Sydney Law Review
STEVEN T YANNOULIDIS[*]
This paper examines the conceptual confusion inherent in the current formulation of the insanity defence. This, it is contended, results from the differing goals and methodologies of law and psychiatry. It is only through the explication of these differences that a reformulation on a more transparent and principled footing may arise. To this end, the main part of the paper seeks to clarify three central tropes or metaphors informing the current law.
Beginning with ‘mental illness’ I show that the law in adopting a ‘disease’ analogy has failed to adequately bring to light the legal and moral nature of the inquiry. Instead, the pathologising of the issue conceals social defence considerations that underlie the law’s concern.
The concept of ‘rationality’ as the characteristic of mental impairment is shown not to be objectively ascertainable. Rather, cognition is to be understood as a process by which pragmatic solutions are affected. The historicity of this purported norm necessitates that the term ‘rationality’ be jettisoned in favour of ‘cognitive competency’.
Criminal responsibility is analysed from the perspective of what lies behind or produces actions. The complexity inherent in human action is seen to involve not only cognitive competence but also an actor’s volitional state. This, it is argued, should be reflected in any ascriptions of criminal responsibility.
In the light of the analysis undertaken it is proposed that in place of the current common law formulation the defence should be put on a legislative footing in the following terms:
A person is not criminally responsible if at the time of the commission of the offence he or she had a mental impairment which included in its symptoms or consequences a loss of cognitive competency to think of the reasons which people are expected to regard as sufficient grounds for refraining from commission of the offence.
It is believed that such a formulation has the advantage of making explicit the claims of both law and psychiatry: the law’s concern with social defence, and psychiatry’s expertise in the assessment of mental impairment.
[*] PhD Candidate, Law Faculty, Monash University, Victoria. The author gratefully acknowledges the assistance of Dr Bernadette McSherry who reading earlier drafts of this paper made several valuable suggestions. Errors and omissions remain the author’s.
This paper examines the nature of the common law defence of insanity, rooted in the M’Naghten Rules, and cognate related defences. It is concerned with one particular goal: the substantive reformulation of the insanity defence. Ancillary to this is a question of methodology: how to proceed with the task of construction subsequent to a critical analysis of the existing defence.
My approach takes as a guiding principle Austin’s remark that ‘words are our tools, and, as a minimum, we should use clean tools’. As such I will analyse the way in which the tropes of ‘mental illness’, ‘rationality’ and ‘criminal responsibility’ not only describe but also inform the insanity defence. In this way it is hoped to reveal the ‘fit’ between words and the realities they describe. It is through a preliminary process of deconstruction that reconstruction is hoped to be brought about.
The substantive part of the paper involves a critical examination of the terms involved. I begin with an overview of the current law. This is followed by an analysis of the three tropes forming the title of the paper. Finally, a reformulated defence of mental impairment is proposed in the light of the foregoing analysis.
Informing both the choice of topic, and methodology, is the work of contemporary theorists in the area of criminal law and philosophy. Such work attempts to understand the nature of the conflicts surrounding liberal conceptions of criminal law, in particular, the contradictions inherent in viewing criminal law as both protecting individual rights and as a system of social control. Social practices within and around the arena of ‘criminal justice’ are viewed by these new critics not, as liberal and critical theorists would have us believe, as contradictory and confused. Rather, policing, prosecuting, sentencing and punishing are seen to be part of a delicately balanced equilibrium. The maintenance of this equilibrium is seen to depend on ‘holding tensions at bay by exploiting logically contradictory discourses but disguising the fact by moving between different levels of, and spheres for, analysis’.
The insanity defence offers in its engagement of two radically different modes of discourse, that of law and psychiatry, a particularly fertile ground for an examination of the nature of this ‘delicately balanced equilibrium’. The defence is shown to operate by means of a strategy of partial conceptual stipulation, (‘disease of the mind’), and an appeal to ordinary usage (‘knowledge of wrong’). While at one level the defence is
confused and contradictory, at another ‘it is part of a complex strategy which allows the criminal law to keep various balls of different shapes and colours in the air at once’. It is with the tracing of these shapes and colours that this paper is concerned.
I begin with a discussion of the substantive law. A general remark on the concept of criminal responsibility is followed by a consideration of the nature of the so-called M’Naghten Rules. This leads to an outlining of the concept ‘disease of the mind’, along with the notions of cognition and volition. The issue of automatism is viewed as a means by which the Rules’ inherent shortcomings and dispositional rigidity are circumvented. I finish this Part with a brief summary of the main points dealt with.
The concept of criminal responsibility rests on the presumption that individuals possess the capacity to make rational choices to act or refrain from acting. Where a person, while understanding the significance of the act, performs it voluntarily and intentionally, he or she will be held responsible.
There are some mental states that impair an individual’s ability to meet these requirements. So, where a person behaves in a state of automatism, involuntariness may exculpate him or her from liability. Where, due to a ‘disease of the mind’, an individual is incapable of knowing that what they do is wrong, that person may be excused on the ground of insanity. Similarly, a ‘disease of the mind’, short of insanity, may preclude an individual from forming the requisite intention for a particular crime.
Cases during the eighteenth century interpreted Hale’s metaphor of the insane being, ‘not as reasonable creatures, their actions [being] in effect in the condition of brutes’, as necessitating a complete absence of reason in order for the defence to apply. It was against such a background that the common law was outlined in the following case.
Daniel M’Naghten was acquitted, on the ground of insanity, of murdering the Prime Minister’s secretary, Drummond, whom he had mistaken for Sir Robert Peel. M’Naghten was under the insane delusion that the Tory party was persecuting him and that his life was endangered. The acquittal created such furore that the House of Lords put certain questions to the judges regarding the scope of the insanity defence, the answers to which became known as the M’Naghten Rules. These form the foundations of the modern law of insanity in the common law world.
The judges were asked five questions relating to insanity as a defence. The answers to the second and third questions contain the famous ‘knowledge of right and wrong’ test:
[T]he jurors ought to be told in all cases that every man is presumed to be 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 the 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 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 what he was doing was wrong.
During the trial of M’Naghten, Tyndal CJ told the members of the jury to acquit if they thought that the defendant did not have sufficient reason to apply the moral distinctions of which he or she may have been aware.  In this fashion the complete incapacity attendant upon the judicial interpretation of Hale’s test was reformulated. 
It is to be noted that the Rules provide those accused with two lines of defence: (i) they must be acquitted if, because of a disease of the mind, they did not know the nature and quality of the act; (ii) or, if they did know the nature and quality of the act, they are to be acquitted if, because of a disease of the mind, they did not know it was ‘wrong’. The primary consideration under either limb is whether the accused was suffering from a ‘defect of reasoning from disease of the mind’. If accused persons were unaware that their act was ‘wrong’ for some other reason, this will generally not amount to a defence, for neither ignorance of the law, nor good motive, normally affords a defence. The M’Naghten Rules have been interpreted as confining the insanity defence within very narrow limits, focusing as they do on the purely cognitive aspect of an accused’s behaviour.
According to the Rules, it must be established that cognitive incapacity, ‘defect of reason’, was caused by a ‘disease of the mind’. As such the mere fact that an accused suffers from impaired reasoning powers is not sufficient. A causal link between this and an underlying ‘disease’ is called for.
The meaning of the term ‘disease of the mind’ is a legal rather than a psychiatric question. The issue is whether the accused’s mental faculties were impaired by illness, not whether he or she was suffering from a recognised mental illness. Insanity is a social judgment founded upon but not precisely representing, a medical diagnosis. Whether a particular condition amounts to a disease of the mind, such that the issue is suitable to go to the jury, is a question of law for the judge. It is up to the jury to decide the question of whether or not the accused was insane at the time of committing the criminal act. The question will involve consideration of whether expert and other evidence establishes the existence of a disease of the mind.
There need be no actual disease in the sense of a deterioration of the material, physical constitution of the mind. All that is required is that the ‘functions of the understanding are through some cause, whether understandable or not, thrown into derangement or disorder’. The expression will cover all conditions clinically recognised as mental illness, including those with an organic base. Hence, purely physical diseases, which are only of incidental concern to psychiatrists, such as arteriosclerosis, epilepsy, and diabetes, may satisfy the requirements of the term. Ultimately however, ‘the condition of the brain is irrelevant as is the question of whether the condition of the mind is curable or incurable, transitory or permanent’.
To this degree the limits to the notion of ‘disease of the mind’ are determined by the concept of pathology. ‘Defects of reason’ stemming from an underlying pathological condition fall within the ambit of the concept. Those ‘conditions of intense passion and other transient states attributable to the fault or to the nature of man’ are excluded from it.
A non-pathological factor may be viewed more as ‘the reaction of a healthy mind to extraordinary external stimuli’.
There are two limbs to the cognitive component of the Rules (‘defect of reason’): knowing the ‘nature and quality’ of the act, and a knowledge of wrongfulness.
At common law the expression ‘nature and quality’ of the act has been taken to refer to the physical nature and consequences of an offence rather than to its moral aspects. It is important that the accused does not know what he or she is doing, or does not appreciate or foresee the possible effects of his or her conduct. Examples include situations where the accused is mistaken about the physical nature of the act, for example, where an insane person cuts another’s throat believing it is a loaf of bread that is being cut. Another example is where the accused decapitates someone just to see what the person looks like without a head. It will be rare that a defendant’s mind is so disturbed as to preclude them from realising the physical nature and quality of the acts and it is normally under the second limb that insanity pleas have been made.
The requisite knowledge of wrongfulness poses an ambiguity: ‘wrong’ according to law, or according to the standards adopted by reasonable people? The question is whether the accused was able to appreciate the wrongness of the particular act at the particular time. It has always been clear that if he or she knew that their act was contrary to law, he or she knew it was wrong for this purpose. The accused is also liable where, while not knowing the act to be contrary to law, knows it was wrong ‘according to the ordinary standard adopted by reasonable men’.
