‘Irresistible impulse’: historicizing a judicial innovation in Australian insanity jurisprudence

Introduction

The historiography of the insanity defence classically pits lawyers against doctors, law against medicine. The M’Naghten Rules adopted in 1843 constituted a definitive judicial commitment to the availability of a defence of insanity, requiring the defendant to prove that at the time a crime was committed he or she was acting under the influence of a disorder of the mind that made him or her incapable of knowing the act was wrong. From the start, the provision was attacked by medical critics who saw it as embedding an archaic formula for determining the influence of an insane condition on the person committing a criminal act. Doctors thought that a person might know what they were doing was wrong but yet be unable to control their actions. The Rules seemed to exclude this possibility.

In England, and in common law systems within the Empire and later Commonwealth, the strict application of the Rules was qualified by the influence of juries as well as different interpretations by judges of what the Rules really meant. As shown by Wiener (2004), by the later nineteenth century insanity defences in England were rising significantly in spite of legal impediments. By the 1930s juries were favouring defendants pleading insanity in unprecedented numbers. In Australia and other jurisdictions the influence of psychiatry on opinion – judicial, bureaucratic and popular – helped drive up the successful deployment of insanity defence, and of unfitness to plead directions (Garton, 1986; Gibbons, Mulryan, McAleer and O’Connor, 1999; Shaw et al., 1992; Walker, 1968; Ward, 2002; Wiener, 2004). As Eigen (2003: 126) has argued, even in conditions which inclined towards judicial dismissal of medically-based insanity defences in the post-M’Naghten period, it was ‘the particular configuration of victim, offender and crime narrative that rendered the judge (and apparently the jury) either solicitous of the prisoner’s torment or contemptuous of an attempt to escape justice’. Contempuous indeed they could be – especially where medical witnesses argued in the mid-nineteenth century the case for ‘moral insanity’ or one of its ‘fellow-travellers’ (as Eigen puts it) such as irresistible impulse, a kind of explanation of criminal acts that many English judges in the later nineteenth century thought could be little distinguished from ‘unresisted’ impulse (Eigen, 2003; Smith, 1981; Wiener, 1990). In the words of an earlier historian of the role of medicine in criminal trials, ‘Alienists used “irresistible impulse” as a physiological term. In court it became a legal term and inevitably sounded like dangerous nonsense’ (Smith, 1981: 108).

Although medical critics saw themselves as the advance guard of scientific modernism in their critique of the M’Naghten Rules, the opposition between law and medicine was by no means one between two unified disciplines. Proverbially, doctors differed; so too did lawyers and judges. The history I want to explore here will touch little on doctors, and mostly on judges. It is a story about the gradual transformation of judicial opinion to accommodate changing understanding of human behaviour, only partly influenced by medicine and psychiatry. While it concerns the development of judicial reasoning, it does not do so with respect to evaluations of how ‘correct’ particular judicial statements of the law may be at any time (cf. Williams, 2000). Rather it seeks to explore the interaction of a range of factors including judicial reasoning, jurisdictional competition, social policy and emergent disciplinary knowledge in shaping a broader conceptualization of the influence of mental illness on behaviour.

The story discussed here has its base in Australia, but it is important to stress that jurisdictional lines do not constrain the development of medico-legal approaches of the kinds I am addressing. The networks of decision-making and debate that I address here centre on three Australians, two of them judges who ‘made very important contributions to the learning on these Rules on the Bench and in their extra-curial writings’ (Waller, 1977: 183). One of them, Sir Owen Dixon, Chief Justice of the High Court of Australia from 1952 to 1964, was a judge of that court from 1929. He was widely acknowledged in his time and later as one of the leading common lawyers of the twentieth century, and his opinions were known, cited and debated in Britain and the USA. He was especially close to Justice Felix Frankfurter of the US Supreme Court – and often in considerable tension with the law lords of the Privy Council (the highest court of appeal for Australian jurisdictions until as late as 1986) over the matters to be considered in this paper in particular (Ayres, 2003). A second figure, Sir John Barry, a judge of the Supreme Court in the state of Victoria from 1947 to 1969 was also a leading criminologist of his day, well known in the USA, Britain and Europe. A third was Barry’s close friend, Norval Morris, a professor from 1964 in the Chicago Law School, where he was founding director of the Centre for Criminal Justice. Morris’s early career at the University of Melbourne helped to establish him as an original and iconoclastic thinker, especially on issues involving insanity and the law. The intellectual networks of such people had frequent cause to debate the law, insanity, punishment and treatment in the years they flourished, the 1950s and 1960s (Finnane, 2006, 2007). Sanford Kadish for example, later a Dean of Law at University of California at Berkeley, had spent a year in the Melbourne Law School in 1955. In 1968 his lecture delivered at Cambridge, England, on the ‘decline of innocence’ challenged vigorously the treatment paradigm being developed as an alternative to punishment by the British social scientist Barbara Wootton, another visitor to Melbourne in the early 1960s, and a warm friend of Sir John Barry (Edwards, 1961; Kadish, 1968; Wootton, Chambers and Seal, 1959). Behind the sober judgments delivered in the criminal courts and published in the law reports lay a constant murmur of debate over the approaches the law should adopt in trying and disposing of violent offenders.

