Automatism

(from Wikipedia) Automatism is a rarely used criminal defense. It is one of the mental condition defenses that relate to the mental state of the defendant. Automatism can be seen variously as lack of voluntariness, lack of culpability (unconsciousness) or excuse (Schopp). Automatism means that the defendant was not aware of his or her actions when making the particular movements that constituted the illegal act. For example, Esther Griggs in 1858 threw her child out of a first floor window believing that the house was on fire, while having a night terror. Brian Thomas strangled his wife in their campervan in a more recent case in Aberporth in an episode of rapid eye movement sleep behaviour disorder (a disorder related to sleepwalking), where he dreamed there was an intruder on top of his wife. The defense of automatism is denying that the person was acting in the sense that the criminal law demands. As such it is really a denial-of-proof – the defendant is asserting that the offence is not made out. The prosecution does not have to disprove the defense as is sometimes erroneously reported; the prosecution has to prove all the elements of the offence including the voluntary act requirement. Automatism is a defense even against strict liability crimes like dangerous driving, where no intent is necessary.

Scope
Automatism is arguably the only defense that excludes responsibility by negating the existence of the actus reus which uniquely allows it to be a defense to both conventional and strict liability offences (although this argument could be extended to the status defense of insanity too). Strict automatism is a denial of actus reus and therefore most commonly used as a defense against strict liability offences. There are a number of reasons why a person may go into a state of automatism, including dissociation or hypo/hyperglycemia. Unconsciousness is the defense of denial of mens rea, which is easier to prove and hence more commonly used for non-strict liability crimes. For example in cases of homicidal sleepwalking the illegal act is typically not denied but the intent to kill is. The defendant will typically be perplexed and confused and will not cover up the episode. Kenneth Parks, after killing his mother-in-law and severely injuring his father-in-law, drove to the police station stating that he thought he’d killed some people. The person’s movements seem purposeful – the sleepwalker interacts with his environment in a limited way. Nonetheless the sleepwalker is not conscious of his actions. The use of the term automatism for these situations causes some confusion, as in these cases it is really the lack of intent on the part of the defendant which denies the mens rea of the offence rather than the actus reus (although this distinction is problematic in many instances), better called unconsciousness. Intention is a problem in crimes of strict liability. Very few people intend to crash their vehicles, so clearly something better than intent is required to define automatism.

Another issue with automatism is that when the issue is raised by the defense as a realistic defense (an evidentiary basis), the prosecution then has to prove beyond reasonable doubt that the defendant was acting voluntarily. This is the case for several other defenses e.g. duress. The justification for this is that voluntary action is part of the definition of the offence, and therefore something under the presumption of innocence the prosecution has to prove. The evidentiary burden was laid down in Hill v Baxter where the defense of automatism failed because there was no good evidence for the alleged black out. Evidentiary burden means that the defendant needs to provide evidence to satisfy the judge that the issue should be put to the jury, which normally requires medical evidence (although R v Woolley,^[3] in which an HGV driver crashed after sneezing, proved an exception).

Exclusions
Because automatism is such a comprehensive defense, there are various exclusions to an automatism defense. The person must not be at fault. The classic example of this is falling asleep at the wheel of a car (Kay v Butterworth). Although one is not responsible for acts done while asleep, one can and is rightly held responsible for driving in a state where one would fall asleep at the wheel. The issue of prior fault applies to many diabetics who suffer hypoglycaemia while driving. Voluntary (and often involuntary) intoxication cannot cause legal automatism.

Causes of insanity will come under the M’Naghten Rules as insane automatism and lead to the special verdict of ‘not guilty by reason of insanity’ rather than a straight acquittal in sane automatism. The M’Naghten Rules require a “disease of mind”, which has been defined in case law as an internal cause. This is medically nonsensical, and does not always bear much relationship to continuing risk which is the main justification. This means that insane automatisms do not require total loss of voluntary control (see below). Thus they are easier to prove in some circumstances, but conversely the burden of proof is on the defendant. Sleepwalking was initially an exception to the internal/external doctrine until the case of R v Burgess.