The High Court, in Stapleton v the Queen considered the situation where the accused knows the act to be contrary to law, but believes that he or she is doing right, and that people would regard the act as right. The Court held that if the accused believed that his or her act was right according to the ordinary standards adopted by reasonable persons, he or she would be entitled to an acquittal. So, wrong means wrong according to the ordinary principles of reasonable people, rather than contrary to law. The Court held that the main question is whether or not the accused was able to appreciate the wrongfulness of the criminal act, such that he or she was able to ‘reason about the matter with a moderate degree of sense and composure’. Where the accused could not so reason, it may be said that the accused could not know that what he or she was doing was wrong.
Where an accused satisfies the cognitive aspect of the defence, that is, knew what he or she was doing and that it was wrong, the claim that due to a disease of the mind he or she could not prevent themselves from doing the act in question, will not satisfy the defence. If as a result of the condition of irresistible impulse the defendant did not know either the nature and quality of the act or that it was wrong, then the defence of insanity is open. For, as the High Court held in Attorney-General (SA) v Brown ‘uncontrollable impulse ... may afford strong ground for the inference that a prisoner was labouring under such a defect of reason from disease of the mind as not to know that he was doing what was wrong.’ On appeal to the Privy Council it was held that for the defence to succeed there must be medical evidence to support the conclusion that irresistible impulse or defect of volition proceeds from mental illness.
Strictly meaning ‘action without conscious volition’, automatism has come to denote in the criminal law, ‘conduct of which the doer is not [fully] conscious’. Acts that occur during this time are considered to be involuntary movements of the body rather than ‘acts’ for the purpose of liability. The question is whether there was an absence of all the deliberative functions of the mind so that the accused person acted automatically. Consciousness is not necessarily fatal to a plea of automatism.
Automatism requires evidence to be disproved by the prosecution. It has arisen as a result of the combination of two pressures. First, it has arisen because of the limitation in the M’Naghten Rules, that the exculpatory mental state of the accused must proceed from a disease of the mind. Secondly, it has developed because a verdict of insanity is not an outright acquittal, but in fact subjects the defendant to indefinite detention, a situation clearly not called for in certain situations. In effect, automatism stems from a conceptual failure to define insanity sufficiently widely, along with an unwillingness to adopt a more flexible approach to issues of disposition of those acquitted.
The distinction between insanity and automatism will turn on whether the state of unconsciousness proceeds from a disease of the mind. If it proceeds from such, the defence is one of insanity or insane automatism, which is irrelevant to the question of voluntariness. If it does not derive from a disease of the mind automatism is available as a ‘defence’. The implications of this distinction are apparent at both a substantive and procedural level. In the instance of sane automatism, the defendant is unconditionally acquitted, while in the instance of insane automatism the defendant is usually subject to detention in a psychiatric institution.
In the instance of insane automatism the onus will be on the defendant to prove, on the balance of probabilities, insanity at the time of the offence. In the case of sane automatism the defendant will only have to satisfy an evidential burden on the issue of voluntariness, the persuasive burden will then be on the prosecution, to negative the assertion of involuntariness beyond a reasonable doubt.
The conditions giving rise to sane automatism include reflexive reactions, spasms, and convulsions. In such situations, quite clearly, the accused’s bodily movements are not the result of an exercise of choice on his or her part but are produced by some other agency. Instances, which are more problematic, are those where the alleged lack of voluntariness stems from a mental state. Some such instances are concussion caused from a blow to the head, sleepwalking, the consumption of alcohol and drugs, hypoglycaemia, or dissociation arising from extraordinary stress, and epilepsy. It should be noted that evidence of some degree of control over bodily movements does not preclude automatism. Where the actions of the accused may be said to be involuntary, the cause of the action does not matter.
The following points may be noted. First, criminal responsibility is held to be predicated on the capacity to make rational choices to act or refrain from acting. Such a capacity must be exercised in a voluntary and intentional fashion. This paradigm will be considered below in Part 5.
Secondly, in relation to mental illness, the Courts’ willingness to interpret the Rules in a more liberal fashion is in fact an acknowledgment of the interrelationship between cognition and volitional impairment. This issue brings to the fore the observation that ‘insanity’ is a social judgment founded upon a medical diagnosis. The notion of ‘disease of the mind’, while determined by the concept of pathology, will be shown to be, in more borderline situations, far more contingent. Norms will be seen to be less descriptive than prescriptive. This will be considered below in Part 3.
Thirdly, the concept of the rational, autonomous individual, which underpins ascriptions of criminal responsibility, is in the instance of automatism severely stretched. Necessitated by the rigidity of interpretation accorded to the Rules and strict dispositional orders, sane automatism in fact requires the fiction of ‘involuntary goal-directed behaviour’. This has no foundation in scientific research nor does it accord with our moral intuitions. This highlights the conceptual difficulties at the heart of the trope of rationality. This will be considered further in Part 4.
In this part I will consider the notion of ‘disease of the mind’. I will begin with the way the law employs the concept of ‘disease’. In particular, I will focus on the way in which the law’s failure to provide a definition in terms of moral responsibility has necessitated recourse to the notion of ‘dangerousness’. I will then consider the concepts of ‘disease’, ‘health’, and ‘illness’, and the extent to which these are sociogenic constructs. Finally, I will turn to psychiatry’s ‘biopsychosocial’ model of mental disorder. It will be shown that this model, while an improvement on the legal conception, suffers from conceptual confusion. The model seeks to describe syndromes in psychological terms, while relying on physiological mechanisms to explicate underlying causes. A possible response to these difficulties is put forward by way of conclusion.
(i) Legal Usage of the Term
For several reasons the law of insanity is not coextensive with the psychiatric conception of mental disorder. First, legal insanity is an excuse for wrongdoing, not a diagnosis of the accused’s mental condition. Excuse of an insane individual is predicated on his or her inability to act rationally, that is, to understand what he or she is doing and why, and to act on the basis of that understanding. As such the scope of the insanity defence in not an empirical matter that can be settled by psychiatrists.
Secondly, the legal framework begins with the premise that the function of understanding is through some cause, whether understandable or not, thrown into derangement or disorder, which may or may not be due to some organic cause. The mere fact of organic damage alone is insufficient. It is the fact of disorder which is decisive, not the cause of it. The legal definition seeks to express a criterion by which an individual may escape responsibility, and further, to exclude those things for which an accused should not be excused, such as violent tempers and so on. In contrast, the psychiatric conception of mental illness is tied to diagnostic categories made up of clinically significant behavioural or psychological syndromes or patterns that occur in an individual. A common misconception is that a classification of mental disorders classifies people, when actually what are being classified are disorders that people have. The legal enquiry is whether illness exonerates the defendant from blame for his or her actions, not whether he or she is suffering from a recognised mental illness.
Thirdly, the law requires that the mental disorder caused the wrongdoing. As the defence of insanity functions as an excuse for wrongdoing, not as a diagnosis of mental condition, the behaviour must be shown to be connected with the psychological state. The psychiatric conception of mental disorder is tied up with the problem of diagnosing and treating severe mental disorders. Psychiatric nosology however, does not carry with it any necessary implications regarding the causes of an individual’s mental disorder. These categories are, moreover, in a constant state of flux, open to revision, reformulation, and challenge.
Fourthly, the common law insanity defence is not equivalent to the so-called ‘product test’ which holds that an accused has a good defence to a criminal charge ‘if his unlawful act was the product of mental disease or mental defect’. The fallacy in this view rests on a conceptual misunderstanding of the way in which diseases affect action. On the basis of the ‘product test’ were a brain lesion to cause a violent impulse, regardless as to whether the defendant could have resisted the impulse, he or she would be free to plead the defence. This however is to confuse causation with excuse. Causation is not an excuse, for all behaviour is
caused. If causation were an excuse, no one would be held responsible for any behaviour, criminal or not. Neither is causation the equivalent to compulsion, a species of excuse.
In the instance of physical diseases, to be excused from blame for a disease entails our being excused from blame for actions stemming from that condition. No responsibility accrues here for the simple reason that responsibility presupposes agent causation. An individual’s movement of a limb during an epileptic seizure is morally neutral, and is described as a movement rather than an action. In the instance of a mental disorder however a different type of causation is at work.
Mental ‘diseases’ affect behaviour through the medium of psychological states, such as thoughts, beliefs and desires. Even if one were able for example, in the case of schizophrenia, to uncover an underlying brain state, it would not follow that the person so suffering is not responsible for the ‘symptoms’ of the illness. This would apply only where the claim ‘Madness is not blameworthy’ is held to entail the belief that ‘madness mitigates blame for wrongdoing’. That is, we may disapprove of a person’s misbehaviour even where we recognise that it is entirely in character. While we do not blame an individual for having the characteristics that he or she possesses, we do blame him or her for acting in character, since we assume that ordinary people are capable of keeping the negative elements of their character in check.
Three tests have been formulated judicially in response to the threshold question of did the accused at the time of the commission of the offence suffer from a ‘disease of the mind’?:
• The recurrence test which holds that if a mental state is likely to recur it should be considered a disease of the mind; • The internal/external test which provides that if the mental state is ‘internal’ to the accused, as opposed to arising from an external cause, it should be defined as a disease of the mind; • The sound/unsound mind test where a disease of the mind is considered on this test to be evidenced by the reaction of an unsound mind to its own delusions or external stimuli.
It is a policy-oriented question that underlies the legal interpretations of the term ‘disease of the mind’. Namely, is this individual likely to cause harm to others if not confined? If so, then his or her mental condition will be considered to be a disease of the mind for the purposes of the insanity defence.
This is overt in the ‘recurrence test’ with its assessment of the potential dangerousness of an individual as the ground for determining his or her suitability for release. It is also implicit in the other tests, based as they are on the assumption that only those with unsound minds or with mental conditions arising from some internal source should be detained.