Irresistible impulse: the Stephen/Dixon/Barry line

‘If the hangman is unemployed, why do we want this defence? Is not fitness to plead, sensibly handled, enough?’ So asked Norval Morris of his judicial correspondent Justice JV Barry of the Victorian Supreme Court in 1969. In R v Weise Barry had delivered judgment in an appeal brought by a man found guilty of murdering his wife. ¹ On appeal, the court found the man not guilty on the ground of insanity, quashed the conviction, and ordered him to be detained at the Governor’s pleasure, under a statutory provision available in the Victorian Crimes Act. As he frequently did in correspondence with legal academic colleagues, Barry sent his judgment to Morris for comment, wondering whether he (Barry) had ‘done anything to elucidate “know” in the M’Naghten Rules’.

Barry’s friendship with Morris, lasting nearly two decades, was grounded in frankness when needed. Morris’s rhetorical question, about the utility of the insanity defence in a jurisdiction which had effectively abandoned the death penalty, was thus in character. So too were his observations on Barry’s attempts to elucidate ‘know’. This was an exercise Morris had long fought in his writing. ‘There is no satisfactory referent, medical or legal, for “mental disease” in the M’N Rules’, he suggested to Barry. ‘If knowledge of wrongness be given the Stephen-Dixon-Barry expanded inter- pretation, I find it hard to think of more than a few exceptional murderers who would not have the defence’. Morris’s suggestion that the insanity defence had become no more than an intellectual game was a proposition that Barry challenged as academic in nature, not grappling with the context in which judges had to make decisions. ‘I agree with your criticism of the Stephen-Dixon-Barry approach,’ replied Barry, ‘but it must be remembered that all three were constrained by the limits of the judicial function. Unlike the carefree academic, we must deal with the law as it is and not as it ought to be, and inch forward slowly’. In R v Weise he felt that he made a small advance by proposing that ‘to “know” means to know sanely and not in the distorted or confused or unreal fashion that a psychotic knows’. ²

Underlying this correspondence was a long judicial history of struggle over narrow or extended definitions of what it was to know something. In the jurisdictional context that Barry was addressing (Australian States, subject to appellate review in the High Court and, until 1986, the Privy Council), he took a stand in Weise on the side of those who wanted to recognize the relevance of a concept of ‘diminished consciousness’. He pitted himself firmly against the view that Chief Justice Herron had put in a NSW case (R v Jenkins) in 1963, that ‘a man is in law, either perceptive or appreciative or is not’. For Herron ‘the law interprets “know” as used in the M’Naghten Rules in a rigid and absolute way’. ³ Such judicial certainty as Herron expressed did not have to await Barry’s judgment in another jurisdiction for correction. Rather, the polarity represented here reflected in microcosm the oppositions in judicial interpretation that characterized the M’Naghten legacy. Along the spectrum that was constituted by ‘a rigid and absolute way’ of knowing and a ‘sane’ way of knowing, judges distributed themselves as they sought to deal with the variety of cases that presented to a court. The narrower view had prevailed in England but had long been shaken in Australia by the influence of Sir Owen Dixon.

‘Knowing what is wrong’

When Norval Morris referred to the Stephen-Dixon-Barry line, what did he mean? What kind of approach to the relations between insanity and criminal responsibility did this line represent? How was this line distinguished from other judicial interpretations? What bearing did this line have on the discourse lurking at the back of the courtroom that aspired to know what insanity was?

There were two elements to the M’Naghten Rules: knowing the ‘nature and quality’ of the act one had done, and knowing that it was wrong. The Rules sought to assess whether the state of mind of an accused had been such as to enable them to know both these things. Both elements had to be satisfied in order for a defence of insanity to be proved as grounds for a verdict of not guilty (prior to 1883 in England, after which the verdict was ‘guilty but insane’). It is also well known that there was major debate around each element, at the time and since. In spite of the debate, the Rules have proved remarkably durable: reviewing the available defences to a charge of homicide in 2004, for example, the Law Reform Commission of Victoria found overwhelming support among legal and medical professionals for retaining the Rules. Here I do not want to discuss what was meant by the nature and quality of an act, but what was meant by knowing that the act was wrong.

When Barry J said in 1969 that the defendant must have known a thing sanely, he placed himself in a line that went back to James Fitzjames Stephen. In A History of Criminal Law in 1883, Stephen had captured the difficulty of the Rules precisely.

The word ‘wrong’ is ambiguous as well as the word ‘know’. It may mean either ‘illegal’ or ‘morally wrong’, for there may be such a thing as illegality not involving moral guilt, and when we come to deal with madness, the question whether ‘wrong’ means ‘morally wrong’, or only ‘illegal’, may be important. (Stephen, 1883: 167)

The example he went on to develop was that of Hadfield, who had attempted to assassinate George III, and whose ‘knowledge of the illegality of his act was the very reason why he did it’. Yet was there any doubt that Hadfield held the kind of delusion that prevented him from exercising ‘an act of calm judgment in the character of the act’? Just because Hadfield knew an act was illegal, could it really be said that he knew such an act was wrong?

