Criminal Law’ (1990) 10 Oxford Journal of Legal Studies 381, 386.

(43) R v Sullivan [1984] 1 AC 156, 172.

(44) See Chapter 6 for a historical discussion in the context of insanity; see also Chapter 3 regarding ‘manifest madness’.

(45) Wiener Reconstructing the Criminal 84. The growing public awareness of insanity was evident in the enactment of vagrancy legislation, such as the Vagrancy Act 1744 (‘An Act…relating to Rogues, Vagabonds etc…’) (17 Geo. II c.5), which meant that the insane could be held under civil laws (see Walker Crime and Insanity in England (Vol 1) 70). Public awareness of insanity was also evidenced in, and heightened by, the voluntary subscription hospitals and asylums for the insane that appeared in the second half of the eighteenth century (see W F Bynum ‘Rationales for Therapy in British Psychiatry 1780–1835’ in A Scull (ed) Madhouses, Mad‐Doctors and Madmen: The Social History of Psychiatry in the Victorian Era (London: Athlone Press, 1981) 40–4; Walker Crime and Insanity in England (Vol 1) 70).

(46) Wiener Reconstructing the Criminal 83–4.

(47) See Smith Lawyers, Legislators and Theorists 56–8.

(48) See Wiener Reconstructing the Criminal 15, and, more generally, 14–26.

(49) Reconstructing the Criminal 47–8.

(50) Reconstructing the Criminal 26; J P Eigen Witnessing Insanity: Madness and Mad Doctors in the English Court (New Haven: Yale University Press, 1995) 79; see also M J Clark ‘“Morbid Introspection,” Unsoundness of Mind, and British Psychological Medicine c.1830–1900’ in W F Bynum, R Porter, and M Shepherd (eds) The Anatomy of Madness: Essays in the History of Psychiatry (London: Routledge, 1988) 83–4; N Rafter ‘The Unrepentant Horse‐Slasher: Moral Insanity and the Origins of Criminological Thought’ (2004) 42(4) Criminology 979, 993–9; H Rimke and A Hunt ‘From Sinners to Degenerates: The Medicalization of Morality in the Nineteenth Century’ (2002) 15(1) History of the Human Sciences 59, 70–3.

(51) Eigen Witnessing Insanity 78–80.

(52) M'Naghten (1843) 10 Cl & Fin 200, extracted in West and Walk (eds) Daniel McNaughton 16.

(53) Extracted in Moran Knowing Right from Wrong 1. The passage of the Prisoners’ Counsel Act 1836 meant that M'Naghten's defence counsel could address the jury.

(54) Extracted in West and Walk (eds) Daniel McNaughton 72.

(55) M'Naghten (1843) 10 Cl & Fin 200, 210. This statement forms the core of the M'Naghten Rules but the Queen's Bench judges also responded to questions about the significance of delusions on the part of defendants claiming insanity and the role of medical evidence where it is given by someone who has not seen the defendant before trial. These parts of the judges’ response were not the focus of judicial or other attention after they were drafted and are no longer considered authoritative: see Walker Crime and Insanity in England (Vol 1) 100; T Ward ‘A Terrible Responsibility: Murder and the Insanity Defence in England 1908–1939’ (2002) 25 International Journal of Law and Psychiatry 361, 374.

(56) D J A Cairns Advocacy and the Making of the Adversarial Criminal Trial, 1800–1865 (Oxford: Clarendon Press, 1998) 178.

(57) Smith Lawyers, Legislators and Theorists 223–5. As Smith chronicles, in relation to the first issue, opinions as to the meaning of the wrongness element of the Rules varied from knowledge of illegality to the broader notion of knowledge of moral wrongfulness. In relation to volitional incapacity or irresistible impulse, the fact that the Rules did not expressly allow for exculpation on the basis of impairment of volitional power, came to produce a ‘steadily accumulating body of medical and, later, legal literature’ that supported the view that the insanity doctrine was too narrowly conceived. See also Walker Crime and Insanity in England (Vol 1) 105; Wiener Reconstructing the Criminal 270 and M J Wiener Men of Blood: Violence, Manliness and Criminal Justice in Victorian England (Cambridge: CUP, 2004) 280.

(58) Smith Lawyers, Legislators and Theorists 257–8.

(59) Lawyers, Legislators and Theorists 223.

(60) Several commentators take this view. See, eg Wiener Reconstructing the Criminal 275.