As a criterion of mental illness this concern with the disposition of those considered dangerous confuses two issues: the utilitarian issue of what to do with an acquittee who may be dangerous in the future and the moral question of the conditions for ascription of criminal responsibility. It is important to recognise that the question of disposition is irrelevant to what the accused did in the past. As such, dangerousness and treatability are incorrectly considered in relation to issues of responsibility as they relate to past acts.
The law has yet to fashion a construct of disease, which reflects concepts of responsibility. At present ‘disease of the mind’ is understood as a disorder of the reasoning process from some cause (in whole or in part internal to the individual) which vitiates responsibility, if it produces deprivation of any of the capacities. This has required resort to the question of future dangerousness and the disposition of the individual accused as a limiting factor. Campbell argues that while ‘[t]his may be social policy ... it is not a legal principle of moral culpability or responsibility’.
A theory of medical norms may hold that the body is composed of certain structured systems (forms) each of which has an assignable range of normal functioning. Defect or disorder of such systems relative to such functioning constitutes a sufficient condition of disease. Yet the precise nature of this functioning, in terms of which disease is to be specified, cannot be determined by noting a series of observable form/function correlations. For example, anginal pain may arise from pulmonary hypertension, not coronary disease. The notion of normal or healthy functioning cannot be simply assigned to the localised processes and structures of the body. What is normal must be construed not as a fixed point but as a range of variations.
Both general and psychiatric medicine, insofar as they suppose themselves to be value-neutral sciences, presuppose that human beings have a set of ‘natural’ functions. This set of ‘natural’ functions, physical and psychological, is said to be essential to our respective natures, not
merely accidental. However, the seemingly normative nature of medical discourse stems from a failure to distinguish the human animal from the human person. It is impossible to speak of psychiatric disease and disorder exclusively in terms of the condition of the human animal. The disorders are formulable only in terms of the mental processes of culturally emergent persons. As such any ascription of ‘natural functions’ to human persons should not be provided, in the context of psychiatry, in a way that ignores the culturally prepared goals of human societies.
Such goals may be formulated in terms of ‘rational minima’, that is, requirements considered least controversial or objectionable to agents endowed with a minimal measure of rationality and which are more or less presupposed by every significant human society. Such minima may include the norms of health and disease. They amount in effect to prudential values to which persons subscribe: avoidance of death, prolongation of life, restriction of pain, gratification of desires, assurance of security of person, body, property, and the like.
In the field of medicine, disease is whatever is judged to disorder or to cause to disorder, in the relevant way, the minimal integrity of body
and mind, relative to these prudential values. Conversely, the criterion of health is the adequate performance of functions, physiological and psychological, in the service of these prudential values.
While performance may be socially assessed, the criteria of health are not primarily social. It is misconceived to equate ill health with social deviation or maladjustment. This follows from the observation that to say that, ‘ “Smith is mentally ill” implies “Smith is socially maladjusted in the society concerned” ’, means that the latter claim states only a necessary condition for the former to have relevance. It does not state a sufficient condition, because ‘Smith is mentally ill’ may be false. Nor does ‘Smith is socially maladjusted in the society concerned’ state either a necessary, or sufficient condition, for the truth of ‘Smith is mentally ill’. Smith can be a social misfit in the society concerned without it being either relevant or true to say that Smith is mentally ill.
(i) Methodological Overview
The Diagnostic and Statistical Manual of Mental Disorders (DSM–IV–TR), replaces the term ‘mental illness’ with ‘mental disorder’ which it defines as ‘a clinically significant behavioural or psychological syndrome or pattern that occurs in an individual and that is associated with present distress (for example, a painful symptom) or disability (that is, impairment in one or more important areas of functioning) or with a significantly increased risk of suffering death, pain, disability, or an important loss of freedom’. Mental disorders may be defined on various levels of abstraction, for example, syndromal pattern, dysfunction, aetiology, each of which is an indicator for a mental disorder, but none of which is equivalent to the concept. 
Psychiatry has clearly adopted a ‘biopsychosocial’ model, which includes recognition of the importance of social, psychological, and biological factors in the aetiologies of mental disorders. In the instance of an individual with schizophrenia, characteristic signs and symptoms include both ‘positive’ symptoms, reflecting a distortion of natural functions, and ‘negative’ symptoms, reflecting a diminution or loss of normal functions, persisting for a period of at least six months (Criterion A and C). No single symptom is pathognomonic of schizophrenia; the diagnosis involves the recognition of a constellation of signs and symptoms. The individual must experience dysfunction in one or more areas of functioning (Criterion B: for example, interpersonal relations, work or education, or self-care). Further, the disturbance must not be due to the direct physiological effects of a substance or a general medical condition (Criterion D and E).
As discussed above, a theory of medical norms may hold that the body is composed of certain structured systems. These ‘regulative principles’ require viewing a human being as an entity whose operating mechanisms conform, more or less, to the ‘natural’ design of the type of organism to which it belongs. Where such functioning is found to be outside what is accepted as the normal range, then this is taken as prima facie evidence of internal malfunctioning. This is expressed as something having gone wrong with the organic machinery of the system. When we speak of psychological reactivity, we are committed to suppose that certain sorts of machinery are at work, and responsible for these reactions. However, unlike the somatic case the details of their operations, or what these mechanisms are, remain uncertain.
Models may be used where underlying causes of relationships, said to exist, are still unknown; that is, when a real theory is not available. Reliance on a statistical model, with its inherent concept of deviancy from a norm, provides psychiatry with the necessary support to buttress complex theories and hypotheses. Specific mechanisms are postulated which are formal, or merely structural, in nature. In any ascription of schizophrenia, for example, an empirical statement is being asserted that not only describes and classifies an individual’s clinical manifestations, it also helps to explain them. For the diagnosis maintains that the clinical picture is the manifestation of the impaired or morbid working of the individual’s psychological machinery.
With mental illness, such formal systems tend to be metaphorically identified. Norms assign operative characteristics to human nature rather than to somatic entities. In this regard, the norms of health and disease tend to correspond with putative norms of ‘happiness’ and ‘non well-being’.
Mental illness is not a straightforward, empirical concept of the sort we find in natural science. Nomological (causal) explanation is roughly the standard in the natural sciences while some kind of hermeneutic understanding seems crucial for social sciences. However, in psychiatric settings theoretical constructs tend to be confused with clinical observations, ‘understanding’ mistaken for ‘explanation’.
(ii) Szasz’s Critique
Szasz’s work has sought to expose the categorical error involved in the conflation of sets of dichotomous terms such as, ‘disease’ or ‘illness’ on the one hand, and, on the other, what he considers to be descriptions of personal behaviour, such as ‘schizophrenia’.  Szasz contends that the former terms, by definition, require a demonstrable histopathology or pathophysiology. The suggestion that schizophrenia is a ‘mental illness’, in the absence of an aetiological basis, which, according to Szasz, does not, and cannot exist, gives rise to the claim that psychiatric nosology is used to aid and abet a ‘myth of mental illness’. Various responses have been made to these claims, two of which will be considered here.
First, a belief that mental illness is a disease process, with an organic or biochemical underlying cause, with certain symptoms, and specific treatments and prognosis, underlies the ‘medical model’ of psychiatric aetiology. The model is best suited to organic states with discoverable underlying causes of physical malfunction. Epistemologically, neither individual symptoms nor a syndrome is sufficient to infer the presence of a particular disease state. As discussed in the preceding section, there is no precise form/function correlation for all diseases allowing abnormal structure and abnormal function to be compared. Hence, it is a presumption that underlying states of malfunction are present on the basis of symptomatology.
The medical model when applied to psychological manifestations or symptoms presupposes that, for example, the ‘disease’ of schizophrenia refers not only to the psychological states making up its symptoms but also to unknown physical states of malfunction producing them. In fact, even in the instance of ‘operational’ conditions, wanting any explanation in organic terms, characterised rather by a set of psychological and behavioural manifestations, the expression ‘disease’ is employed. The confusion inherent in such use of metapsychological language is succinctly stated by Radden who observes that ‘a metaphorical extension has come to acquire a literal meaning’. In effect, psychiatric explanations are held to give causes of mechanical movements or processes, rather than giving reasons for actions. A more cautious convention would view the medical model as presupposing an analogy, not an identity, between madness and physical disease: madness is a ‘disease’ not a disease.
Secondly, McAuley views Szasz as seeking to exploit a conceptual difficulty lying at the heart of contemporary discourse in psychiatry, by means of a false dichotomy held to exist between the terms ‘organic’ and ‘behavioural’. Diseases need not be characterised by a clearly discontinuous change in bodily functioning, traceable to some discreet cause. They may be systemic, having their origins in the internal state of the organism, and arise as a transformation and breakdown of otherwise normal functions. In the domain of mental impairment, we may refer to genetically influenced variations in brain organisation as dispositions to varying forms of mental disorder. Where these biological dispositions appear at the extreme end of the spectrum of abnormality, we find the ‘disease’ entities of psychiatry becoming clearly defined. This re-conceptualisation of disease broadens it to include, in addition to physical lesions and their symptoms, a constitutional proneness or predisposition. Even in the absence of an underlying brain dysfunction, there may exist an operative tendency-to-disorder in addition to the prospect that an organic substructure may eventually be discovered. It is only a wider concept of mental illness which is called for at this descriptive phase of psychiatry’s development, one not tied down to known organic pathology and/or aetiology. Szasz’s dichotomy of ‘organic’ and ‘behavioural’ is, on this view, a misdescription of modern psychiatric thought and practice.
Fulford argues that viewing mental impairment along the lines of a scientific model of bodily disease, a ‘science-half-field-view’, is not so much wrong as incomplete. A ‘full-field’ view, by contrast, would recognise, in addition to facts and failure of functioning, value and failure of action.