The most contentious qualifier is that there must be a total loss of control. In Attorney-General’s Reference No 2 of 1992,^[5] this definition of legal automatism was confirmed. A lorry driver had crashed, and his defense (backed up by expert evidence) was that the monotony of motorway driving had caused him to go into a state of ‘driving without awareness’ where although he could make minor adjustments to follow the road he was not truly conscious of driving. This followed the decisions of Watmore v Jenkins and Broome v Perkins where diabetic drivers who had driven 3 miles or more were held to not have the total loss of control necessary for the defense of automatism. This definition is problematic, and the Law Commission, Butler Committee and leading legal academic RD Mackay have all argued that this definition is too restrictive.

Classically automatisms in the legal sense have been defined as spasms, reflexes, convulsions or acts committed in a state of unconsciousness e.g. sleep. However there have been cases where the automatism defense was successful when none of these apply. In R v T^[9] the defendant had been raped a few days prior to committing a robbery. She was clearly conscious of what she was doing, but in a dissociative state due to post-traumatic stress disorder from being raped. However when Isitt^[10] drove away from a collision and evaded a police roadblock in a dissociative state, the defense was not successful. The Canadian court expressed a logical way of distinguishing such cases in Rabey.^[11] There the defendant went into a dissociative state due to being spurned. It was held that such a commonplace occurrence was not the sort of external stimulus that would cause legal automatism (although the insanity plea would be open to him).

Omissions
The problem with omissions and automatism is that the strict legal definition requires total loss of control. A person may well not be able to avoid a crash even though he has some residual control. In this case, the law is imposing liability for failing to do the impossible. Several commentators including HLA Hart^[12] have suggested that responsibility for omissions must be framed with reference to the actor’s capabilities at the time, rather than the objective test.

Research regarding Automatism

Sleep related violence, alcohol and sleepwalking.
Ebrahim, Irshaad, Fenwick, Peter
Brain: A Journal of Neurology, Vol 135(11), Nov, 2012. pp. 1-2.
Abstract:
Comments on an article by F. Siclari et al. (see record 2010-26185-004). Although Siclari et al. cite some relevant studies, in our opinion, they do not adequately discuss the possible causal link between alcohol ingestion and disorders of arousal. Siclari et al. stated that criteria for establishing the putative role of an underlying sleep disorder in a specific violent act have been proposed by Mahowald and coworkers and Siclari et al. partially reproduce, as Table 4 at page 3505, material from that earlier publication including the statement: ‘Alcohol or drug intoxication precludes the use of disorder of arousal in forensic cases’. More generally, there is a long history of clinical and forensic case reports linking alcohol intake to sleepwalking, automatism and sleep related violence. Siclari et al. reference a study by Hartmann where alcohol ingestion was shown to precipitate sleepwalking in a clinically diagnosed sleepwalker but do not acknowledge the implications of this case report. Although not quoted, a further example exists in the case of ‘R versus Lowe’.

R v Leonboyer: The Role of Expert Witnesses in Psychological Blow Automatism Cases.
Wells, Helene, Wilson, Paul
Psychiatry, Psychology and Law, Vol 11(1), Apr, 2004. pp. 167-173.
Abstract:
Automatism is a defence in criminal violence that raises some critical issues about the role of the expert witness in the determination of guilt and innocence. The recent Australian case of R V Leonboyer illustrates some of these issues including psychiatrists and psychologists going beyond their area of expertise, establishing reasonable doubt on the basis of possibilities rather than probabilities and failing to differentiate between automatic behaviour and its causes. Though there may be a necessity to utilise the experience of psychologists and psychiatrists to explain the complex issues arising from the defence of automatism, there are particular problems in using such expert witnesses in this under-researched and complex area.

