Home > Jurisprudence >

The evolution of the idea of criminal responsibility in the Western legal tradition          

                                        Criminal responsibility – definition of the concept                                                                                                   
               Responsibility in pre-state societies                                                                        
                                        The notion of proportionality                                                                                    
                                        Stopping the feud –compensation                                                                           
                                        Increasing centralisation of authority                                                                       
                                        The idea of guilt                                                                                                       
                                        Culpability and early English law                                                                           
                                        Developments in English law before the nineteenth century                                 
                                        Criminal responsibility in the nineteenth and twentieth centuries                          
                                        Criminal responsibility – developments in continental law                                    
                                        Criminal responsibility accepted                                                                           
                                        Difficulties in ascertaining and giving effect to criminal responsibility               
                                        Challenge to the idea of criminal responsibilty                                                     




This essay traces the nature and history of criminal responsibility, a fundamental requirement of European criminal justice. The primitive tradition of ‘payback’ recognized only harm, and with the monetisation of the payback-response a monetary equivalent for injury was sought. The quantum was related to harm not guilt. This was transformed by two historical developments: the introduction of the centralised state, which maintained order without the anarchic threat of payback, and the new idea of guilt – an idea originating in medieval Europe in the teachings of the church. The essay examines in some detail how that idea was introduced in common law and continental criminal justice until in the mid-twentieth century the concept of responsibility was rejected or substantially modified by psychologists and criminologists. This essay is linked to the essay The introduction of western law into Papua New Guinea and especially the endnote The elimination of self-help – the analogy of Anglo-Saxon society.  


Criminal responsibility – definition of the concept  

My purpose is to describe the gradual process by which the concept of criminal responsibility came to occupy a central place in the western legal tradition.

A criminal offence comprises two elements: an act or omission and a state of mind. The act or omission must have been willed. An involuntary act is a contradiction in terms.  In law, it is not an act at all. What is done by an epileptic in a state of epilepsy or by a somnambulist in a somnambulistic state cannot constitute a criminal offence.  By their nature such cases as these are rare. One took place in Victoria in 1950. A mother, Mrs Cogdon, was charged with the murder of her daughter. The evidence showed that she loved her daughter even immoderately.  Mrs Cogdon had killed her daughter whilst in a somnambulistic state, dreaming that the daughter was being attacked by soldiers. This occurred at the time of the Korean war. Mrs Cogdon was acquitted. The act of killing was held to be involuntary.

The other constituent in criminal responsibility is the mental element attaching to an offence. A person will not ordinarily be regarded as responsible, even though the act occurs voluntarily unless the consequences were intended, or, if unintended, the act was accompanied by some other mental state by virtue of which responsibility can be attributed. Intent or mens rea is the presumed requirement in all common law offences. It is also the presumed requirement for any offence newly created by statute unless the contrary is clearly indicated or unless the offence belongs to a class of regulatory offences not ‘criminal in the real sense’.

A person who, whilst insane, commits what would otherwise be a criminal offence is not criminally liable.*

The nature of legal insanity was laid down in McNaghten's case in 1843. A person is legally insane if ‘whilst 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 he was doing what was wrong.’

These then in outline are the main ingredients of ‘criminal responsibility’ (1).*

To examine the concept one would have to go into detail and more closely analyse the nature of 'intent' and its relationship to foresight of the consequences; the defence of mistake (including whether the mistake must be mistake of a fact only and/or whether it must be a reasonable mistake) and the defences of self‑defence, duress, necessity and provocation. My purpose is however, is to trace the evolution of the concept not to analyse it.

Although the idea of responsibility has become central to the administration of criminal justice, it does not follow that the qualifications put upon it in recent times are unjustified. It may be accepted that the human will is not as ‘free’ as was once supposed, that a humane criminal law should take account of the forces determining criminal behaviour, that the difference between normality and abnormality is not clear‑cut and that insanity is not as cognitive as was once thought.