For Stephen there was no doubt about the meaning the word ‘wrong’ in the context of a criminal law charge. ‘There is no offence’, he went on, ‘in answer to a charge of which madness is likely to be set up as a defence, as to the moral character of which any question can arise … A person who disbelieved in all moral distinctions, and had ridded himself of all conscience, would know that murder is wrong.’ Given this natural law understanding, then the key question became one relating to the capacity of insane persons to control their conduct. A narrow interpretation of the M’Naghten Rules would limit the insanity defence to a question of madness ‘regarded merely as a possible cause of innocent mistakes as to matter of fact and matters of common knowledge’. Wedding himself to what he characterized as a wider interpretation, Stephen became the advocate of a law ‘that no act is a crime if the person who does it is at the time when it is done prevented either by defective mental power or by any disease affecting his mind from controlling his own conduct’ (Stephen, 1883: 167–8). The attempt was fruitless, though he considered that judicial interpretation of the Rules on the wider lines he had suggested might avoid the need for the kind of legislative provision that he drew up in 1878. ‘Knowledge that an act is wrong’ might be the best test of responsibility if the words were construed within the understanding he posited, namely that ‘it is as true that a man who cannot control himself does not know the nature of his acts as that a man who does not know the nature of his acts is incapable of self-control’ (Stephen, 1883: 171).

Although Stephen’s views on these matters struggled to find acceptance among the judges of his own country (for many of whom the role of the law was to insist on the restraint of impulse, not to find excuses for failure to control passions ⁴ ), their influence on others was enduring. By way of the imperial circuitry of appeal courts in the old Empire and the succeeding Commonwealth, the broader view eventually worked its way back into the law of England. To see how this happened, we must consider the judicial opinions and extra-judicial writing of Owen Dixon of the High Court of Australia.

Dixon and ‘the impossible legacy’

For Dixon, the English laws that Australia inherited and had to work with were profoundly unhelpful in the matter of insanity and crime. The legacy that he discussed in a 1957 lecture on this ‘discreditable chapter of the law’ was one which had ‘gone astray’ in three respects in the nineteenth century. The first step, the decision taken after the Hadfield case, to add a rider to a verdict of acquittal specifying that it was on the grounds of insanity, confused the functions of criminal law and lunacy administration. Second, the M’Naghten Rules had imprisoned the ‘common law in a formula … [and] had deprived the common law not only of its capacity for development, but even of its accustomed flexibility of application’. In the third stage, the trial of Maclean for the attempted assassination of Queen Victoria had led to the Trial of Lunatics Act of 1883, with its seemingly contradictory requirement of the verdict ‘guilty but insane’ (Dixon, 1965: 224) . As Dixon (1965: 218) commented on this last development, ‘a prisoner who at his trial hears a verdict of guilty but insane pronounced may be forgiven if he fails to recognise that he has been acquitted’.

Dixon was much attached to what he called ‘basal principle’ in common law, and he is considered to be the judicial apostle of what he called ‘strict legalism’, a standard that might even in his own case be considered as an ideal type rather than an adequate accounting for the scope of his judicial reasoning. His own judicial practice, at least in respect of the way he dealt with the defence of insanity, seemed to call into question his gloomy view of the prison created by the M’Naghten Rules. In truth he played an important role in interpretation of the Rules in a way that seemed to demonstrate their capacity for development and even flexibility of application. ⁵ That his judgments also occasionally ran into the brick wall of the Privy Council does not disprove the cogency of this view of his role. Indeed, the degree to which Dixon persisted in a line of thinking that eventually led him to a decisive rejection of Privy Council authority was, I suggest, consistent with his readiness to develop the law under circumstances where he felt its flexibility was warranted and not inconsistent with common law principle. When Norval Morris spoke of the Stephen-Dixon-Barry interpretation (see Note 2), it was precisely such a flexibility and development that he identified, even if it was along a line that he felt should now be abandoned.

Tension between the direction of the High Court in Australia and what the Privy Council was prepared to concede was highlighted in 1959 in a case that attracted a good deal of attention behind the scenes. In Attorney-General(SA) v Brown the Privy Council overruled the High Court’s quashing of a criminal trial verdict, refusing to admit that a defence of irresistible impulse, in the absence of medical evidence on the question, might constitute a symptom of insanity sufficient to exculpate a defendant on a charge of murder. In March 1959 John Whelan Brown was found guilty of the murder of Neville Lord, a young manager of a sheep station in South Australia. There was no motive for the crime, which occurred when Brown inexplicably took a gun he found lying in his room and shot Lord in the bedroom of his home. Brown admitted to the police that he knew what he was doing and that it was wrong, but in a critical piece of evidence added ‘but I couldn’t help myself’. Medical opinion was divided on the question of possible schizophrenia, and the medical witness for the defence advanced an explanation of Brown being in a state of ‘simple schizophrenia’ at the time of the shooting, so that he did not know that what he was doing was wrong.