Automatic sources of aggression.
Todorov, Alexander, Bargh, John A.
Aggression and Violent Behavior, Vol 7(1), Jan-Feb, 2002. pp. 53-68.
Abstract:
This paper reviews research on automaticity with particular relevance to aggression. Once triggered by environmental features, preconscious automatic processes run to completion without any conscious monitoring. The basic experimental technique for studying automatic processes is priming. This paper reviews studies showing that priming, including subliminal priming, of mental constructs related to aggression leads to reliable effects on perceptions, judgments, and behavior. Specifically, after such priming, people perceive ambiguous behaviors as more aggressive and tend to act more aggressively themselves as well. It also reviews studies showing that: a) prolonged exposure to violence can result in the development of chronic accessibility of aggressive constructs affecting how the social environment is interpreted, and b) even goal-directed behavior can be automatically triggered by situational features if this behavior is consistently and frequently enacted in the same situation.

Sleep violence: Forensic science implications: Polygraphic and video documentation.
Mahowald, Mark W., Bundlie, Scott R., Hurwitz, Thomas D., Schenck, Carlos H.
Journal of Forensic Sciences, Vol 35(2), Mar, 1990. pp. 413-432.
Abstract:
Presents 4 case studies of males (aged 16–73 yrs) and reports that sleep-related violence can be associated with a number of diagnosable and treatable sleep disorders, including night terrors/sleepwalking, nocturnal seizures, REM sleep-behavior disorder, sleep drunkenness, and psychogenic dissociative states occurring during the sleep period. Potentially violent automatized behavior, without consciousness, can and does occur during sleep. Violence resulting from these disorders may be misinterpreted as purposeful suicide, assault, or even homicide. Sleep-related violence must be added to the list of automatisms. A classification system of both waking and sleep-related automatic behavior is proposed, with recommendations for assessment of such behavior.

Who needs capacity?
Buchanan, Alec
International Journal of Law and Psychiatry, Vol 40, May, 2015. pp. 1-5.
Abstract:
The UK Law Commission’s Discussion Paper, Criminal Liability: Insanity and Automatism, recommends introducing the concept of capacity to the insanity defence. The concept of capacity has an established role in those parts of the law that concern the validity of the decisions that people make, for instance in composing a will or entering into a contract. Making mental capacity a criterion for criminal responsibility in a mentally disordered defendant, however, is potentially problematic. First, the term capacity already has several different meanings in the literature on the jurisprudence of mental abnormality. Second, using the concept of capacity in the way that the Law Commission proposes poses difficulties that relate to the provision of testimony by expert witnesses.

Mental incapacity and criminal liability: Redrawing the fault lines?
Peay, Jill
International Journal of Law and Psychiatry, Vol 40, May, 2015. pp. 25-35.
Abstract:
The proper boundaries of criminal liability with respect to those with questionable mental capacity are currently under review. In its deliberations in the areas of unfitness to plead, automatism and the special verdict of not guilty by reason of insanity the Law Commission for England and Wales have been cognizant of particular difficulties in fairly attributing criminal responsibility to those whose mental capacities may or may not have impinged on their decisions, either at the time of the offence or at trial. And they have referenced the potential breaches of the European Convention on Human Rights (ECHR) posed by the state of our current laws. However, in their efforts to remedy these potential deficiencies is the Law Commission heading in a direction that is fundamentally incompatible with the direction embodied by the United Nations Convention on the Rights of People with Disabilities (CRPD)? Whether one must cede sensibly to the other, or whether some compromise might emerge, perhaps through an extension of supportive services or through the development of disability-neutral criminal law, forms the subject of this paper.