Nevertheless, the idea that a person should only be punished if he or she is ‘responsible’ and to the extent of that responsibility is essential to what we mean by a civilised system of criminal justice. We have come now to think of that as axiomatic.                                                                                                                                                                                                                                                                                  

Responsibility in pre‑State Societies

The idea of responsibility played no part in primitive social regulation.  Before the advent of the State in a developed form, order was maintained within the kinship group by the strength of kinship ties and the resolution of disputes between kin by mediation.  Peace between different groups was kept, if at all, by self-redress.  The chief means of self-redress was revenge or reparation for the loss suffered by the group.  Accordingly, when we speak of liability, it is not compensation in any objective sense of equivalence to which we refer.  Nor was it punishment of the offender.  Liability to the group consisted of satisfaction of group feelings for the loss sustained.  'Payback', the pidgin term used in Papua New Guinea, captures the notion. In the Highlands it nearly always meant reprisal by physical violence. In most primitive societies an attempt was made to modify violence by reparation in the form of gifts (2).  But whether vengeance was exacted or reparation received, the purpose was compensatory not punitive. The purpose was restoration of the loss.  Little or no consideration was given to the personal culpability of the doer of the act.  It may be doubted whether punishment, as a separate and isolated notion, was ever thought of.  Insofar as punitive notions entered into the matter they were likely to fix upon the inanimate object causing the injury or death. The deodand, as it was called, would be delivered to the victim or the kin of the victim. Thus, in Alfred's time, if a tree had caused injury it would be delivered up. Under the Hebrew code a more punitive response is evident. An ox, which had gored a victim, would itself be stoned.

The kin of the victim might be satisfied with vengeance against the wrongdoer’s kin group although the wrongdoer himself went free.  Gore, in 'Justice versus Sorcery’, tells of an extreme case in Papua in the early part of the century. It concerned the Menyama people. A man killed his wife when angered by her. The woman belonged to a village or group some distance away. When her two brothers heard of their sister's death they determined on 'payback'. They set off for the murderer's village, but before reaching it, saw a man of the same language group as their sister's murderer. Satisfied that his death would be sufficient 'payback' they arrowed him to death.  It is evident that the wrong was thought to have been done not by the individual on his own account but vicariously on behalf of the group. This notion survived in a ‘midway’ form into later times. Thus, in Roman law, if a theft or damage to property was committed by a son or slave within the potestas, action would lie against the head of the family who could surrender the person who did the wrong or, alternatively, pay damages (3).

In the absence of the State therefore the Peace was kept by fear of reprisal being taken by a potential victim or by the victim’s kin.

It is clear enough that in the end this was hopelessly inadequate.  Each wrong would, in its turn, engender a violent response. What followed would be a continuing renvoi of violence. Writing of Papua New Guinea, James Sinclair, a District Officer of considerable experience, said:  "If a man of one group is killed that group did not rest until they had killed a member of the opposing group. So started a feud for retaliation by the injured group itself that brought further retaliation and so the vicious vendetta spiralled.”*

The notion of proportionality

The inadequacy and counter‑productiveness of 'payback' were recognised by traditional societies. This was illustrated by various measures taken to halt or mitigate the endless violence accompanying the blood feud. Such were the Play duels of the Arnhem Land aborigines, in which a slight injury inflicted by a spear would satisfy kin feelings, and the shouting matches among the Siane and other peoples of Papua New Guinea (4).

Another approach to limit retaliatory violence was by seeking to equate the reparation exactly to the loss suffered – the talionic principle.

Exodus laid down an ‘eye for an eye, tooth for tooth, hand for hand, foot for foot, burn for burn, bruise for bruise’. Gibbon, in the 'Decline and Fall', said of the Roman Twelve Tables that ‘they approved the inhuman and unequal principle of retaliation: and the forfeit of an eye for an eye, a tooth for tooth, a limb for limb, is rigorously exacted unless the offender can redeem his pardon by a fine of three hundred pounds of copper.’ The talion was never introduced in English law in a precise form (although the early writer Britton (1291) advocated an out and out equation of ‘member’ for ‘member’).

The talionic concept, as embodied in the Hebrew and other Middle Eastern codes, clearly evidenced proportionality. But the equation was strictly referable to injury caused. It in no way related to the degree of guilt or to the wrongdoer's responsibility for the injury.

We see in the Islamic Sharia of today a code in which the talionic idea lingers.