An appeal to the South Australian Supreme Court was unsuccessful, but in the High Court Brown won a reprieve. The core of the decision, written by Dixon, was that the trial judge had erred when he told the jury, in summing up, that if they felt Brown had been acting under an uncontrollable impulse, this was no defence in law and they should find him guilty. Dixon rebutted this reasoning: ‘to treat his domination by an uncontrollable impulse as reason for a conclusion against his defence of insanity is quite erroneous. On the contrary it 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 what he was doing was wrong.’ ⁶ Dixon had long maintained that mind was indivisible and that the M’Naghten Rules might be interpreted to allow the possibility that uncontrollable impulse might affect a capacity to ‘comprehend the wrongness of an act’. When the High Court quashed the conviction and ordered a new trial, the Crown took the case on appeal to the Privy Council. There the High Court decision was overturned, with the Privy Council refusing to accept what its judgment represented as Dixon’s inference that (with or without evidence) irresistible impulse must be recognized as a symptom of insanity under the M’Naghten Rules. ⁷

The decision was greeted with disdain by Morris, who wrote to Barry shortly after reading reports of the Privy Council judgment: ‘It looks as if Dixon’s excellent work over the years on this topic has been frustrated.’ Having re-read the High Court’s decision he found it hard to believe the Privy Council’s reversal: ‘I hear that they dug out the very oldest brandy at the Adelaide Club when this news came through’, he added, a reference explained by Morris’s contempt for the conservative judicial and legal culture in this city where he was Dean of the only Law Faculty. ⁸ Morris was so exercised by the judgment that he determined to include a lengthy discussion of it in a forthcoming US publication to which he was contributing an article on insanity defences in Australia (Morris, 1961). ‘Nowhere in its long judgment’, he concluded after working on this, ‘does the Judicial Committee [of the Privy Council] come to grips with the main reason for the High Court’s ordering a retrial, namely, that the High Court interpreted the trial judge as having directed the jury that if they found that the accused was labouring under an irresistible impulse at the time of the killing then they should find him guilty.’ Yet there was some comfort in the emergence in the judgment of an agreement that ‘irresistible impulse and M’Naghten insanity can co-exist’. ⁹ This con- solation found its way a few years later into a more measured assessment of the outcome in his Studies in Criminal Law, co-authored with Colin Howard. In this book the presentation of the outcome in Brown welcomed the clarification of the role of medical evidence in such trials (Morris and Howard, 1964: 51).

Barry, a judge of the Victorian Supreme Court, and therefore someone who had the responsibility to conduct such criminal trials, found the judgment equally unsatisfactory. Apart from its ‘turgid, repetitious style [which] is usually a sign of a confused mind’ and the ‘lame’ justification for granting special leave to appeal against a High Court judgment, the language was ‘very loose’. In any case, Barry considered that the concept ‘irresistible impulse’ was not very helpful. He disliked the phrase, he told Morris, and looked to other language in psychiatry to address the need – for example, [Fredric] Wertham’s notion of ‘catathymic crises’, in which an overwhelming derangement of mental processes meant that the ‘subject cannot in any rational sense be said to know what he is doing or to have any faculty of restraint’. ¹⁰ It was this kind of reasoning that led Barry in Weise to reaffirm the importance of a ‘sane’ kind of knowing, one that relied on an authoritative charge to the jury by Dixon in 1934 in Porter, which I discuss below.

Dixon’s own reaction to the Privy Council’s reversal of his High Court judgment is not clear but may be guessed. By this time he was already impatient with the Privy Council’s intrusion on Australian jurisdiction; in discussions with the Australian Prime Minister shortly before the Privy Council’s judgment, he ‘favoured entrusting the High Court with the power to give leave as a condition precedent’ to Privy Council appeals from Australian courts. He was encouraged in his opinion by Felix Frankfurter of the US Supreme Court, who thought the Judicial Committee’s decision in the Brown case poor, though only partly because he thought the Australian judiciary should have the last say in such a matter. More significantly, the High Court’s Brown judgment, rightly attributed by Frankfurter to Dixon, was one which he considered showed ‘due regard for more modern psychiatric thinking without jettisoning the rule in M’Naghten’s case’. Frankfurter thought it in line with Dixon’s ‘other applications of M’Naghten’s Rule in your excellent charge in R v Porter, and your opinion in Sodeman v the King’ (Ayres, 2003: 272–3).

Dixon leaned towards more modern thinking, as Frankfurter put it, but his judgments on these matters did not open the door to any unalloyed dependence on medical opinion. Rather, he had expanded the concept of knowing what was right or wrong to include something broader than the generally narrow constructions that prevailed in the English appeal courts (e.g. R v Windle 1952), while also refining to a high degree the legal understanding of what it meant to impute intention to an action. To see these shifts, and the way in which they allowed medical evidence to find a way into the court, it is necessary to go back to Dixon’s earlier judgments, especially the definitive Porter (1933), Sodeman (1936) and Stapleton (1952).