The evaluative connotations of mental illness reflect differences in our evaluations of mental phenomena, in contrast to the relatively uncontentious value judgments in physical illness. Yet these connotations represent legitimate variations in what counts as mental illness. Rather than leading to the medicalisation of life’s problems, as Szasz would contend, this non-descriptivism makes explicit the evaluative logical element in the diagnosis of mental illness. As such, it sounds a warning for any immediate descriptive attitude sought to be advanced.
In the full-field view, mental impairment is to be understood not as defective cognitive functioning, but rather as a defect in ‘reasons for action’, that is, a defect in ‘practical reasoning’, or a defect in the reasons we have for the things we do. Actions are defined by means of the reasons for which they are done. If the reasons themselves are defective, there is a constitutive, rather than merely executive, failure of action. There is, in this sense, no action at all. This will be considered in Part 4.
The above discussion of the term ‘disease of the mind’ has brought to light the following key issues. First, differences in the nature of the enquiry with which law and psychiatry are concerned were noted. In particular the fact that legal insanity is an excuse for wrongdoing, not a diagnosis of the accused’s mental condition. Policy considerations were seen to inform the legal definition of ‘disease of the mind’. These social defence considerations lead to dispositional issues being decided on grounds of ‘dangerousness’. That is, what the accused is likely to do in the future, not what he or she may have done in the past.
The psychiatric conception replaces ‘mental illness’ with ‘mental disorder’. Adopting a biopsychosocial model, diagnostic categories serve both a descriptive and explicative purpose. While an improvement on the legal explanation, psychiatry also adopts an essentially scientific model of bodily disease. Szasz’s argument attacks the conceptual confusion inherent in the formulation of mental processes in terms of mechanistic function.
While the concept of ‘disease’ is readily applicable to physical states, being a defect or disorder of a structured system from an assignable range of normal functioning, it has only a metaphoric application to mental disorders. It is for this reason that in any reformulation of the insanity defence the term ‘disease of the mind’, or its cognate predicate ‘mental illness’, be jettisoned. The legal or moral question is whether or not mental responsibility was substantially impaired. Functional impairment is the feature characterising those we describe as insane. The nature of this dysfunction is best understood in terms of a defect in practical reasoning, or reasons we have for the things we do. Accordingly, any reformulation of the insanity defence should incorporate the following definitional element:
‘Mental Impairment’ is to be understood as a defect in practical reasoning, characterised by a distortion of natural functions and a diminution or loss of normal functions.
‘Natural’ denotes functions with either an organic or non-organic basis. ‘Normal’ is to be defined in terms of societal ‘prudential values’ as discussed above. In this way both the evaluative and normative aspects of psychiatry’s diagnostic categories are made explicit, without recourse to mechanistic explanations. Additionally, our moral intuition that agents have some capacity to act for reasons is clearly stated. The concept of ‘practical reasoning’ will be explored in the next section.
In its focus on symptoms, the medical model views the M’Naghten Rules formulation, ‘defect of reason’, as applying to the intellectual operations, or ‘cognitive functions’, of the mind. ‘Mental illness’ has been characterised as an extreme and prolonged inability to know and deal in a rational and autonomous way with oneself and one’s social and physical environment. ‘Autonomy’ is defined as ‘having and freely actualising a capacity for making one’s own choices, managing one’s own practical affairs and assuming responsibility for one’s own life’. ‘Autonomy’ will be examined in Part 5 in relation to ‘criminal responsibility’. In this Part, I will consider the meaning of the term ‘rationality’.
The first section will seek to outline the process of ‘practical reasoning’, which several authors have sought to employ in explicating the concept of rationality. The second will attempt to show the precise manner in which the concept is used in the context of mental impairment. Finally, I will turn to a critical examination of the concept of ‘rationality’.
The ability to reason has come to be associated with that of decision-making. This accords with the interdependence of the two abilities, their ‘mutual recursiveness’. This account is also integral to ‘folk psychology’, the view most individuals accept about mental life. Most people believe that their conscious feelings and judgments control their actions, and that these are readily available to them through introspection.
People use practical reasoning to try to achieve their goals in the actions they perform, and when they do this they display what has been termed, a ‘rationality of purpose’. This pattern of practical reasoning can be expressed as follows:
1. x desires q (for example, let it be the case that the wine be chilled).
2. x believes that doing p will produce q (for example, if I open the window, then the wine will be chilled).
3. x does p (for example, I open the window).
If in fact x does p because of a desire for q and in the belief that doing p will produce q, then we can say that p was an intentional action explained by x’s belief/desire set. When a person acts according to this type of practical syllogism, he or she causes an action, which is rationalised by his or her desires and beliefs. An explanation rationalises an act when it portrays that act as the rational thing to do given the actor’s beliefs and desires.
A concern of those supporting this analysis is to show that the legal conception of a person is the same as the moral one. Both conceptions are instantiated by the notion of the ‘person as practical reasoner’. Acts are explained in terms of personal agency, a person bringing about some state of affairs, rather than through purely causal accounts. What results is a theory of rational agency, serving legal and moral interests, framed in terms of the actor’s reasons for acting.
This theory seeks to attribute both logical and causal force to practical reasoning. The logical force is found in the wants represented in the first premise, in conjunction with the beliefs about the effective means to fulfilling those wants included in the second premise. These provide the reasoner with good reason (rationalisations) to want to perform the action indicated by the conclusion. The causal force is found in the agent’s wanting to achieve the end stated in premise 1 and he or she believing that the action identified in premise 2, will provide an effective means to that end. This belief will cause him or her to decide to act on that want or directive, represented by 3, as a means to satisfying the want stated in 1. The process represents a pattern of inference by which the individual reasons from his or her beliefs and desires for acting to an intentional act. The act is identified as an effective means, to the end represented, by his or her reasons for acting.
In essence, practical reasoning’s central premises are: (1) that whether a bodily movement is an action depends on what a person knows he or she can do and on what he or she knows him or herself to be doing; and (2) that reasons (belief/desire sets) can both cause and rationalise behaviour independently of other factors.
I will now turn to an examination of the implications that this analysis of action has for questions of mental impairment.
(i) The Nature of Mental Impairment
Affective disorders without substantial cognitive dysfunction will not ground a mental impairment defence. As such this enquiry will focus on the paradigmatic case of psychotic disorders such as schizophrenia. DSM-IV-TR defines the differential diagnosis of these disorders to include psychotic symptoms. Psychosis ‘refers to delusions, any prominent hallucinations, disorganised speech, or disorganised or catatonic behaviour’. Delusions are defined as ‘a false belief based on incorrect inference about external reality...[in the face of] incontrovertible proof or evidence to the contrary.’ An ‘hallucination’ is defined as ‘a sensory perception that has the compelling sense of reality of a true perception but that occurs without external stimulation of the relevant sensory organ’.
Inferences about disorganised thinking are based primarily on the individual’s speech. This may manifest as ‘loose associations’, slipping from one topic to another; ‘tangentiality’, where answers to questions are either obliquely related or completely unrelated; and in severe cases, disorganised ‘word salad’. Catatonic symptoms, however, are non-specific and may occur in other mental disorders.
Major disorders of thought processes involve disturbances in three areas: cognitive focus, reasoning and concept formation. Cognitive focus is the failure to effectively select relevant aspects of a stimuli field or to adjust attention in response to changing situations. Disturbed reasoning involves over-generalised thinking, the drawing of conclusions without evidence and combinative thinking, an inappropriate condensation of impressions and ideas into beliefs and conclusions. Concept formation disturbance involves ‘over-inclusiveness’: ‘the inability to conserve conceptual boundaries with the result that there is an incorporation of irrelevant ideas’. This failure to distinguish between false and veridical perceptual experiences and beliefs has been termed a failure of ‘reality testing’.
(ii) Rationality and Thought Disorder
Mental impairment has been viewed as constituting a failure of epistemological justification. This maintains that beliefs are justified if they are the product of correct cognitive processes, where the correctness of these foundations for beliefs is an inherent feature of the cognitive processes involved. Delusional thought processes will, in the absence of such justification, generate ‘false’ beliefs. It is the way a belief is formed and held, not its truth or falsity, which will determine its irrationality.
It will be noted that this criterion relates knowledge not to the content of an individual’s beliefs but rather to cognitive process. An effective practical reasoner is one who reasons, from wants and beliefs, to necessary and sufficient conditions for satisfying those wants. Such reasoning requires at least some unspecified degree of competence at recognising intuitive implication and inconsistency. In light of this, consider the position of an individual experiencing severe mental impairment.
A person experiencing lack of cognitive focus may attend to peripheral rather than central features of an event or situation, and thereby fail to consider implications, inconsistencies, or the significance of an act for his or her wants.
Impaired reasoning may result in an unwarranted interpretation of situations leading to an irrelevancy of action. Alternatively, it may lead to combinative thinking giving rise to not merely contradictory thoughts, but cognitive processes allowing inconsistency to guide the individual’s actions. The thought processes that allow the individual to hold inconsistent beliefs, prevents him or her from responding appropriately, to the immediate implications and inconsistencies in thinking about their conduct.
Impairment of concept formation involves two types of distorted thought. First, abstract concepts may have no meaning, or the individual may not be able to accurately apply those concepts to concrete events. Secondly, idiosyncratic interpretations and symbolism can lead one to vest events with an uncalled-for significance.
A failure of ‘reality testing’ involves both a failure to accurately perceive the external world and one’s relationship to it. This inability is not confined to hallucinations. It denotes a cognitive failure, which includes thought disorder leading to faulty reasoning and judgment, as well as purely perceptual errors (hallucinations). A person with schizophrenia is just as likely to draw erroneous conclusions from accurately perceived objective evidence, as he or she is to mistake the flux of his or her own consciousness, for such evidence. Further, he or she may rely on private sensations, with no objective correlative in the real world, as a guide to action.