The Sharia embodies the doctrine of qisas: that is 'retaliation' by the entire community.  The qisas had their origins in the Old Testament 'eye for an eye' etc. doctrine. The Sharia punishment for murder or attempted murder is beheading; for theft – amputation of the right hand; for adultery and false accusation of adultery or blasphemy – stoning to death, and for drunkenness severe punishment to the point of death. Together these punishments are known as the hadd. The meaning of the Arabic word is historically interesting. It means 'limitation'. Muhammad was faced with ending the constant feuds among the Arabs which had dragged on for generations. These were to be terminated by the imposition of the qisas.

The punishments imposed by the Sharia are fixed and to that extent ‘limited’ but they are also sufficiently severe to assuage the desire for revenge.*

Stopping the Feud – Compensation

Proportionality or the talionic principle was thus perhaps the first important step in limiting violence. In the long run, however, the monetisation of reparation was more effective This made it possible to split the compensatory and punitive elements in response to a wrong. As long as 'punishment' was subordinate to and merged in the reparative objective, the concept of responsibility, could not begin to develop (5).

It was thus hoped to stop the feud by inducing the injured man or the kin of the deceased to accept monetary compensation. Acceptance of compensation was at first voluntary in both Roman and Anglo‑Saxon law. In Anglo‑Saxon law the injured man could accept the bot and the kin relatives of a deceased could accept the wer. If neither payment of the bot or the wer were made, the feud could be continued.

The Anglo‑Saxon law laid down by King Aethelbert in the sixth century specified a Table of compensation. The bot was referable to the injury mentioned in the table: ‘if an ear is struck off, the bot is twelve shillings; if an ear is pierced, only the amount of three shillings; if an eye is knocked out, fifty shillings ... a broken jaw, twenty shillings’ and so on.

Next, however, acceptance of compensation became obligatory. No doubt this was due to the advancing authority of the King. It was, though, as Maitland said, ‘no easy task to force the injured man or the slain man's kinsfolk to accept a money compensation instead of resorting to reprisals’. But by Alfred's day it was unlawful to begin a feud until an attempt has been made to exact monetary reparation. Under the law of Edmund the slayer was to have twelve months to pay the wer before he could be attacked.

There was an early and important development to that which has been described.  The offender was required not only to pay the bot to the injured and those who had suffered loss.  He must also pay something to the King. This payment was known as the wite.  In other words, the offender had not only injured the kin but had broken the King's Peace. He must therefore also buy back the Peace which he had broken. In time, certain offences became botless because of their peculiar gravity. They were simply punished by the King.  We have here the beginnings of punishment by the State.

Increasing Centralisation of Authority

 Certain conditions were necessary for progression towards recognition of individual responsibility. It was first necessary for the individual wrongdoer to be separated from the group to which he belonged.  Secondly, the idea of punishment of the wrongdoer for an offence had to be isolated from reparation to the victim or the victim’s kinship group.  The bots and the wite symbolised the beginnings of this and, although the wite seemed almost as if it too were compensatory (so as to compensate the King for his Peace having been broken), it carried within it the notion of punishment.

In the twelfth century this elaborate system of bots and wites disappeared quite suddenly. Maintenance of the King's Peace became uppermost. It began perhaps with a relaxation of kin bonds and the replacement of them by vassalage. To some extent bots and wites had become discredited because the amounts payable were too unrelated to the offender’s capacity to pay. But, in England, what proved critical was the deliberate extension of the King's Peace throughout the Kingdom.*

About this time, or shortly afterwards, the term 'felony' was first used. This term related to the worst crimes. Importantly, they were crimes which were botless. No bot could redeem them. Wilful homicide became a capital crime. The kinsfolk of the slain lost their right to a wer or to compensation of any sort.

This extinction of the feud and substitution of Royal authority was not universal throughout Europe. In parts of Europe other than England the blood feud continued – longest in Friesland, Lower Saxony and parts of Switzerland where it continued until the sixteenth century.