Dixon and M’Naghten

The distinctiveness of the Australian approach was centred on Dixon’s reasoning in three cases. We have already noted his later (1957) view of the M’Naghten Rules which he considered had imprisoned the common law in a formula, blocking its capacity for development. Some indication of the way in which he might approach an insanity defence early in his career in the High Court was evident in his participation in a 1932 discussion on the plea of insanity. At one of the early meetings of the Medico-Legal Society, a Melbourne psychiatrist Reg Ellery ¹¹ presented a paper on the plea of insanity, arguing for a wholesale reform of the law’s approach. Ellery (1932) sought the replacement of retributive punishments and selective exculpation of offenders able to mount an insanity defence by a system of treatment under a regime of compulsory psychiatric examination of all offenders.

In discussion, Owen Dixon, who had been a High Court judge for three years, was the first to comment and did so at some length. He avoided the policy issues raised by Ellery’s proposal, choosing instead to comment on the resistance to adaptation of the law that had resulted from the M’Naghten heritage, as it had been interpreted by the English Court of Criminal Appeal. In ways that he would follow in some of the judgments discussed below, Dixon argued that the flexibility of the common law had been reduced by the M’Naghten Rules to a ‘formula which has proved incapable of adaptation to widening knowledge and changed conceptions of mental phenomena’. In a brief account of the judicial history of the Rules, he went on to argue that ‘modern Judges’ had in fact not initially treated the Rules as such a formula but that:

the legal principle [of the insanity defence] retained sufficient vitality of its own to enable them to apply it to conduct which the prisoner was without any capacity to control or direct. A formidable number of Judges, sitting alone, had given considered rulings which departed from the Macnaughton [sic] formula. But unfortunately, the English Court of Criminal Appeal in 1926 finally declared that such departures were inadmissible. ¹² (Ellery, 1932: 43)

To emphasize the point, Dixon suggested that, far from lagging behind medical thought, lawyers before M’Naghten had been in advance of community sentiment. It was the English Court of Criminal Appeal which had restrained recent ‘judicial tendencies towards a more rational and liberal rule’, by an approach ‘which rightly or wrongly considered the Macnaughton [sic] formula remained conclusive’ (Ellery, 1932: 43). Not only are these comments valuable for suggesting the consistency of Dixon’s views that would be developed at greater length in 1957. More importantly and interestingly, Dixon’s approach in the cases which he adjudicated over the next few years suggest a determination to continue the ‘more rational and liberal’ line of interpretation that had been followed by many judges in England prior to the constraints embraced by the Court of Appeal in the 1920s.

In the first of these, Dixon was in the unusual role, for him, of being the trial judge. Although a judge of the High Court, he was sitting in the Court’s original jurisdiction (under which the Court still heard criminal matters in the Australian Capital Territory) in a matter involving the trial of a man for murdering his 11-month-old son. After a period of protracted conflict with his wife, including a period of separation, Bertram Porter had administered the poison strychnine to his son, of whom he had care, after threatening he would kill himself and the infant. Porter entered a defence on the grounds of insanity, and medical evidence was heard. In directing the jury, Dixon set out the matters with a clarity that was of enduring influence, the direction still being reprinted in criminal law casebooks in the USA as well as Australia (Bronitt and McSherry, 2001; Fisse and Howard, 1990; Kadish and Schulhofer, 2001). First, Dixon drew the jury’s attention to the rationale for the insanity defence: what was the ‘utility of punishing people if they be beyond the control of the law for reasons of mental health?’ ¹³ He emphasized that the defence was nevertheless to be contained to those who were something other than simply abnormal, of ‘peculiar’ disposition or ‘peculiarly tempered’. The deterrent value of the criminal law, Dixon stressed, should not be undermined by a loosening of the bonds of responsibility. For this reason it was essential to be clear about what was the ‘standard of mental disorder’ to be applied by a jury in assessing an insanity defence; Dixon distinguished this from the questions that the medical profession faced, or even those charged with the administration of lunacy.

In developing the standard to be considered by the jury, Dixon went on to outline the standard requirements of the M’Naghten Rules. The jury was required to make a judgment about the condition of the mind at the time of the act; and for an insanity defence to be proved, the ‘state of mind must have been one of disease, disorder or disturbance’, one that affected the understanding, rather than requiring a demonstrable change in the brain or ‘physical constitution of the mind’. It was, however, in elaborating two key words in the M’Naghten Rules that Dixon sought to refine the interpretation in ways that would shape later views. In assessing whether the defendant ‘knew that what he was doing was wrong’, the jury had to consider not ‘right or wrong in the abstract’, but the meanings of these terms in the specific circumstances of the killing of an individual. If the man could not ‘reason about the matter with a moderate degree of sense and composure it may be said that he could not know that what he was doing was wrong’. Further, what was meant by ‘wrong’? This was wrong ‘having regard to the everyday standards of reasonable people’. If the ‘man whom you are trying had such a mental disorder or diseased intelligence at that moment [of administering the strychnine] that he was disabled from knowing that it was a wrong act to commit in the sense that ordinary reasonable men understand right and wrong and that he was disabled from considering with some degree of composure and reason what he was doing and its wrongness,’ the jury should find him not guilty on the grounds of insanity.