This failure to distinguish between inter-subjective and intra-subjective experience and meaning is itself a ‘fundamental want of reason’. It amounts to more than a mere failure to reason well. It precludes reasoning itself.
The objects of the person with schizophrenia’s thought have such a radical privacy that they are not connected in an apparent and predictable way with public, external events or stimuli. Nor does the person with schizophrenia share with others the responses required for a shared language. It is the absence of these qualities that deprive the individual with schizophrenia’s speech, while psychotic, of language status. For it is language based upon shared meaning which grounds communication.
Hence, thought within such speech cannot be described as ‘rational’ or ‘reasonable’, for this requires communicability. Such a shared communicable thought presupposes the ability to test reality, and is the basis upon which logic proceeds.
In everyday life we do not reason in order to be logical, but are logical in order to achieve our goals. This would seem to suggest that practical reasoning somehow accords with our rational designs. Yet practical reasoning has been subjected to analysis showing that ‘normative theory and psychological facts pass each other by’. The correlation between practical reasoning and rationality will be considered in this section.
Effective practical reasoning requires at least three types of cognitive competency. First, the actor must be able to form accurate beliefs about the relevant wants and circumstances. Second, the actor must, through an associative process, select an act likely to promote his or her target want from amongst a set of other relevant wants and beliefs. Finally, the actor requires an accurate reasoning process that will allow him or her to draw warranted conclusions about the probable relationships among the various wants, acts and consequences.
The notion of rationality has no ideal definition. However, any working definition should be multi-level and incorporate the following:
• Intelligible desires and beliefs, that is, not wildly inconsistent or incoherent; • Beliefs and desires which are consistent with one another; • Beliefs and desires that reflect ‘transitivity’, that is, those that are intelligible and free from direct contradiction, logically cohere, and are implied by one another.
On this basis, to be ‘insane’ is to act on reasons that are not only unintelligible, but inconsistent and incoherent as well.
No one conception of rationality may be said to prevail over any other in support of the above. Even so, most advocates of practical reasoning view rationality as objectively ascertainable. For example, Moore writes that those who fail to act for rational ends ‘lack an essential attribute of being a person’.
Empirical evidence however, suggests that in fact people, due to cognitive limitations, are more likely to achieve goals by satisficing. That is, individuals find adequate solutions rather than optimise or maximise utility across all possible choices and outcomes. The search for norms of verification is illusory, as all that may be attained is an account of how people in fact behave.
The idea of rationality operates as an unqualified conditional. A person is either fully rational or only minimally so. The acquisition of desires and beliefs are judged independently of their role in practical reasoning when ascertaining the threshold question of minimal rationality. This however is to preclude the possibility of discriminating between psychotic and non-psychotic behaviour, which any criterion should be able to perform.
Focusing on cognitive processes, rather than content, we may say that even sane people entertain unreasonable or irrational beliefs. Faulty judgment, an everyday weakness, finds expression in an insane individual as delusional thinking. Yet delusional thinking need not be an indicator, on its own, of irrationality. For a person may suffer from an apparently delusional belief and maintain this, despite evidence to the contrary, yet not be considered to hold anything other than a false belief, not an irrational one. The belief may be maintained as well grounded where it is based upon idiosyncratic but countervailing ‘evidence’ provided by hallucinations, though these need not go together. The presence of delusions cannot be appealed to in any general way to demonstrate irrationality in insanity.
Practical reasoners deny the holding of metaphysical and religious beliefs as fit subjects for consideration with regards to questions of rationality. The orthodox belief in human immortality differs from, for example, an individual with schizophrenia’s messianic belief, only in respect of its widespread acceptance. The difference between the two lies not in the content of the beliefs, but in the way the beliefs are formed and held.
A reliance on expressions such as ‘widespread belief’, ‘widely shared’, and ‘reasonable view’ as criteria precluding analysis lacks intellectual rigour. This is to propose that rationality be confined to an arbitrarily defined area of ideas, and that only uncontested ideas and beliefs count. It is also to ignore the philosophic or religious roots of beliefs that people have about themselves and their environment. In this manner rationality is seen to be the prerogative of a majority.
The concept of rationality conceals a normative and moral judgment. It is the contents of the observed person’s motives and beliefs, which are comparatively assessed. Rationality is both socially and historically contingent. This applies equally to our formal modes of rationality, as well as to our conceptions of rationality. That is, we effectively determine what kind of thought and action will be rational, and in this sense constitute rationality as a mode of human activity.
Norrie has noted the way in which the reasoning legal subject, within a world of legal rights, protected by the rule of law, provides a mechanism of ‘ideological legitimation’. By positing a test of insanity on the basis of rationality, ‘[l]egal discourse ... decontextualis[es] ... social acts and individualis[es] social conduct’. This is to conceal the link between social context and behaviour and portray conduct as the result of exclusively individual, rather than social, pathology.
Irrationality, as the characteristic of mental impairment, accords with our intuitive notion of what it means to be insane. There must be a direct relationship between the rationality of one’s reasons and the ability to perceive and understand the nature or wrongfulness of what one is doing. It is these latter capacities, which ensure that our beliefs and desires are indeed rational.
However, a belief in an objectively ascertainable view of rationality is illusory. Empirical studies reveal that the way people do reach decisions has less to do with practical reasoning as with pragmatic decision-making. Further, the concept of rationality is seen to be socially and historically contingent, it being the content of a person’s motives and beliefs, which are scrutinised.
The argument that to be insane is to act on reasons, which are not only unintelligible, but also inconsistent and incoherent, denies clinical reality. For many psychotics have hallucinatory belief systems which are internally coherent.
If the reasons for behaviour are unintelligible, this should be sufficient to ground an insanity defence. Yet the concept should be limited to desires and beliefs inconsistent with those held by others in the individual’s societal group. In this fashion, incorrect but intelligible desires, such as jealously-motivated acts, or beliefs such that stealing while you are poor is justified, will not be deemed to be irrational. Moreover, such limitation will make clear that it is the person’s mental state, and not his or her behaviour, which is the object of the enquiry.
It is suggested that a reformulation of the insanity defence should incorporate the following definitional element:
‘Cognitive competency’, as indicative of practical reasoning, is measured by the extent to which the desires or beliefs, which motivated the offence, are unintelligible in light of what the individual’s society commonly desires and believes.
It will be noted that ‘cognitive competency’ is used in place of either ‘irrationality’, or Foucault’s ‘unreason’ (‘deraison’). I avoid ‘irrationality’ due to the breadth of pejorative connotations it carries, as well as the prevailing belief in its normative value. Foucault’s ‘unreason’
has been contrasted with ‘madness’ (‘folie’). The former represents an understanding of insanity as a form of illogicality or unreasonableness allied to the ordinary, imperfectly rational thought of sane people. Madness, on the other hand, is seen as a medical condition, and as such, alien and beyond comprehension. Yet insanity is not a status, as for example childhood is, but rather an excuse. It is not allied to ordinary rational thought, but is in fact the total absence of rational thought. While an insane person may have periods of remission, a child cannot be said to have intermittent periods of being ‘doli capax’. A particularly infelicitous expression, ‘unreason’, is best avoided.
I now turn to an examination of the concept of responsibility and its implications for the mental impairment defence. I begin with a consideration of two conflicting accounts of what grounds ascriptions of responsibility, choice and character. An examination of the nature of mental impairment, however, reveals that the conflict between the two theories is spurious. For what underlies criminality is neither character nor choice per se, but wrongful action, which manifests an incorrect attitude to the law. Finally, the interrelationship between cognition and volition will be analysed, and certain conclusions drawn.
Modern legal theory, while acknowledging the importance of physical actions, seeks to understand ascriptions of culpability in terms of what lies behind or produces the actions. In particular theorists view criminal liability as being founded on either ‘choice’ or ‘character’.  That is, what justifies conviction and punishment is viewed as being either the defendant’s wrongful choice, or some defect of character that the criminal conduct reveals. Both these notions of responsibility are instantiated in respective theories of excuse.
Representative of the ‘choice’ or ‘capacity’ theory of excuse is the work of H LA Hart. Hart holds that actors may be punished only where they had ‘the normal capacities, physical and mental, for abstaining from what [the law] forbids, and a fair opportunity to exercise these capacities’. The theory is but an instance of the more general principle of moral responsibility ‘that a person is not to be blamed for what he [sic] has done if he could not help doing it’. The justification for the choice theory offered by Hart is grounded in considerations of fairness and justice, constraints ‘which civilized moral thought places on the pursuit of the utilitarian goal by the demand that punishment should not be applied to the innocent’.
An actor is responsible where that actor’s choice is not made impossible by factors over which he or she has no control, that is, where the actor’s choice is a cause of his or her behaviour. Such ‘impossible’ factors are incapacity in the agent or the lack of fair opportunity to use a non-defective capacity.
As to why criminal liability depends on choice, Hart holds that the law is ‘a choosing system’ that aims ‘to guide individual’s choices as to behaviour’. This the law does by presenting him or her with reasons for exercising choice in the direction of obedience, but leaving them to choose. Such a system maximises ‘the efficacy of the individual’s informed and considered choice in determining the future’ by making his or her choice ‘one of the operative factors determining whether or not’ he or she is punished. Further, choice is viewed as the central condition of responsibility. Cardinal to Hart’s case for legal excuses is the belief that where excuses are absent, offenders act in accordance with their own values, attitudes and beliefs.
Character theorists argue that choice theorists’ contentions are defensible only on the basis of two essential premises: that we can separate the choosing-self from what motivates its choices, and that we can take a behavioural view of a person’s character traits. Such views, it is contended, do not accord with either our common sense, nor with current tenets of moral psychology.
First, the character theorist argues that the proper focus of the criminal law is not on the action of the defendant. Conviction is warranted given the validity of the inference we may make to an undesirable character trait. It is that character trait which the law condemns and punishes. Where there is an acquittal, given the absence of a fault element or some defence or excuse, the inference from act to character trait is not available. Choice is neither a necessary nor a sufficient condition of criminal liability.