What is Automaticity? An Analysis of Its Component Features and Their Interrelations.
Parent Book Series:
Frontiers of social psychology
Moors, Agnes, Houwer, Jan De
Social psychology and the unconscious: The automaticity of higher mental processes. Bargh,
Abstract:
The authors provide a detailed analysis of the concept of automaticity and its component features. First, as have previous such analyses (e.g., Bargh, 1994; more recently Conrey, Sherman, Gawronski, Hugenberg, & Groom, 2005), they argue against a simple automatic vs. controlled processes dichotomy, in which a given process of interest is considered either entirely automatic or entirely controlled. Instead, on both empirical and theoretical grounds, the authors conclude in favor of deconstructing the process in terms of its component features (i.e., intentional or not, controllable or not, efficient vs. consumptive of limited attentional capacity, in awareness or not), because many, perhaps most, processes of interest are actually mixtures of automatic and controlled features. In harmony with the deconstruction approach, the authors provide an analysis of the concept of automaticity in terms of its component features, but they go further to analyze each component feature as well. But the authors do not leave their deconstructed Humpty Dumpty just sitting there on the pavement in pieces, they try to put him back together again, by addressing the important next question of whether there might be some combinations of features which are more likely than others, and further, whether any combination of these features is possible. For example, processes we are aware of are also more likely to be controllable, and those we are not aware of are unlikely to be intentional. The chapter discusses which of these combinations are more versus less likely to co-occur. Overall it provides much needed clarity as to the basic conceptual issues at stake in contemporary automaticity research.

Clarifying automatism.
Yeo, Stanley
International Journal of Law and Psychiatry, Vol 25(5), Sep-Oct, 2002. pp. 445-458.
Abstract:
This article examines the ruling in the case of Stone v The Queen as regards the presumptions of voluntariness and of mental disorder automatism. The clear pronouncement in Stone is that the crux of automatism is involuntariness comprising a complete lack of capacity to contain one’s conduct. Consequently, unconsciousness, whether total or impaired, is not essential for automatism to exist. This clarification produces several important practical outcomes regarding the use of expert evidence. One is that such evidence should be directed to showing whether the defendant’s conduct was involuntary as opposed to whether he or she was unconscious or semiconscious at the time of the alleged crime. Another is that, in cases where mental disorder automatism is pleaded, expert evidence should be used to help determine whether the defendant’s alleged criminal conduct was involuntary, and not whether he or she lacked understanding of the nature of the conduct or its wrongfulness. An analysis of two previous leading Canadian decisions on automatism in light of the clarifications to the law on automatismachieved by Stone follows.

Diminished capacity and automatism as a defense.
Barnard, Philip G.
American Journal of Forensic Psychology, Vol 16(2), 1998. pp. 27-62.
Abstract:
Presents a case study of the use of automatism as a diminished capacity defense of a male patient charged with first degree arson. Automatism is a complex legal concept which relates to diminished capacity. The case cited involved many significant physical and psychological elements including a severe reactive depression as well as lack of sleep and nutrition resulting in a dissociative episode during which a crime was committed. The forensic psychologist involved evaluated, treated, and testified at the criminal trial and the subsequent deposition for a civil lawsuit.

Diminished capacity defenses considered.
No authorship indicated
Mental & Physical Disability Law Reporter, Vol 8(6), Nov-Dec, 1984. pp. 527-528.
Abstract:
In Polston v. Wyoming, 685 P.2d 1 (Wyo. Sup. Ct. 1984), the Wyoming Supreme Court upheld a ruling that a mayhem defendant who lost his memory was not entitled to present a defense of automatism because there was no evidence that he lacked criminal intent. In Pennsylvania v. Davis, 479 A.2d 1077 (Pa. Super. Ct. 1984), a Pennsylvania appeals court upheld a rejection of a diminished capacity defense based on lay and expert testimony. In Gurganus v. Florida, 451 So. 2d 817 (Fla. Sup. Ct. 1984), the Florida Supreme Court ordered a new trial for a murder defendant who had not been permitted to present expert testimony on the combined effects of drug and alcohol consumption on his ability to form the required specific intent.