The relative clarity of Dixon’s direction no doubt lent it the kind of authority that resulted in it becoming the standard reference point for directions to the jury in insanity defence cases. Barry played a key role in giving the direction wider circulation: the charge to the jury was published in 1935 in the first volume of Proceedings of the Medico-Legal Society of Victoria, under Barry’s editorship ([Dixon], 1935); and the following year he drew it to the attention of defence counsel in the Sodeman case (Bourke and Sonenberg, 1969: 80; Waller, 1977: 178). Via that route the direction entered the law books more generally. In 1943, while not yet a judge, Barry noted its significance as a ‘model for judges who have to perform the task … of directing juries intelligently upon the defence of insanity’ (Barry, 1943). Subsequently, as a judge, he readily turned to Porter to guide his own practice. One of his early unreported judgments was often later cited, along with Porter, in support of the proposition that ‘disease of the mind’ was not limited to physical evidence of deterioration but must stop short of being a ‘mere idiosyncrasy, such as bad temper’. ¹⁴

The grounding of Dixon’s understanding of ‘wrong’ in a way that linked it more to Stephen’s than to narrower Privy Council interpretations became clearer in two cases which were heard on appeal in the High Court of Australia, one of them proceeding also to the Privy Council. In the first of these, the case of Arnold Karl Sodeman, the defendant was found guilty of the murder of a young girl, a killing that was the last of a series to which he confessed. Medical evidence about Sodeman’s mental instability was very strong, but the jury found him guilty and he was sentenced to death, as was required under Victorian law at the time. On appeal to the High Court, Sodeman was unsuccessful, with the court of four judges equally divided. The care which Dixon had exercised in Porter found effect in the judgments delivered by the two judges, Dixon and Evatt, who favoured a new trial for Sodeman. Both judges found that the trial judge had misdirected the jury, and Evatt cited Dixon’s Porter judgment as a model of the kind of ‘clear, accurate and elaborate summing up’ that judges should give when directing juries required to consider the abstractions of the M’Naghten Rules. Dixon did not cite himself, but his detailed reflection in Porter on what was necessary in considering such insanity cases again informed his shaping of the considerations of which contemporary juries needed to take account. Both judges found reason to criticize the trial judge’s dismissal of the possible relevance of ‘irresistible impulse’. Each indicated his readiness to set the High Court on a path that would differ from prevailing judicial authority in England. Dixon phrased his view of this possibility very carefully, but Evatt was blunt: not only was the Australian High Court not necessarily bound, ‘even by decisions of the Court of Appeal [in England]’, but ‘it would be unsatisfactory if the common law of England … must be regarded as forever unable to adjust its rules to modern medical knowledge and science’. Both considered that evidence (of the kind that had been led by medical experts in Sodeman’s case) that in effect described something called ‘irresistible’ or ‘uncontrollable’ impulse might afford (in Dixon’s words) ‘the strongest reason for supposing that he is incapable of forming a judgment that they [the defendant’s actions] are wrong’. For Evatt, the trial judge had been misleading in emphasizing ‘an antithesis between diseases of the mind leading to irresistible impulse and the conditions described in M’Naughten’s Case’, for the latter might not only accompany, but even be inferred from, a disease of the mind producing an ‘“irresistible impulse”’. Both Evatt and Dixon were inclined to take seriously the contribution of changing scientific ideas, Dixon being one of the founding members and active participants in Melbourne’s Medico-Legal Society (Dixon, 1937; Lunney, 2005). Evatt’s views were unabashedly clear when he came to sum up the gap between what he observed happening in the trial of Sodeman and what it was necessary for courts to note. As he concluded ‘it is quite out of accord with modern research in psychology to assert an absolute gap between cognition and conation’. ¹⁵

Sodeman was unfortunate in the High Court for the absence of a fifth judge meant the divided court judgment must result in the failure of his appeal. On further appeal, the Privy Council turned down the attempt to embrace the doctrine of irresistible impulse, principally on the grounds that accepting otherwise would lead to ‘different standards of law in England and in the Dominions’. The Privy Council judgment appeared in the All England Reports where it was noted that ‘this case raises the question whether the rules in M’Naghten’s case are quite consistent with modern medical theory’. ¹⁶ This emphasis was more relevant to Evatt’s statement of the matter than to Dixon’s, for whom the logic of reasoning in the interpretation of what ‘wrong’ and ‘know’ mean was perhaps more central.

In Sodeman the Privy Council thus explicitly rejected an attempt to expand the M’Naghten Rules to include irresistible impulse to do an act, even where a person knew the act was wrong. The loosening of the Rules by way of adumbrating what ‘know’ and ‘wrong’ meant was in practice perhaps encouraged by the diffusion of Dixon’s charge to the jury in Porter. The distance between Australian and English authority lengthened in the early 1950s when once again Dixon delivered a judgment that broadened the judicial reading of what was involved in an insanity defence. This time the judgment entailed a very specific rejection of English case law. In R v Windle, the English Court of Criminal Appeal in 1952 rejected an appeal by a man whose insanity defence to a charge of murdering his wife had failed when Justice Devlin ruled that there was no issue to be put to the jury. In what Walker (1968: 114–15) described as the pendulum reaching its furthest point on the issue of legally or morally wrong, the Lord Chief Justice decided that ‘wrong’ in the M’Naghten Rules meant ‘contrary to law’, i.e. ‘illegal’.