Secondly, it is agreed that some of our character traits can constitute motivations to act, because of a particular set of values, feelings, desires, or aversions. Yet, a moral character trait may also refer to far more than a simple disposition to do the act required by some moral rule. It may also encompass dispositions to feel the appropriate emotions and moral motivations to do what the governing moral norm requires. The moral agent cannot simply make a rational choice, to experience the appropriate moral sentiments when he or she wishes to act morally. These moral emotions are the product of prior moral education, experience, and deliberation by the agent about what he or she should strive to be.
Objections to the character theory may be in large part due to the way it has been formulated by advocates. First, there is a supposition that the point of the criminal law and punishment is not directly to deter and punish harm-causing actions, but to deter and punish bad character, in and of itself. Yet this is to elevate character theory from a theory about moral culpability into a theory of criminal liability. The latter, however, must be underpinned by some version of the ‘harm principle’: liability is only predicated on harm caused. In the absence of any harm attributable to character per se, there can be no argument grounding liability on the basis of the formation or display of bad character.
Secondly, the assumption that culpability is attributable to a character trait renders the theory unable to explain liability for negligently caused harm, for example, the uncharacteristic ‘slip’. Even where the harm was intentionally inflicted, the theory would hold that this would go to the question of culpability, rather than, as is the case, towards the issue of sentencing.
Finally, the theory asks the criminal law to inquire into the accused’s character. That is, the character theory places a gloss on the law, asking the law to go beyond the fault element in ascriptions of culpability, and consider possible motives where no defence is offered.
Such criticism presents the character theorist with an opposite problem to that faced by the choice theorist. The latter’s account of criminal liability is seen to be too ‘thin’: there is more to being criminally
responsible than the making of choices. On the other hand, the character theorist, in finding his or her account on the moral significance of character, provides a far too rich account of criminal responsibility.
Not all who raise a defence of mental impairment fail to make a choice. In fact it may be the case that the accused may have acted differently had he or she so chosen. In response to this, a choice theorist may attempt to excuse mental impairment by arguing in one of two ways. First, that such responses are not really choices. The mentally impaired individual is deemed to be unable to engage in practical reasoning, or to conform his or her conduct to any judgments he or she might make.
Alternatively, the choice theorist may hold that mental disorder should excuse only where the agent could not have chosen to act otherwise than they did.
Yet, capacity to make the necessary choice, as a criterion of excuse, fails to take into account the fact that such capacity is a presupposition of moral responsibility and criminal liability, not the object of either. Further, to argue that liability rests on instrumental rationality, or practical reasoning, is not adequate. While some mental disorder manifests a lack or impairment of cognitive capacity, this is not the case in all instances.
Controversially, a person with an Anti-Social Personality Disorder (ASPD) may not appreciate the normative aspects of his or her actions. A person suffering severe depression may fail to have insight into the way in which his or her normative understanding is distorted by the pathological affective state. And a person suffering from a compulsive volitional disorder may have his or her ability to conform to conduct seen to be right effectively precluded. Any attempt to explain these kinds of incapacity must involve an appreciation of what is entailed in the concept of a person.
Self-conception relies on our ability to integrate the personal, impersonal, and non-personal points of view. That is, an integrated individual is able to look at themselves as an ‘I’, a ‘someone’, and a ‘something’, and express these perspectives in thought and action. This integration involves the ability to shift back and forth from one perspective to another, comparing and evaluating one’s beliefs about other people and one’s environment as well as criticising and correcting one’s beliefs and intentions to act in certain ways. This capacity to adopt and integrate these different perspectives is not merely a cognitive or intellectual capacity. Paralleling this development of logic is the development of cooperation and personal autonomy in the moral sphere.
As we have seen, it is a structuring principle of a person’s motivational system that beliefs and intentions are inter-subjectively supportable. Where such a system is closed, the idiosyncratic procedures forming the individual’s belief system are said to lack epistemic justification. Not subject to external or internal checks, the individual is unable to evaluate his or her beliefs, intentions or desires. Such an individual has lost his or her capacity to conform their behaviour to the law. The reasons for this incapacity stem either from the holding of false beliefs, incorrect rules of inference, or the maintaining of inconsistent beliefs. The agent’s conduct is therefore ascribed to the disorder, and the offender excused as not autonomous, as not having entered into the process of forming intentions and beliefs. A character theorist would contend that the person’s action does not manifest any character at all.
It could be argued however, that the criminal law is concerned not with character per se, but with criminal action which is sufficient for liability. Even acts ‘out of character’ result in liability. Criminal action is not merely evidence from which guilt, as some undesirable disposition, is to be inferred, but constitutes criminal guilt. This applies equally to the choice theorist’s claim that an action’s criminal character is determined purely by what the agent ‘chose’ to do.
This observation gives force to the argument that, in fact, the conflict between choice and character theorists rests, ultimately, on a spurious distinction between what a person ‘is’, and what a person ‘does’. As Silber has argued, ‘what a man does is a function of what he, in the context of his situation, is, and what he is within this context is revealed by what he does’. What makes a person criminally liable is not choice or character, but wrongful action, which, as the action of a responsible moral agent, manifests in and by itself some inappropriate attitude towards the law and the values the law protects.
It was noted above that a person with a compulsive disorder is precluded from controlling his or her behaviour. Such behaviour is referred to as ‘akrasia’ and is characterised as ‘acting against one’s better judgement’. The individual’s beliefs and desires all support one cause of action, and yet, the person persistently does another. In such cases, the agent cannot refrain from doing, or repeating, a particular action. Such cases of non-cognitive impairment are to be distinguished from those in which the agent’s desires coincide with his or her behaviour. In the latter instance, Hall argues, there is no real way of determining whether the actor ‘could not’, or ‘did not’, refrain from action. In such cases the attribution of responsibility is considered necessary as a socio-political premise.
Much of the problem relating to controversial non-cognitive impairments stems from confusion as to the relationship between volition and cognition. Traditional volitional standards have been misguided because they have sought a volitional basis for exculpation that stands separately from the cognitive one. Yet, the relationship is not that of two distinct processes. Volition is the exercise of the faculty, or function, by which one engages in conscious and intentional action, as a result of decision or choice, through deliberation. Volitional impairment involves some disruption of the practical reasoning process through which an actor selects a goal. Severe cognitive disturbance distorts this process of practical reasoning. Major cognitive psychopathology therefore, disrupts the volitional process. It does so by distorting the causal process through which the actor’s reasons for acting serve as structuring causes for his or her decision to perform the act he or she selects.
It follows that a cognitive interpretation of a mental impairment defence that addresses distortion of cognitive process, rather than content, provides an accurate account of both cognitive and volitional aspects of major psychopathology. A similar observation was made by Stephen, who defined self-control as ‘a power to attend to distant motives and general principles of conduct, and to connect them rationally with the particular act under consideration’. This link, between cognition and volition, allows Stephen to hold ‘that a man who by reason of mental disease is prevented from controlling his own conduct is not responsible for what he does’.
Stephen’s formulation makes clear that, on this interpretation, the crucial question becomes whether a mental disorder prevented the accused from controlling his or her actions. The fact that an accused person could control his or her behaviour, in the sense of being able to engage in practical reasoning, however, does not vitiate a defence of volitional insanity. Such would be the case where the derangement of an accused’s mind precludes his or her accessing the kind of moral information required for the development of a sense of self, as discussed above. In such an instance we would say that, while the cognitive component is functioning, the individual’s motivational-affective state is not.
It has been argued that while some mental disorder may be described in terms of a defect in instrumental rationality, this does not account for all cases. In particular, non-cognitive disorders, such as the compulsive’s inability to control his or her behaviour, where this does not accord with his or her desire, pose conceptual difficulties.
The difficulty can be seen to stem from a failure to note the interrelatedness between cognition and volition. Severe cognitive impairment impacts on volitional behaviour by distorting the causal process through which an actor selects a goal.
Cognitive ability to engage in practical reasoning is a necessary, not a sufficient, condition of volitional behaviour. For a sense of personal agency is predicated not only on an intellectual, but also on an inter-subjectively supportable, system of beliefs and intentions. This latter has been shown to stem from an integration of differing viewpoints.
In the light of the complexity inherent in human action, any conception of responsibility should take into account more than simply the actor’s cognitive competence. It should equally seek to comprehend the manner in which this cognitive element interrelates with the actor’s volition or ‘motivational-affective’, state. A person may distinguish between right and wrong and yet not be able to choose between them.
Such a conception would accord with the test stated by Dixon J in R v Porter: criminal responsibility will attach according to whether ‘D could not think rationally of the reasons which to ordinary people make his act morally right or wrong’. Yet in stating such a test, two refinements should be made. The expression ‘rational’ should be replaced by ‘cognitive competence’, and the enquiry should turn on whether the conduct amounted to an offence, not its morality. This is a purely legal question.
A reformulation of the insanity defence should include the following element in relation to the question of criminal responsibility:
A person is responsible for his or her actions where he or she had adequate cognitive competency to think of the reasons which people are expected to regard as sufficient grounds for refraining from commission of the offence.
Such reasons are: the prohibited nature of the conduct; the risk of punishment and the absence of offsetting advantage; and the ability to exercise choice.
Such a formulation, rooted in a principle of understanding, effects two immediate advantages. First, the ‘knowledge of wrongness’ rule is displaced by a test of competency for understanding the imperatives of
compliance. So where individuals are deluded into thinking that they are satisfying a divine command, or higher objective, even though they know
that the conduct is legally or morally forbidden, they will have a defence. For, such individuals may be said to lack the capacity to think competently about whether there is an overriding advantage in so acting. They may also be unable to think of their selves as capable of exercising choice.