Marc Bloch described the position elsewhere in Europe as follows:

"In short, except in England, where, after the Conquest the disappearance of any legal right of vengeance was one of the aspects of the royal ‘Tyranny’, they confined themselves to moderating the more extreme manifestations of practices which they were unable or perhaps unwilling to stop altogether.”*

In England, the idea of punishment, as distinct from reparation of the group, had finally emerged. The demarcation between ‘crime’ (an offence against the State) and ‘tort’ (a civil remedy for recovery of compensation by private persons for a wrong) had begun. That demarcation remained for long indistinct but for our purposes we can from the thirteenth century speak clearly enough of 'punishment'.

It did not follow though, because it was now the individual with whom the King’s criminal courts were concerned and the question was punishment of the individual, that the group could be dispensed with. The King's power, although greatly enhanced, was not yet sufficient for that. The group was needed for law enforcement. And so the group was made liable for punishment if it failed to yield up an offender in its midst. Group liability of this kind has long disappeared in Europe (except for the notorious Nazi sippenhaft) but in thirteenth century England a whole village might be 'amerced' to the King if the offender belonging to the village was not apprehended.

The disappearance of this practice came with the establishment of Royal authority. This was not easy in those days of poor communication and, indeed, in England there was no effective police force until the nineteenth century. Nevertheless the more compact areas of Europe enabled Royal authority to take over law enforcement from the group or community at a relatively early date.  It is interesting that group liability continued to be a feature of Chinese law for a long time and was to be found in both the T’ang and Ch'ing codes. It may be that, although centralisation took place in China at an early date, the geographical area to be controlled was too large for effective law enforcement from the centre. Hence the dependence of the State upon the clan for enforcement.

The Idea of Guilt

 Punishment of the individual as distinct from compensation of the group is only a condition – although a necessary condition – for giving effect to personal criminal responsibility. What was needed also was the idea of ‘guilt’.

This did not come about as naturally as we might suppose. For such a notion acknowledges exculpation of the individual if there is a lack of guilt even where serious harm may have eventuated from the conduct in question. But it is precisely this harm – disturbance of the Peace – which the King was concerned to prevent.

The church, on the other hand, was concerned not with harm but with sin.  It was the saving of souls, not the Peace, to which it gave its concern. Sin was very much a matter of the sinner's state of mind. The now revered legal maxim – ‘actus non facit nisi mens sit rea’ – was used first by St Augustine.

It was thus the church through canon law which was most immediately responsible for introducing the idea of ‘guilt’ into European law.

It was not wholly christian in derivation. The church had absorbed stoic teaching and stoic ethics. The stoics emphasised the state of mind of the doer of the act. This was to some extent transposed into law. Cicero said, ‘Take care that punishment does not exceed the guilt’.*  Although the canon law was the first code of law with a fully developed concept of guilt it drew upon the later Roman law in achieving this(6).*  It is evident from the Digest that the developing Roman law of the Empire was paying increasing attention to intent and capacity. Thus Ulpian, as recorded in the Digest, said that ‘madmen and infants’ were not capable in law and that the essence of certain offences was that they had been done intentionally. The canon law drew upon Roman law but departed from it where the older formalism would have overridden intention. In the canon law ‘bad faith’ was critical no matter what the form of the law. As Gabriel Le Bras has written, canon law in the criminal field provided for the ‘precise investigation in any given case of the intention’.*

Le Bras adds a qualification:

"It must be added... that while in general the Canon law maintained the principle of personal responsibility for faults, it did not altogether escape the tendency, common to all medieval legal systems, which is, in determining penalty, to take account of the group as much as of the individual and to obtain reparation … by demanding it from the innocent if it could not be paid by the guilty”.*

Canon law helped to form European criminal law both directly and through the reception of Roman law by continental countries. Its influence was very much greater on continental law, which adopted canonical procedures, than in England.

Canon law was, however, an instrument of the church. As we have said, the ‘guilt’ with which it was concerned was the ‘guilt’ of the sinner. Wrong and sin were not clearly distinguished. Crimen and peccatum were often used synonymously in the early texts. Wrongs were defined in the theological order of mortal and venial sin. It was this theological twist to the matter of intention and the offender's state of mind, which substantially detracted from the contribution of the canon law to the evolution of criminal responsibility. The sanctions were directed not to punishment but to the amendment of the sinner. Penance was required to be done so that the guilty sinner would repent and so that atonement might be obtained. We see all of this most clearly in the supreme sin of heresy where the state of the heretic's soul was alone in question.