In the same year the Australian High Court in the case of Stapleton had its opportunity to dismiss an approach that was now a very long way from Australian judicial thinking. In the Northern Territory a man had shot and killed a policeman in the course of a manic outburst following domestic argument with a female partner. On trial before Justice Kriewaldt he was convicted and sentenced to death. On appeal his conviction was quashed by the High Court, and a new trial was ordered. The unanimous judgment of the three judges hearing the appeal was written and delivered by Dixon, by this time Chief Justice of the Court. The centrepiece was an extended history of the meaning of wrong in insanity jurisprudence. Dixon started by citing Stephen’s judgements in Davis (1881) – as well as the later case of Kay (1904) – as precedent for his own reasoning in Porter that the test of knowing wrongness was dependent on being able to reason ‘with a moderate degree of sense and composure’. Dixon then identified a critical fault in the judge’s charge to the jury, in so far as the trial judge had stated the M’Naghten test as raising the question ‘whether the accused knew that firing a shot at another person was against the law’. In so directing, the trial judge had followed the recent decision of the English Court of Appeal in Windle. Dixon was firm in rebuttal: ‘that decision is one that we are not prepared to accept or act upon’. There followed an extraordinarily detailed consideration of the case law before and since M’Naghten, the effect of which was to establish authority for Dixon’s view that wrong meant more than just ‘illegal’ and that the decision in Windle ‘should not be followed’. In directing the jury’s attention to the narrowest judicial view of ‘wrong’, the trial judge in Stapleton had disadvantaged the appellant. While he did not dwell in this case, as he had in Sodeman, on the relevance of uncontrollable impulse, Dixon insisted that it was precisely in cases where the accused was so disturbed as even to treat his action ‘as one of inexorable obligation or inescapable necessity’, that the law’s traditional view of wrong should be respected. ¹⁷ This was a determined dismissal of the narrower view of wrong that now seemed to have taken hold in England.

Thus, when it came to hearing the appeal of Brown in 1959, Dixon was ready to scrutinize with considerable finesse the curt dismissal by the trial judge (and the South Australian Supreme Court) of the availability of an insanity defence. The Brown decision (a unanimous decision again delivered by the Chief Justice) took up the meaning of ‘uncontrollable impulse’ more directly in this case than in the three earlier judgments we have discussed. This was because the trial judge had seriously erred, so Dixon argued, in leading the jury to think that since uncontrollable impulse was not a defence in English law, any evidence of such impulse as a cause of the accused’s action in firing a gun without other apparent motive should lead to a finding of guilty. As Dixon insisted, such a view ignored the authority attached to the interpretation he had advanced in a number of leading cases: that evidence of 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’. In so arguing, Dixon this time cited the opinion of Lord Greer in argument during the English appeal case of R v Ronald True (1922); the outcome of this case had led to the appointment of the Atkin Committee which had recommended (fruitlessly as it turned out) a legislative recognition of the defence of irresistible impulse. ¹⁸

Dixon’s views in Stapleton had a resonance well beyond the Australian jurisdictions. In 1961 New Zealand, which had effectively legislated the M’Naghten Rules in its 1893 Criminal Code, amended ‘wrong’ by qualifying it with ‘morally’; this change was judicially interpreted by the New Zealand Court of Appeal as giving legislative authority to the line followed in Stapleton and Porter, to be preferred over the narrower English construction (Waller, 1977: 172). Indeed the New Zealand Court’s judgment was an enthusiastic endorsement of the different view taken ‘on this side of the world’. Dixon’s direction to the jury in Porter was referred to ‘as a classic in this country’ and used by Justice Turner to expound the meaning of right and wrong. Even though no higher court in that country had explicitly authorized the decision, Stapleton was said to be generally accepted by judges in New Zealand, and ‘juries were directed accordingly. ¹⁹ Also, in Canada in 1976 a bare majority of the Supreme Court followed Windle, over the dissent of four judges who preferred the view adopted by Dixon in Stapleton, ²⁰ another symptom of the wider effect of this departure from English judicial opinion (Stuart, 2001).

Did it matter?

The position espoused by Dixon and Barry was close to the formulation of the law enacted in the Australian jurisdictions that adopted the Criminal Code drafted by Sir Samuel Griffith and adapted from the earlier attempt to enact an English code. A comparative study of Australia insanity law published by Barry in the Canadian Bar Review in 1943, four years before he became a judge, drew out the major difference in statutory law between the ‘Code’ states and the others (Barry, 1943). Three States had effectively instantiated the notion of irresistible impulse, along the lines suggested by Stephen in 1879 and again in 1883. The other states relied alone on the narrower formulations of the M’Naghten Rules. But did it matter when it came to trial?