(61) See, for example, J P Eigen, ‘Lesion of the Will: Medical Resolve and Criminal Responsibility in Victorian Insanity Trials’ (1999) 33(2) Law and Society Review 425, 438–9; R D Mackay Mental Condition Defences in the Criminal Law (Oxford: Clarendon Press, 1995) 95; Smith Lawyers, Legislators and Theorists 220. In Keith Smith's words, the M'Naghten Rules constituted ‘a synthesized “restatement” of earlier case‐law, while asserting more strongly a narrow knowledge or cognitive basis of the defence, and, by implication, denying the relevance of ineffective volitional control’ (Lawyers, Legislators and Theorists 220).

(62) J P Eigen ‘Sense and Sensibility: Fateful Splitting in the Victorian Insanity Trial’ in R A Melikan (ed) Domestic and International Trials 1700–2000: The Trial In History Volume II (Manchester; Manchester University Press, 2003) 9.

(63) ‘Sense and Sensibility’ 167.

(64) It was in the context of these cases that a defendant's actions were first described as ‘automatic’: see ‘Sense and Sensibility’ 144–6. Several of the defendants identified by Eigen were granted insanity defences while others had their charges dismissed or received directed acquittals on miscellaneous grounds or miscellaneous verdicts such as ‘not guilty on the grounds of unconsciousness’ (‘Sense and Sensibility’ 138–9, 145). The place of insanity verdicts in these cases attests to the continuing flexibility and capaciousness of the insanity doctrine in the decades after M'Naghten.

(65) R Smith Trial by Medicine: Insanity and Responsibility in Victorian Trials (Edinburgh: Edinburgh University Press, 1981) 164; see also N Lacey ‘In Search of the Responsible Subject: History, Philosophy and Social Sciences in Criminal Law Theory’ (2001) 64(3) Modern Law Review 350, 365; Wiener Reconstructing the Criminal 91. This emphasis on personal responsibility raised the profile of those, such as the criminally insane, who could not attain it. Criminal and, to a lesser extent, civil law, was used as an instrument for fostering ‘self‐disciplining and gratification‐deferring personalities in the population at large’ (Wiener Reconstructing the Criminal 91). Significantly, the principle of individual responsibility which came to stand at the centre of the law was as much aspirational as actual (Reconstructing the Criminal 54).

(66) Reconstructing the Criminal 84.

(67) In Wiener's words, ‘if ordinary men were now expected to master their passions, then the only successful path to avoid a guilty plea was likely that of showing the prisoner to be not ordinary’: see M J Wiener ‘Judges and Jurors: Courtroom Tensions in Murder Trials and the Law of Criminal Responsibility in Nineteenth Century England’ (1999) 17 Law and History Review 467, 504.

(68) Wiener Men of Blood 287. The idea of a distinctive and confined class of defendants who lay outside the bounds of humanity could be ‘readily reconciled’ with Victorian expectations of personal self‐discipline (Men of Blood 287).

(69) R v Kemp [1957] 1 QB 399, 408.

(70) R v Clarke (1972) 56 Cr App R 225, 228.

(71) As the Butler Committee noted in 1975, this component of the insanity doctrine excludes many individuals, such as those with mood disorders, who can only be described as ‘mad’ (United Kingdom, Report of the Committee on Mentally Abnormal Offenders (Cmnd 6244, 1975) (‘Butler Report’) para 18.6; see also S Dell, ‘Wanted: An Insanity Defence that Can be Used’ [1983] Criminal Law Review 431). For the Butler Committee, the ‘main defect’ of M'Naghten insanity lay in the fact that it relies on a ‘now obsolete belief in the pre‐eminent role of reason in controlling social behaviour’ (para 18.6).

(72) See R v Byrne [1960] 2 QB 396 and Chapter 9.

(73) R v Kemp [1957] 1 QB 399, 407. In Kemp, the Court brought arteriosclerosis within the bounds of ‘disease of the mind’ on the basis that ‘hardening of the arteries is a disease which is shown on the evidence to be capable of affecting the mind in such a way as to cause a defect, temporarily or permanently, of its reasoning, understanding and so on’ (408 per Justice Devlin).

(74) R v Kemp [1957] 1 QB 399, 406–7 per Lord Devlin; R v Sullivan [1984] 1 AC 156, 172 per Lord Diplock.