Culpability and early English Law

English common law, which was relatively unaffected by canon law, was slow to accept the idea of moral guilt. To a large extent under the early law a man acted at his peril. Even in the late Middle Ages the statement by Chief Justice Brian that ‘the thought of man shall not be tried for the devil himself knoweth not the thought of man’* would seem to have been generally accepted.

One should not, however, overstate this. Even in the period of the wer and wite an unintentional homicide would result in the payment of the wer only.  The Leges Henrici also make passing reference to St Augustine's statement – ‘actus non facit nisi mens sit rea’ – even though the only application of the maxim contemplated was for the offence of perjury. (It would seem that at this time it was seriously envisaged that a person might be convicted of perjury although believing the false statement to have been true.)  In the thirteenth century, the great medieval writer on English law, Bracton, who was heavily influenced by the canonist, Bernard of Pavia, stressed moral guilt and stated that homicide would not be committed unless intention to injure were present. Nor, according to Bracton, could an infant or madman be held criminally liable.

Bracton's influence was great but he was, in this respect, ahead of his time. English law, unlike the canon law, had no graduated series of punishments. Flexibility of punishment is essential if effect is to be given to the idea of guilt. English law had no such flexibility. ‘For this reason’, as Maitland has written, ‘in Bracton's text we may see Bernard's doctrine of homicide floating on the surface of, and scarcely mingling with the coarser English law, which hardly knew what to do with the manslayer who was not guiltless but did not deserve to be called a felon and put to death.’*


Developments in English Law before the Nineteenth Century

In Bracton's day the only homicide justifiable under English law was the carrying out of a lawful sentence of death. Death caused by misadventure or in the course of self defence was still regarded as felonious. The Justices could neither acquit nor grant a pardon. But in time it became the practice from them to recommend that the King grant mercy. This practice was recognised in 1278 by the Statute of Gloucester. A pardon would not detract from the rights of the deceased's kin. And the Leges Henrici clearly recognised the right of the kin to the wer.

In the centuries that followed, self‑defence and misadventure became defences proper. But even in the eighteenth century the onus still lay upon the accused to establish these defences. It was decided in R v Oneby (1727) that the accidental character of death could, on a charge of murder, only be raised by way of defence. Sir Michael Foster, in the Discourse on Homicide (1762) stated that: ‘in every charge of murder, the fact of killing, being first proved, all the circumstances of the accident, necessity or infirmity are to be satisfactorily proved by the prisoner...’

The medieval law had also made no distinction between murder and manslaughter: that is between intentional and recklessly caused homicide. The difference in these offences came about in a curious way.

The relationship between Church and King required giving to ‘clerks’ what was described as the benefit of clergy. An accused entitled to benefit of clergy could plead his clergy and escape execution even on a charge of murder. Any person who could read a verse was considered to be a ‘clerk’. For over a century a convicted murderer could, by reading a verse, escape punishment. Obviously such a scandal could not continue forever. A series of statutes in the fifteenth century were to result in the modern distinction between murder and manslaughter. This was done by denying benefit of clergy to clerks who committed murder with ‘malice aforethought’.

From the beginning of the sixteenth century the great concern of English criminal law became the nature of ‘malice aforethought’.

One difficulty lay in defining the mental state required for each of the different offences: murder, manslaughter and larceny. Another question was the position which was to obtain where an unintended death had occurred in the course of the commission of an unlawful act. What, for example, should be the case if A, intending to steal B's poultry, shoots at them and accidentally kills C?

Thinking was heavily influenced by institutional writers such as Staunford, Lambard, Coke and Hale. Coke said that every homicide, however unintentional, if caused in the course of committing an offence, was murder because the unlawful purpose supplied the malice. Under the influence of Hale murder became restricted to death occurring in the course of the commission of a felony and thereafter it became restricted further to the case of a felonious act likely to endanger life.

In the seventeenth century English law began to consider the defences arising from incapacity to form the required mental state. It was settled in the middle ages that madness, if it existed when the crime was committed, negated liability. But what left the lawyers of that age puzzled was the case of partial insanity. No difficulty arose if the accused suffered intervals of madness but what of the case where the accused had not wholly lost his or her reason? Hale took the view that partial insanity would not excuse because most ‘felons are under a degree of partial insanity when they commit these offences.’