The patchy figures available in Australian jurisdictions suggest that the broader definition of ‘wrong’ and the greater readiness to acknowledge a concept of irresistible impulse may have had only limited effect in trials and outcomes. The very mixed consequences of mounting an insanity defence, especially in jurisdictions which had abolished capital punishment or diminished its use, were considered a significant disincentive to use of the defence at all – validating Morris’s challenge to its rationale. In Victoria it appears there were only four successful defences in the 10 years after the Sodeman case, in which he had ultimately failed to win his appeal on the grounds of insanity. The imperative of avoiding the gallows undoubtedly was an incentive to use the insanity defence but, in jurisdictions where mercy pleas would usually result in commutation anyway, this incentive may have diminished by the twentieth century. The resort to insanity pleas could be revived by local circumstances as appears to have occurred in Victoria in the 1960s and 1970s. A long battle over the government’s determination to hang Robert Peter Tait in 1962 was only concluded by the success of lawyers in having the man certified under new mental health laws, resulting in his last-minute reprieve (Burns, 1962). A few years later a bitter public campaign failed to save from the gallows Ronald Ryan, who was certainly not in a position to plead insanity. Both cases may have accelerated the attempts by lawyers to look to an insanity defence wherever possible. Over the half century from 1936, the only time successful insanity defences reached double figures were in the years between Victoria’s last hanging (1967) and the formal abolition of the sanction (1974). After abolition the successful use of the defence declined dramatically to rarely more than one or two cases a year (Law Reform Commission of Victoria, 1990: 89). ²¹

Striking evidence of the link between the availability of the death penalty and the use of the insanity defence was provided in a 1970s review of the disposition of mentally disordered persons in Australia. Considering the ratio of persons in custody who had been found not guilty on grounds of insanity to the total number of prisoners serving life sentences, Freiberg (1976) found significantly lower ratios for those jurisdictions (Queensland and New South Wales) which had abolished the death penalty. Significantly one of these states, Queensland, was a Code state, where some of the ambiguities of the M’Naghten Rules had been avoided and the lack ‘of capacity to control his actions’ had been included as an available element of the defence. Freiberg (1976) wondered ‘whether the percentage of insanity verdict’ cases in Victoria might decrease following the abolition of capital punishment’ the year before. Two decades later it was possible to review the impact; Freiberg (1994) then confirmed his earlier suspicion that there was a very considerable decline in the use of the insanity defence following the abolition of capital punishment.

When Norval Morris asked that question in 1969, ‘when the hangman is unemployed, why do we need this defence?’, ²² he was addressing a reality, but his rhetorical question underestimated the complex of reasons sustaining the defence. Two years earlier Morris (1967) had set out his case against the insanity defence, in an article in the Southern California Law Review. In a deterministic world, mental illness (so Morris suggested) was only one of a wide variety of factors that might be argued to disadvantage those who committed crimes. What about a ‘ghetto defence’, in a country where it could be shown conclusively that being brought up in particular urban locales weighed heavily on one’s life chances and one’s freedom to act within the law. To this Kadish (1968) replied that it was no objection to an inadequate proceeding to suggest that it be broadened to capture even more people. More fundamentally Kadish challenged Morris, Wootton and others who wanted to shift all the decisions about mental illness and offending into the arena of dispositions and appropriate treatment, to consider whether they missed the positive protection afforded by the insanity defence (with its consequences of lengthy incarceration) to those who were free of the burden of mental illness. What consequences, he wanted to know, would follow from making potentially all who crossed the courtroom or police station threshold the subjects of invasive and indeterminate treatment? In any case such arguments remained only remotely connected to what faced real defendants in courtrooms. There the fate of those few who advanced the insanity defence continued to be subject to all the uncertainties opened up by the House of Lords in 1843. This article has shown that attempts to nail down those ambiguities could result in quite disparate approaches in different jurisdictions, even within one country. Interacting with other factors (not least in Australia the assertive independence from English judicial authority entertained by Dixon), the impact of medicine and science in judicial thinking, seen by many medical professionals after 1843 as too little or too poorly understood, can instead be seen as shaping in much more subtle ways the emerging distinctions the judges sought to make in interpreting what the Rules meant. Even as he repeated his commitment to basal principle, Chief Justice Dixon of the High Court of Australia loosened the ties of principle to accommodate the extra-legal thinking that bore on matters before the court. When he insisted that wrong meant something more than legally wrong, he confessed that the law could not always get it right, and affirmed that on some matters, juries and common-sense notions had to guide outcomes. Even before it became possible to determine exactly why a person’s disordered mind made it impossible to control an impulse to commit a violent crime, Dixon insisted that the wrongness of an action had also to mean something more than merely being wrong according to law. An openness to the persuasive evidence of medical science assisted such a change in judicial thinking – although this was different from medicine being hegemonic in the court-room. The readiness of the Australian High Court to move away from the constraints of English judicial authority, during decades when this possibility was slowly developing, facilitated the adoption of a notion that irrepressible or uncontrollable impulse might be a factor demanding attention in a defence against a charge of homicide.