(75) Bratty v Attorney‐General for Northern Ireland [1963] AC 386, 412. Bratty, who was charged with the murder of a woman he strangled with a stocking, testified that he had a ‘terrible feeling’ and that a ‘sort of blackness’ came over him at the time of the offence (extracted in Bratty 388) and his counsel adduced evidence of psychomotor epilepsy.

(76) These include alcoholism (DPP v Beard [1920] AC 479 at 501; Attorney‐General for Northern Ireland v Gallagher [1963] AC 349), arteriosclerosis (R v Kemp [1957] 1 QB 399), hyperglycaemia (R v Hennessy [1989] 1 WLR 287), psychomotor epilepsy (referred to in Bratty v Attorney‐General for Northern Ireland [1963] AC 386, 403 per Lord Kilmuir), and sleepwalking (R v Burgess [1991] 2 QB 92).

(77) R v Hennessy [1989] 1 WLR 287, 292.

(78) Re hyperglycaemia, see R v Hennessy [1989] 1 WLR 287; re sleepwalking, see R v Burgess [1991] 2 QB 92; re epilepsy, see R v Sullivan [1984] 1 AC 156.

(79) R v Burgess [1991] 2 QB 92, 101.

(80) R v Burgess [1991] 2 QB 92, 101.

(81) See T H Jones ‘Insanity, Automatism and the Burden of Proof on the Accused’ (1995) 111 Law Quarterly Review 475, 498; see also E Lederman ‘Non‐Insane and Insane Automatism: Reducing the Significance of a Problematic Distinction’ (1985) 34 International and Comparative Law Quarterly 819, 824; P Robinson Structure and Function in Criminal Law (New York: Clarendon Press, 1997) 92–4. I discuss the significance of a disability‐based system in Chapter 2.

(82) Bratty v Attorney‐General for Northern Ireland [1963] AC 386, 412.

(83) R v Burgess [1991] 2 QB 92, 99.

(84) As a number of writers have argued, the absence of a disposal power under the automatism doctrine has resulted in a broad approach to the phrase ‘disease of the mind’: see, for example, Cross, ‘Reflections on Bratty's Case’ 239; Mackay Mental Condition Defences 58; A Norrie Crime, Reason and History: A Critical Introduction to Criminal Law (London: Butterworths, 2001) 182.

(85) Bratty v Attorney‐General for Northern Ireland [1963] AC 386, 410.

(86) Hill v Baxter [1958] 1 QB 277, 285–6; see also R v Sullivan [1984] 1 AC 156, 172 per Lord Diplock. As Norrie concludes, ‘a broad definition of mental disease opens up the possibilities for the court to order a form of social control where the alternative, under a finding of non‐insane automatism, is that the accused walks free’ (Crime, Reason and History 182).

(87) See, for example, Quick [1973] 57 Cr App R 72.

(88) Mackay Mental Condition Defences 98; see also Norrie Crime, Reason and History 182.

(89) As R D Mackay writes, the effect of the narrow approach to the first and the third limb of M'Naghten is ‘to exclude the vast majority of mentally disordered persons from the realm of the insanity doctrine, as inevitably in most cases they will know what they are doing and that the offence they are committing is legally wrong’ (Mental Condition Defences 100).

(90) Walker Crime and Insanity in England (Vol 1) 101.

(91) R v Codère (1916) 12 Cr App R 21, 26–7.

(92) A E Gotlieb ‘Intention, and Knowing the Nature and Quality of an Act’ (1956) 19(3) Modern Law Review 270, 272.

(93) See Norrie Crime, Reason and History 180–1. To fit within M'Naghten, a defendant's delusions must be directly related to their criminal act (Crime, Reason and History 181).

(94) See Lacey ‘Responsibility and Modernity in Criminal Law’ 267–8 for discussion.

(95) R v Codère (1916) 12 Cr App R 21, 27.

(96) R v Windle [1952] 2 QB 826, 833.

(97) In a persuasive argument based on nineteenth‐century English authorities, the High Court of Australia concluded that, contra Codère and Windle, the notion of ‘wrong’ employed in M'Naghten was the same as ‘right and wrong’ or ‘good and evil’ (R v Stapleton (1952) 86 CLR 358, 369). In England and Wales, there is some indication that wrongness is interpreted more flexibly in practice (Mackay Mental Condition Defences 104; T Ward ‘Magistrates, Insanity and the Common Law’ [1997] Criminal Law Review 796, 803).