Secondly, the issue is one of competency, not mental impairment. Understanding is thus not vitiated where the accused’s state was one of dissociation from extreme stress, or substance induced hallucinations. In such instances policy considerations become explicit in the law’s concern with public security. This concern would find expression in evaluations of ‘dangerousness’ and the need for preventive detention. This would be best handled under mental health legislation providing involuntary commitment and mandatory supervision.
It was noted in Part 2 (C) that insanity is a social judgment founded upon, but not precisely representing, a medical diagnosis. In establishing the defence it has been held that what constitutes a ‘disease
of the mind’ is a question of law for the judge. Whether the accused was insane is a question for the jury, aided by expert and other evidence. Usually, the expert evidence is that given by a psychiatrist. It is my contention that what results is a conceptual confusion stemming from law and psychiatry’s differing methodologies. It has been the analysis of this confusion, that has formed the subject of this paper. The submissions for a proposed reformulation have sought to make explicit what is often only implicit in the Rules.
Criminal responsibility has been reformulated in terms of a ‘principle of understanding’. A legal question, it operates as a threshold to the defence of insanity. It makes no reference to mental impairment but re-conceptualises the issue as one of competency to meet an expected standard of behaviour. Where such is not met the accused is deemed not responsible. Yet, in order to preclude the release of potentially dangerous individuals, the law is to engage the notion of ‘dangerousness’. In this fashion social defence concerns will be met.
In relation to the jury question, the law seeks the aid of experts. It is argued that it would be conducive to the transparency, though not necessarily surface clarity, of the law that psychiatry’s evaluative implications be made explicit. In this way psychiatric nosology may be shown to be both societally and historically contingent.
Morris has remarked that the M’Naghten Rules while ‘woolly, semantically confused, psychologically immature nonsense’ are ‘means whereby juries work rough justice in a difficult peripheral area of law and morality’. What has been attempted above is a restatement of the Rules in accordance with a clearer understanding of the foci of law and psychiatry. Accordingly it is proposed that a restatement of the insanity defence be legislatively enacted in these terms:
A person is not criminally responsible if at the time of the commission of the offence he or she had a mental impairment which included in its symptoms or consequences a loss of cognitive competency to think of the reasons which people are expected to regard as sufficient grounds for refraining from commission of the offence.
‘Mental Impairment’ is to be understood as a defect in practical reasoning, characterised by a distortion of natural functions and a diminution or loss of normal functions.
‘Cognitive competency’ as indicative of practical reasoning, is to be measured by the extent to which the desires or beliefs, which motivated the offence, are unintelligible in light of what the individual’s society commonly desires and believes.
‘Reasons’ include, the prohibited nature of the conduct; the risk of punishment and the absence of offsetting advantage; and the ability to exercise choice.
The history of the insanity defence could be recapitulated as the struggle between two different imperatives: social defence, the law’s concern, and scientific objectivity, purportedly measured by a notion of value-free truth existing in the domain of psychiatry. It is believed that the underlying methodological differences between law and psychiatry will be rendered, by being made explicit, more suited to the fulfilment of a single agenda. This agenda must include as a goal the explication of both principle and policy issues as they appear in forensic settings.
 John Austin, ‘A Plea for Excuses’ in Vincent C Chappell (ed), Ordinary Language (1964) at 46.
 Nicola Lacey, ‘Contingency, Coherence, and Conceptualism’ in RA (Antony) Duff (ed), Philosopy and the Criminal Law (1998); Alan Norrie, ‘Simulacra of Morality? Beyond the Ideal/Actual Antinomies of Criminal Justice’ in RA (Antony) Duff (ed), Philosophy and the Criminal Law (1998).
 Nicola Lacey, ‘A Clear Conception of Intention: Elusive or Illusory?’ (1993) 56 MLR 621 at 636.
 Id at 641.
 Vallance v The Queen  HCA 42; (1961) 108 CLR 56.
 R v Falconer  HCA 49; (1990) 171 CLR 30.
 Simon Bronnit & Bernadette McSherry, Principles of Criminal Law (2001) at 236–237.
 Hawkins v The Queen  HCA 28; (1994) 179 CLR 500.
  EngR 875; (1843) 4 St Tr (ns) 847.
 Finbarr McAuley & J Paul McCutcheon, Criminal Liability: A Grammar (2000) at 645, citing 1 Hale PC at 31–32.
 R v M’Naghten  EngR 875; (1843) 4 St Tr (ns) 847.
 McAuley & McCutcheon, above n10 at 646. n
 John C Smith, Smith and Hogan: Criminal Law (8th ed, 1996) at 202 (n81).
 R v Kemp  QB 399 at 406.
 R v Sullivan  AC 156.
 Glanville Williams, Textbook of Criminal Law (2nd ed, 1983) at 644.
 R v Kemp  1QB 399 at 406; R v Falconer  HCA 49; (1990) 171 CLR 30 at 48–49.
 R v Porter  HCA 1; (1933) 55 CLR 182 at 188.
 R v Bromage  1 Qd R 1 at 6.
 R v Porter  HCA 1; (1933) 55 CLR 182 at 189.
 Owen Dixon, ‘A Legacy of Hadfield, McNaghten and Maclean’ (1957) 31 ALJ 255 at 260.
 R v Kemp  QB 399.
 Bratty v Attorney-General for Northern Ireland  AC 386.
 R v Quick  EWCA Crim 1;  QB 910.
 R v Kemp  1 QB 399 at 407.
 McAuley & McCutcheon, above n10 at 683.
 Dixon, above n22 at 260.
 R v Radford (1985) 42 SASR 266 at 274 (approved R v Falconer  HCA 49; (1990) 171 CLR 30).
 R v Codere (1916) 12 Cr App R 21; Sodeman v The Queen  HCA 75; (1936) 55 CLR 192 at 215 (Dixon J); Willgloss v The Queen  HCA 5; (1960) 105 CLR 295.
 John W Cecil Turner, Kenny’s Outlines of Criminal Law (19th ed, 1966) at 76.
 Ranjit Murugason & Luke McNamara, Outline of Criminal Law (1997) at 270.
 R v Codere (1916) 12 Cr App R 21 at 27.
  HCA 56; (1952) 86 CLR 358 (approving R v Porter (1933) 55 CLR).
 Contra Windle v The Queen  2 QB 826; See Smith, above n14 at 209.
 Stapleton v The Queen  HCA 56; (1952) 86 CLR 358 at 367 citing R v Porter  HCA 1; (1933) 55 CLR 182 at 189 (Dixon J).
 R v Porter  HCA 1; (1933) 55 CLR 182.
 Sodeman v The Queen  HCA 75; (1936) 55 CLR 192.
 (1959) ALR 808.
 Id at 814.
  AC 432 at 449–450 (Lord Tucker).
 Bernadette McSherry, ‘Mental Impairment and Criminal Responsibility: Recent Australian Legislative Reforms’ (1999) 23 Crim LJ 135 at 142–144.
 R v Cottle  NZPoliceLawRp 16;  NZLR 999 at 1077 (Greeson P).
 Ryan v The Queen  HCA 2; (1967) 121 CLR 205 at 214 (Barwick CJ).
 Charles R Williams, ‘Development and Change in Insanity and Related Defences’  MelbULawRw 28; (2000) 24 MULR 711 at 717–718.
 Id at 719.
 Rabey v The Queen  2 SCR 513 at 524 (Dickson J); R v Falconer  HCA 49; (1990) 171 CLR 30.
 Re Wakefield (1958) 75 WN (NSW) 55.
 R v Cogdon  Res Jud 29.
 R v O’Connor (1980) 146 CLR 64; R v Meddings  VicRp 42;  VR 306.
 R v Quick  EWCA Crim 1;  QB 910; contrast R v Hennessy  EWCA Crim 1;  2 All ER 9 (hyperglycaemia).
 R v Tsigos [1964– 5] NSWR 1607; R v Radford (1985) 42 SASR 266.
 R v Sullivan  AC 156.
 Edward M Coles, ‘Scientific Support for the Legal Concept of Automatism’ (2000) 7 Psychiatry, Psychology and Law 33 at 45.
 McAuley & McCutcheon, above n10 at 638–644.
 Jennifer Radden, Madness and Reason (1985) at 28.
 R v Porter  HCA 1; (1933) 55 CLR 182 at 189 (Dixon J); R v Kemp  QB 399 at 407 (Devlin J).
 Ian G Campbell, Mental Disorder and Criminal Law in Australia and New Zealand (1988) at 128.
 American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders (DSM–IV–TR) (4th ed, text revision, 2000) at xxxi.
 McAuley & McCutcheon, above n10 at 640.
 R v Kemp  QB 399 at 407 (Devlin J).
 Michael Gelder, Dennis Gath & Richard Mayou, Oxford Textbook of Psychiatry (2nd ed, 1989) at 75, 96–98.
 Michael T Pathe & Paul E Mullen, ‘The Dangerousness of the DSM–III–R’ (1993) 1 JLM 47 at 51.
 Durham v United States  USCADC 293; 214 F 2d 862 (1954), 874.
 McAuley & McCutcheon, above n10 at 641–644.
 Stephen J Morse, ‘Excusing the Crazy: The Insanity Defense Reconsidered’ (1985) 58 Southern California LR 777 at 789; Compare, Michael Moore, Law and Psychiatry: Rethinking the Relationship (1984) at 362.
 See Radden, above n56 at 30–33.
 McAuley & McCutcheon, above n10 at 643.
 Bernadette McSherry, ‘Defining What is a “Disease of the Mind”: The Untenability of Current Legal Interpretations’ (1993) 1 JLM 76 at 82–89.
 Bratty v Attorney General for Northern Ireland  AC 386.
 R v Rabey (1977) 37 CCC (2d) 461; R v Hennessy  EWCA Crim 1;  1 WLR 287.
 R v Falconer  HCA 49; (1990) 171 CLR 30.