(98) See, for example, Butler Report para 18.22. For instance, Glanville Williams argues that the M'Naghten test amounts to ‘little more than an assertion that the defendant cannot be convicted without the necessary mental element’ (G Williams Textbook of Criminal Law (London: Stevens, 1978) 593). This argument is supported by the fact that the insanity doctrine is available only to offences with a mens rea element: DPP v H [1997] 1 WLR 1406, 1409.

(99) As Celia Wells points out, the proposition that insanity negatives mens rea is true only where mens rea refers to a subjective mental element (C Wells ‘Whither Insanity?’ [1983] Criminal Law Review 787, 794).

(100) See A Ashworth Principles of Criminal Law (Oxford: OUP, 2009) 143–4; J C Smith ‘Insanity—Available as a Defence in Summary Trials’ [1997] Criminal Law Review 129, 133; Ward ‘Magistrates, Insanity and the Common Law’ 802.

(101) See Colvin ‘Exculpatory Defences in Criminal Law’ 394; P Robinson ‘Criminal Law Defences: A Systematic Analysis’ (1982) 82 Columbia Law Review 199, 205. This idea about the more global impact of mental incapacity on a particular individual connects to my discussion of the formal qualities of ‘madness’ on the mental incapacity terrain: see Chapter 3.

(102) See further Chapter 8 on infanticide.

(103) DPP v Beard [1920] AC 479, 506 per Lord Birkenhead. See further Chapter 7 on intoxication.

(104) Woolmington v DPP [1935] All ER 1, 8 per Lord Sankey. The Woolmington Court also stated that the persuasive burden of proof of insanity lay with the defendant. I discuss this in Chapter 6 on evidence and proof of exculpatory incapacity.

(105) R v Davies [1913] 1 KB 573.

(106) T Ward ‘Law, Common Sense and the Authority of Science: Expert Witnesses and Criminal Insanity in England, CA 1840–1940’ (1997) 6(3) Social and Legal Studies 343, 354; see also Smith Lawyers, Legislators and Theorists 323; T Ward ‘The Sad Subject of Infanticide: Law, Medicine and Child Murder 1860–1938’ (1999) 8(2) Social and Legal Studies 163, 168. The Atkin Committee was broadly supportive of the M'Naghten formulation of insanity, concluding that M'Naghten was ‘in substance, sound’, but proposed explicitly encompassing irresistible impulse as a basis for exculpation (United Kingdom Report of the Committee on Insanity and Crime (Cmd 2005, 1924), 7–8). The subsequent Criminal Responsibility (Trials) Bill 1924, which proposed enlarging the M'Naghten Rules to include defendants ‘wholly incapable of resisting the impulse to do the act’, was defeated. See for discussion Smith Lawyers, Legislators and Theorists 323; T Ward ‘A Terrible Responsibility’ 372.

(107) R v Kopsch (1925) 19 Cr App Rep 50, 51–2 per Lord Hewart.

(108) Homicide Act 1957, s 2, as amended by the Coroners and Justice Act 2009; R v Byrne [1960] 2 QB 396. See Chapter 9 for discussion.

(109) See Mackay Mental Condition Defences 97 for discussion of the need for a distinction between what he refers to as insane and non‐insane automatism.

(110) In David Garland's analysis, this era marked the ‘most vigorous’ developments in penal‐welfare strategies: see D Garland The Culture of Control: Crime and Social Order in Contemporary Society (Oxford: OUP, 2001) 34, and more generally. For a discussion of these broader social developments on the treatment of ‘madness’ in prison, see T Seddon Punishment and Madness: Governing Prisoners with Mental Health Problems (London: Routledge, 2007) 33–9.

(111) See generally N Rose Governing the Soul: The Shaping of the Private Self (London: Free Association Press, 1999); N Rose Inventing our Selves: Psychology, Power, and Personhood (Cambridge: CUP, 1996).

(112) United Kingdom Royal Commission on Capital Punishment 1949–1953 Report (Cmd 8932, 1953) para 333.

(113) Smith Lawyers, Legislators and Theorists 327. While this approach was ‘supremely pragmatic’ (and perhaps captures jury practice), it unnerved a number of commentators (including a minority of members of the Commission) who argued against the Commission's proposal and ‘little if any support emerged for the Report's proposal’: see Lawyers, Legislators and Theorists 327–8.

(114) Aiming to ‘avoid the use of medical terms about which there may be disputed interpretations or whose meaning may change with the years’ (Butler Report para 18.17), the reformulated defence would have been comprised of two parts: (a) did the defendant know what he or she was doing?; or (b) was he or she suffering from ‘severe mental illness or severe subnormality’ at the time of the act? ‘Severe subnormality’ was to be defined as in the Mental Health Act 1959 to refer to ‘a state of arrested or incomplete development of mind’ (Butler Report para 18.30). ‘Severe mental illness’ was to be defined to encompass significant impairment of intellectual functions, mood, thinking, perception and delusional beliefs (Butler Report para 18.35). The language of the proposed new verdict—‘severe mental illness or severe subnormality’—would have meant that, in practice, expert witnesses would have decided the availability of insanity (Mackay Mental Condition Defences 137), and unlike the Committee's proposals on disposal of insane defendants (Butler Report paras 18.42–18.45), which were implemented in the Criminal Procedure (Insanity and Unfitness to Plead) Act 1991, the proposals about the substance of insanity were not implemented.

(115) Attorney‐General's Reference (No 2 of 1992) [1994] QB 91, 105 per Lord Taylor of Gosforth.

(116) R v Hennessy [1989] 1 WLR 287, 291; see also S Prevezer ‘Automatism and Involuntary Conduct’ [1958] Criminal Law Review 440, 440–1.

(117) In Isitt, Lawton LJ stated that ‘it is a matter of human experience that the mind does not always operate in top gear. There may be some difficulty in functioning’: R v Isitt (1978) 67 Cr App R 44, 48. For discussion, see N Morris ‘Somnambulistic Homicide: Ghosts, Spiders and North Koreans’ (1951) 5 Res Judicatae 29, 32; see also N Levy and T Bayne ‘Doing without Deliberation: Automatism, Automaticity and Moral Accountability’ (2004) 16(3) International Review of Psychiatry 209.

(118) In Watmore v Jenkins, the defendant was charged with dangerous driving following a hypoglycaemic episode and acquitted. His acquittal was overturned on appeal on the basis that he had not suffered a ‘complete destruction of voluntary control’ ([1962] 2 QB 572,587). In Broome v Perkins (Broome v Perkins (1987) 85 Cr App R 321), the Court concluded that automatism was unavailable because, at certain stages of the journey, ‘the defendant's mind was controlling his limbs and that thus he was driving’ (333). Likewise, in Attorney‐General's Reference (No 2 of 1992) [1994] QB 91, the defendant, who adduced expert evidence that he was in a condition of ‘driving without awareness’, was not able to rely on a defence of automatism because he had merely ‘reduced’ control (105).

(119) I take up this issue of differences of degree versus differences of kind in the context of diminished responsibility: see Chapter 9.

(120) See, for example, R v Harrison‐Owen [1951] 2 All ER 726; Attorney‐General's Reference (No 2 of 1992) [1994] QB 91; S Yeo ‘Clarifying Automatism’ (2002) 25 International Journal of Law and Psychiatry 445, 446. The early definition of automatism offered in Watmore v Jenkins (Watmore v Jenkins [1962] 2 QB 572) reflects this approach to the doctrine. In that case, Winn J opined that automatism was ‘a modern catch‐phrase which the courts have not accepted as connoting any wider or looser concept than involuntary movement of the body or limbs of a person’ (586).

(121) See, for example, R v Stripp (1979) 69 Cr App R 318, 323 per Ormrod LJ. For discussion, see I D Elliott ‘Automatism and Trial by Jury’ (1967–1968) 6 Melbourne University Law Review 53, 60; P Fairall ‘Automatism’ (1981) 5 Criminal Law Journal 335, 335 and P Fairall ‘Voluntariness, Automatism and Insanity: Reflections on Falconer’ (1993) 17 Criminal Law Journal 81; B McSherry, ‘Epilepsy, Automatism and Culpable Driving’ (2002) 21 Medicine and the Law 133; G Williams, Textbook of Criminal Law (London: Stevens, 1978) 608–9. In Charlson (R v Charlson [1955] 1 WLR 317), Barry J compared the defendant's case with that of an epileptic: ‘the actions of an epileptic are automatic and unconscious, and his will and his consciousness are not applied to what he is doing; he is not in control of his actions’ (320).