 Bernadette McSherry, Insanity, Automatism and Criminal Responsibility: Towards a New Defence of Mental Impairment (Unpublished Doctoral Thesis, 1996) at 44–45.
 McSherry, above n69 at 89.
 Campbell, above n58 at 129.
 Id at 132.
 J Margolis, ‘The Concept of Disease’ (1976) 1 Journal of Medicine and Philosophy 238.
 See BA Farrell, ‘Mental Illness: A Conceptual Analysis’ (1979) 9 Psychological Medicine 21 at 27.
 Compare DSM–IV–TR, above n59 at 297, 309.
 A Lewis, ‘Health as a Social Concept’ (1953) 4 British Journal of Sociology 109 at 122.
 Gelder, Gath & Mayou, above n62 at 106.
 Margolis, above n77 at 251.
 Lewis, above n80 at 124.
 Farrell, above n78 at 24.
 DSM–IV–TR, above n59 at xxxi.
 Id at xxx.
 RD Miller, ‘Hidden Agendas at the Law–Psychiatry Interface’ (1990) Journal of Psychiatry and Law 35 at 51.
 DSM–IV–TR, above n59 at 298–302:
Criterion A: Characteristic Signs and Positive: distortions in thought content (delusions)
Symptoms perception (hallucinations) language, and self monitoring of behaviour
Negative: affective flattening, alogia, and avolition
Criterion B: Marked social or occupational Dysfunction in one or more major areas of dysfunction functioning; eg, interpersonal relations, work or education, or self-care
Criterion C: Persistence of symptoms for at During the 6 month period, there must be at least
least 6 months 1 month of symptoms meeting Criterion A
Criterion D: Disturbance not better Not better accounted for by Schizoaffective accounted for by other Disorder Disorder or a Mood Disorder With Psychotic
Criterion E: Not due to other physiological Not due to the direct physiological effects of a effects substance or a general medical condition
 Farrell, above n78 at 27.
 Sacha Bem & Huib Looren de Jong, Theoretical Issues in Psychology (1997) at 24.
 Id at 17.
 RE Kendell, ‘The Concept of Disease and its Implications for Psychiatry’ (1975) 127 British Journal of Psychiatry 305 at 309.
 Moore, above n66 at 281.
 Gelder, Gath & Mayou, above n62 at 100.
 Thomas Szasz, Ideology and Insanity (1974) at 15.
 Thomas Szasz, The Myth of Mental Illness (1972) at 45–47, 85–86.
 Finbarr McAuley, Insanity, Psychiatry and Criminal Responsibility (1993) at 65–73; Radden, above n56 at 20–26; Michael Roth, ‘Schizophrenia and the Theories of Thomas Szasz’ (1976) 129 British Journal of Psychology 317.
 RM Veatch, ‘The Medical Model: Its Nature and Problems’ (1973) in Rem B Edwards (ed), Psychiatry and Ethics (1982) at 88–108.
 DSM–IV–TR, above n59 at 305.
 Radden, above n56 at 18.
 Moore, above n66 at 281.
 Radden, above n56 at 18.
 McAuley, above n97 at 68–73.
 Roth, above n97 at 321–322.
 McAuley, above n97 at 67.
 Id at 67–68.
 Kenneth WM Fulford, ‘Value, Action, Mental Illness, and the Law’ in Stephen Shute, John Gardner & Jeremy Horder (eds), Action and Value in Criminal Law (1993) at 279–310.
 Id at 290–291.
 Id at 298–299.
 Moore, above n66 at 244–245.
 Herbert Fingarette, ‘Insanity and Responsibility’ (1972) in Edwards, above n98 at 504.
 Rem B Edwards, ‘Mental Health as Rational Autonomy’ in Edwards, above n98 at 70.
 Philip N Johnson-Laird & Eldar Shafir, ‘The Interaction Between Reasoning and Decision Making: An Introduction’ (1993) 49 Cognition 1 at 2.
 Jonathan StBT Evans, David E Over & Ken I Manktelow, ‘Reasoning, Decision Making and Rationality’ (1993) 49 Cognition 165 at 169.
 Moore, above n66 at 13–14.
 Herbert Fingarette, The Meaning of Criminal Insanity (1972) at 196–197.
 Robert F Schopp, Automatism, Insanity and the Psychology of Criminal Responsibility (1991) at 118–119.
 Robert F Schopp, ‘Depression, the Insanity Defence, and Civil Commitment: Foundations in Autonomy and Responsibility’ (1989) 12 International Journal of Law and Psychiatry 81 at 98.
 DSM–IV–TR, above n59 at 297.
 Id at 821.
 Id at 823.
 Id at 300.
 Id at 301.
 Schopp, above n117 at 185.
 Gelder, Gath & Mayou, above n62 at 10–12.
 Schopp, above n117 at 186.
 Gelder, Gath & Mayou, above n62 at 306.
 Radden, above n56 at 78.
 Schopp, above n117 at 183.
 Id at 192.
 Id at 195–201.
 Compare Radden, above n56 at 66.
 McAuley, above n97 at 35–36.
 Radden, above n56 at 66.
 Evans, Over & Manktelow, above n114 at 170.
 Johnson-Laird & Shafir, above n113 at 3.
 Schopp, above n117 at 188.
 Morse, above n66 at 782.
 Christopher Slobogin, ‘A Rational Approach to Responsibility’ (1985) 83 Michigan LR 820 at 830–831.
 Jan Pols, ‘Mental Illness and Rationality: A Critique’ (1989) 12 International Journal of Law and Psychiatry 71 at 72.
 Moore, above n66 at 178.
 Evans, Over & Manktelow, above n114 at 169, 185.
 Anton Mooij, ‘Philosophy and Forensic Psychiatry’ (1986) 9 International Journal of Law and Psychiatry 117 at 121–122.
 Radden, above n56 at 73.
 Robert L Arrington, ‘Practical Reason, Responsibility and the Psychopath’ (1979) 9 Journal for the Theory of Social Behaviour 71 at 81.
 Radden, above n56 at 68–69.
 DSM–IV–TR, above n59 at 821.
 Radden, above n56 at 73; Morse, above n66 at 784; Moore, above n66 at 105.
 Pols, above n140 at 76.
 Marx W Wartofsky, ‘The Dialectics of Rationality: The Historicity of Norms in Psychiatry and the Law’ (1980) 3 International Journal of Law and Psychiatry 279 at 283.
 Alan Norrie, Crime, Reason and History (2nd ed, 2001) at 176.
 Id at 189.
 Moore, above n66 at 100–108.
 Michel Foucault, Madness and Civilization: A History of Insanity in the Age of Reason (translated from the French by Richard Howard) (2001) at 59.
 Radden, above n56 at 2.
 Ronnie D Mackay, Mental Condition Defences in the Criminal Law (1995) at 81–85.
 Antony Duff, ‘Choice, Character, and Criminal Liability’ (1993) 12 Law and Philosophy 345.
 Herbert LA Hart, Punishment and Responsibility (1968).
 Id at 152.
 Id at 174.
 Id at 80.
 Id at 44.
 Id at 46–47.
 Chin L Ten Crime,Guilt and Punishment (1987) at 100.
 Peter Arenella, ‘Character, Choice and Moral Agency’ in Ellen F Paul (ed), Crime, Culpability and Remedy (1990) at 78–79.
 Duff, above n158 at 364.
 Arenella, above n166 at 79.
 Jeremy Horder, ‘Criminal Culpability: The Possibility of a General Theory’ (1993) 12 Law and Philosophy 193 at 206.
 Samuel Stojlar, Moral and Legal Reasoning (1980) at 72.
 Horder, above n169 at 208.
 Michael Moore, ‘Causation and the Excuses’ (1985) Cal LR 1091 at 1137.
 Duff, above n158 at 355.
 Mackay, above n157 at 82.
 RA Duff, ‘Psychopathy and Moral Understanding’ (1977) 14 (3) American Philosophical Quarterly 189 at 198–199.
 DSM–IV–TR, above n59 at 350.
 Sir James Stephen, History of the Criminal Law of England: Volume 2 at 167; compare J Hall, ‘Mental Disease and Criminal Responsibility’ (1958) 33 Indiana LJ 212 at 223.
 RJ Lipkin, ‘Free Will, Responsibility and the Promise of Forensic Psychiatry’ (1990) 13 International Journal of Law and Psychiatry 331 at 349–355.
 Radden, above n56 at 141–146.
 Jeremy Horder, ‘Cognition, Emotion and Criminal Culpability’ (1990) 106 LQR 470 at 480.
 Duff, above n158 at 379.
 John R Silber, ‘Being and Doing: A study of Status Responsibility and Voluntary Responsibility’ (1967) 35 UChiLR 47 at 65 (emphasis in original).
 Duff, above n158 at 380.
 DSM–IV–TR, above n59 at 667.
 Radden, above n56 at 133.
 McAuley, above n97 at 50.
 Jerome Hall, ‘Mental Disease and Crime’ (1945) 45 Columbia LR 677 at 702–703.
 Pols, above n140 at 79.
 Schopp, above n117 at 202.
 Id at 203, 246.
 Stephen, above n177 at 170.
 Id at 167.
 McAuley & McCutcheon, above n10 at 663, 672.
 JM Livermore & PE Meehl, ‘The Virtues of M’Naghten’ (1967) 51 Minn LR 789 at 807.
 Id at 808.
 McAuley & McCutcheon, above n10 at 649.
  HCA 1; (1933) 55 CLR 182 at 189 approved Stapleton v R  HCA 56; (1952) 86 CLR 358.
 Brent Fisse, Howard’s Criminal Law (5th ed, 1990) at 432, 454, 463, 466–468.
Bernadette McSherry, ‘Criminal Detention of those with Mental Impairment’ (1999) 6 JLM 216 at 220.
 Norval Morris, ‘ “Wrong” in the M’Naghten Rules’ (1953) 16 Mod LR 435 at 437.