(122) Williams Textbook of Criminal Law 611.

(123) Bratty v Attorney‐General for Northern Ireland [1960] AC 386, 401, quoting the Court of Criminal Appeal judgment in Bratty. Similarly, in R v Sullivan [1984] 1 AC 156, 173 Lord Diplock referred to actions which were ‘unconscious and thus “involuntary” in the legal sense of that term’.

(124) Robinson Structure and Function in Criminal Law 85–6.

(125) Morris ‘Somnambulistic Homicide’ 30; see also S Prevezer ‘Automatism and Involuntary Conduct’ 445.

(126) See Ashworth Principles of Criminal Law 88–90.

(127) See Robinson Structure and Function in Criminal Law 31–6: see also W Wilson Criminal Law: Doctrine and Theory 2nd edn (London: Longman, 2003) 227. This disjuncture between the voluntariness requirement and exculpatory automatism has been regarded as something of a paradox in criminal law. Norrie argues that the criminal law employs a ‘narrow physical conception of the nature of involuntariness’ for the purposes of the law of actus reus which ‘squeezes out subjective excuses’ derived from broader moral bases (Crime, Reason and History 110). But, Norrie argues, consideration of prior fault as required by the automatism doctrine has the effect of ‘re‐moralising’ involuntariness for the purposes of exculpation, ensuring ‘convictions in situations where the judges believe that control must be affirmed, and application of the technical test might secure an acquittal’ (Crime, Reason and History 111). Norrie concludes that the ‘structural dynamic of excluding and re‐admitting substantive moral issues’ lies at the root of the modern criminal law (Crime, Reason and History 58).

(128) I discuss this point about the more global impact of mental incapacity on particular individuals in my analysis of the formal qualities of ‘madness’ on the mental incapacity terrain: see Chapter 3.

(129) These include cerebral concussion (Bratty v Attorney‐General for Northern Ireland [1960] AC 386, 414 per Lord Denning), a blow from a stone or an attack from a swarm of bees (referred to in Hill v Baxter [1958] 1 QB 277, 283 per Lord Goddard), hypoglycaemia (R v Quick [1973] 57 Cr App R 722; R v Budd [1962] Crim LR 49; R v Bingham [1991] Crim LR 433) and dissociation (per the Canadian decision of R v Rabey (1980) 15 CR (3d) 225; R v T [1990] Crim LR 256).

(130) Bratty v Attorney‐General for Northern Ireland [1963] AC 386, 410.

(131) See R v Kemp [1957] 1 QB 399 and Bratty v Attorney‐General for Northern Ireland [1963] AC 386.

(132) Bratty v Attorney‐General for Northern Ireland [1963] AC 386, 404 per Lord Kilmuir; see also at 410 per Lord Denning.

(133) R v Roach [2001] EWCA Crim 2698). In Roach, the Court stated that ‘the legal definition of automatism allows for the fact that, if external factors are operative upon an underlying condition which would not otherwise produce a state of automatism, then a defence of (non‐insane) automatism should be left to the jury’ ([28] per Potter LJ). This rule allowing an individual to raise both automatism and insanity is subject to the rules of evidence and procedure governing exculpatory incapacity. See Chapter 6.

(134) E Baker ‘Human Rights, M'Naghten and the 1991 Act’ [1994] Criminal Law Review 84, 89.

(135) In R v Hennessy [1989] 1 WLR 287, the defendant was charged with one count of theft (of a car) and one count of driving while disqualified. Hennessy claimed that he was in a state of automatism at the time of the offences, admitting that he had failed to take his insulin for ‘two or three days’ before he committed the offences. The Court of Appeal upheld Hennessy's convictions and the ruling of the trial judge that his state of mind was caused by a disease (diabetes) and so the defence of automatism was not available.

(136) R v Quick [1973] 57 Cr App R 722; R v Bingham [1991] Crim LR 433. Quick, an insulin‐dependent diabetic who was charged with assault occasioning actual bodily harm, claimed that he was hypoglycaemic at the time of the assault because he had taken insulin but had not eaten much food. On appeal, the Court of Appeal ruled that the defence of automatism should have been left to the jury because a ‘malfunctioning of the mind of transitory effect caused by the application to the body of some external factor such as violence, drugs [such as insulin], including anaesthetics, alcohol and hypnotic influences cannot fairly be said to be due to disease’ as is required for the ‘disease of the mind’ element of the M'Naghten Rules (922 per Lord Justice Lawton).

(137) This has been widely recognized: see for instance M Goode ‘On Subjectivity and Objectivity in Denial of Responsibility: Reflections on Reading Radford’ (1987) 11 Criminal Law Journal 131, 141; J Edwards ‘Automatism and Criminal Responsibility’ (1958) 21 Modern Law Review 375, 377; R F Schopp Automatism, Insanity and the Psychology of Criminal Responsibility (Cambridge: CUP, 1991) 83–4.

(138) R v Tolson (1889) 23 QBD 168; Bratty v Attorney‐General for Northern Ireland [1963] AC 386, 409 per Lord Denning and at 415 per Lord Morris.

(139) R v Burgess [1991] 2 QB 92, 101 per Lord Lane CJ; see R D Mackay and B J Mitchell ‘Sleepwalking, Automatism and Insanity’ [2006] Criminal Law Review 901 for discussion.

(140) Butler Report para 18.22; see also I Ebrahim et al ‘Violence, Sleepwalking and the Criminal Law: Part 1: The Medical Aspects’ [2005] Criminal Law Review 601, 603; P Fenwick, ‘Somnambulism and the Law: A Review’ (1987) 5(3) Behaviourial Sciences and the Law 343, 350 and P Fenwick ‘Epilepsy, Automatism and the English Law’ (1997) 16 Medicine and the Law 349, 351. The distinction drawn by Lawton LJ in Quick between hypoglycaemia and hyperglycaemia has been described as ‘medically naïve’ (C Howard and P T D'Orban ‘Violence in Sleep: Medico‐Legal Issues and Two Case Reports’ (1987) 17 Psychological Medicine 915, 923).

(141) R v Burgess [1991] 2 QB 92, 98 per Lord Lane CJ.

(142) R v Hennessy [1989] 1 WLR 287, 294.

(143) R v Hennessy [1989] 1 WLR 287, 294.

(144) See, for example, Lederman ‘Non‐Insane and Insane Automatism’ 824; Mackay Mental Condition Defences 41, 58; Schopp Automatism, Insanity 81–3; W Wilson et al ‘Violence, Sleepwalking and the Criminal Law: Part 2: The Legal Aspects’ [2005] Criminal Law Review 614, 617. As these commentators explain, policy concerns have meant that the automatism doctrine, which results in a complete acquittal, has been narrowly drawn, while the insanity doctrine requirement of a ‘disease of the mind’ has been expansively drawn, ensuring that dangerous defendants are caught by the latter and subject to the disposal powers attached to the special verdict.

(145) Goode ‘On Subjectivity and Objectivity’ 143.

(146) Hill v Baxter [1958] 1 QB 277, 285.

(147) As a normative matter, this position has been advocated by several commentators: see, for example, S Kadish ‘Excusing Crime’ [1987] California Law Review 257, 266–7, 286–9. It has also been suggested that the restriction around prior fault should apply to insanity: see E W Mitchell Self‐Made Madness: Rethinking Illness and Criminal Responsibility (Aldershot: Ashgate, 2003).

(148) In Bratty v Attorney‐General for Northern Ireland [1963] AC 386, the House of Lords stated that the automatism doctrine must be viewed as separate and distinct from the plea of voluntary intoxication (414 per Lord Denning; see also R v Lipman [1970] 1 QB 152). The jurisprudence on intoxicated offending is marked by the idea that a defendant is culpable for voluntary intoxication. In Majewski (DPP v Majewski [1977] AC 443), Lord Edmund‐Davies contrasted cases of voluntary intoxication by drugs or alcohol with those excuses in which ‘the actor is wholly free from fault in relation to the onset of a mental state’ (487). Further, the idea that the defendant is at fault for becoming intoxicated in the first place is one of the rationales for the particular rules comprising the law on intoxicated offending (see, for example, R v Hardie (1985) 80 Cr App R 157, 162 per Parker LJ). See further Chapter 7.

(149) R v Bailey [1983] 1 WLR 760. Bailey was charged with wounding with intent to cause grievous bodily harm under the Offences Against the Person Act 1861. At first instance, the recorder directed the jury that self‐induced incapacity could not form the basis of an automatism doctrine and the defendant was convicted. On the basis that ‘there may be material distinctions between a man who consumes alcohol or takes dangerous drugs and one who fails to take sufficient food after insulin’, the Court held that Bailey's self‐induced automatism (or intoxication) could be taken into account in relation to any offence, not just a ‘specific intent’ offence: R v Bailey [1983] 1 WLR 760, 764–5 per Lord Griffiths.

(150) R v Bailey [1983] 1 WLR 760, 765.

(151) See further Chapter 2.

(152) According to Robinson, this special role is that of a ‘catch‐all excuse’, which may be used when the defendant's control is ‘impaired by a disability other than one of those recognized in the traditional excuse defences’, such as insanity and intoxication (Robinson Structure and Function in Criminal Law 166). This means that the function of the defence is to ensure that certain actors who have caused harm are not liable on the basis that they cannot control their conduct (Structure and Function in Criminal Law 31, 36; see also Kadish ‘Excusing Crime’ 259). Robinson argues that no particular cause is stipulated for a defence of involuntariness because ‘total lack of volition is an obvious and convincing ground for exculpation’ (Structure and Function in Criminal Law 166).

(153) Structure and Function in Criminal Law 167. Where a defendant has raised an issue of impaired capacity at the time of the offence, but for some reason this is insufficient to ground a defence of automatism, it may be taken into account on sentence. This has been important in cases involving driving offences (see, for example, Broome v Perkins (1987) 85 Cr App R 321). In the Court of Appeal decision of R v Isitt (1978) 67 Cr App R 44, the Court considered an appeal from a dangerous driving conviction. Isitt had a hysterical fugue while driving. As discussed above, the Court of Appeal upheld his conviction because, ‘physically at any rate’, he had been driving the van, so he was precluded from relying on automatism. The Court noted that, although the psychiatric evidence did not amount to a defence, it might be relevant by way of mitigation (49).

(154) R v T [1990] Crim LR 256. The prosecution had argued that post‐traumatic stress disorder was a ‘disease of the mind’ and the only defence open to T was insanity. In response to this argument, the Court stated that post‐traumatic stress involving a normal person was not a ‘disease of the mind’ (257). The prosecution had also argued that T's ability to open the pen knife, which was used during the course of the robbery, demonstrated that she had retained partial control over her actions. The Court concluded that T was acting as though she was in a ‘dream’ (257). The prosecution's arguments might be thought to have defeated T's argument on one of two grounds: the external factor requirement and/or the ‘total loss of self‐control’ requirement of the automatism defence.

(155) The Court of Appeal's decision is in contrast with the driving offences decisions, referred to above, in which anything short of ‘total loss of self‐control’ was fatal to a claim for exculpation via the automatism defence.

(156) Quick (1973) 57 Cr App R 722, 734.

(157) The Butler Committee worked on this assumption but recommended extending the defence to magistrates’ courts (para 18.19). This recommendation was not implemented.

(158) R v Horseferry Road Magistrates ex parte K [1997] QB 23, 46.

(159) DPP v H [1997] 1 WLR 1406, 1409. If a (potentially) insane defendant appearing in a magistrate's court is charged with a strict liability offence, he or she may receive an ordinary acquittal. Alternatively, the defendant may be subject to an inquiry into his or her unfitness and, if the defendant is found to have ‘done the act or made the omission charged’, magistrates have discretion to make a hospital or guardianship order (Mental Health Act 1983, s 37(3)).

(160) See 〈http://www.justice.gov.uk/lawcommission/areas/insanity.htm〉 (last accessed 9 September 2011), and 10th Programme of Law Reform, Law Commission No 311 (London, 2008), available at 〈http://www.justice.gov.uk/lawcommission/publications/programmes‐law‐reform.htm〉 (last accessed 9 September 2011). In foreshadowing its review of insanity, the Law Commission specifically identified the inappropriateness of the label ‘insanity’ for intellectual and physical disorders such as epilepsy.

(161) See for discussion, P O'Malley Risk, Uncertainty and Government (London: Glasshouse Press, 2004) and other work by the same author.

(162) See for discussion, L Zedner ‘Pre‐crime and Post‐criminology?’ (2007) 11(2) Theoretical Criminology 261–81 and V Tadros ‘Crimes and Security’ (2008) 71(6) Modern Law Review 940–70.

(163) See J Peay Mental Health and Crime (London: Routledge, 2010) 175–86.

(164) See N Rose Powers of Freedom: Reframing Political Thought (Cambridge: CUP, 1999) 261 and 260–3 more generally.