Drunkenness also led to difficulties. ‘If a person that is drunk’, it was said in 1551, ‘kills another, this shall be felony, and he shall be hanged for it, and yet he did it through ignorance, for when he was drunk he had no understanding nor memory; but inasmuch as that ignorance was occasioned by his own act and folly and he might have avoided it he shall not be privileged thereby’*

Criminal Responsibility in the nineteenth and twentieth Centuries

During the nineteenth century the idea of criminal responsibility acquired its present significance. In England this was in part due to the natural evolution we have described. But, in addition, the intellectual conditions of the age were congenial to its development. Speaking of these Roscoe Pound said:

"In the nineteenth century ... law became a realisation of the idea of liberty and existed to bring about the widest possible individual liberty. Liberty was the free will in action. Hence it was the business of the legal order to give the widest effect to the declared will and to impose no duties except in order to effectuate the will or to reconcile the will of one to the will of the others by a universal law. What had been a positive, creative theory on the basis of intention became a negative restraining, one might say pruning theory. Except on the basis of intention liability could flow only from culpable conduct…”*

We left self‑defence and misadventure in the eighteenth century as having become defences instead of mere grounds for pardon. But it remained uncertain whether these excuses must be found specifically or whether a general verdict might be given. In his Discourse on Homicide, Foster expressed the view that a general verdict was admissible. The point is not merely technical. For it raises the question whether the issue of guilt, including the mental element, is a single issue or whether it is to be broken up into a number of separate issues, in which case the general question of guilt need not be proved or decided.  The general verdict became established in English criminal law.

Absolute madness was a defence.  Erskine's successful defence of Hadfield on a charge of attempting to murder George III confirmed that.  But just what constituted insanity was left uncertain until the rules were authoritatively laid down in McNaghten’s case in 1843.

Daniel McNaghten had shot and killed Edmund Drummond, Sir Robert Peel's private secretary in mistake for Peel.  McNaghten suffered from delusions. He was acquitted on the grounds of insanity. The acquittal was highly controversial leading to a debate in the House of Lords. As a result, the Judges were asked by the Lords to state the rules of law relating to insanity.  The Judges thus declared the rules extrajudicially. The primary rule was that a person would be excused on the ground of insanity ‘if through disease of the mind he or she either did not know what he or she was doing or did not know that it was wrong’.

These nineteenth century developments in the law of insanity contributed to the view that drunkenness should be allowed to modify or negative criminal liability.

The law was in a dilemma where drunkenness was self‑induced. In 1909, in R v Meade, it was said that the presumption a man intends the natural consequences of his act, may be rebutted by showing that his mind was ‘so affected by the drink he had taken that he was incapable of knowing what he was doing was dangerous, i.e. likely to inflict serious injury.’ But in 1920, in Director of Public Prosecutions v Beard, the House of Lords held that ‘inasmuch as mens rea or another state of mind is, with very few exceptions, a necessary constituent of all crimes, the true rule is that, if drunkenness has produced in a person accused of a crime an incapacity for form a particular intent necessary for the commission of that crime, he cannot be convicted and that it is in those circumstances only that drunkenness, not amounting to insanity is a defence’.

The elevation of the idea of criminal responsibility in English law culminated in the rejection by the House of Lords of the old eighteenth century proposition that the onus lay on the defence to prove accident. This was declared in 1935 in Woolmington v Director of Public Prosecutions.  The trial judge had on a charge of murder directed the jury that once the Crown had proved the killing of the person it must be presumed to be murder and that it was for the prisoner to prove circumstances which would excuse the homicide as accident. The House unanimously held this to be a misdirection. The conviction was quashed.

The Lord Chancellor, Lord Sankey said:

"Throughout the web of English criminal law one golden thread is always to be seen; that it is the duty of the prosecution to prove the prisoner's guilt... if at the end of and on the whole of the case, there is reasonable doubt, created by the evidence given by either the prosecution or the prisoner, as to whether the prisoner killed the deceased with a malicious intention, the prosecution has not made out the case and the prisoner is entitled to an acquittal."

As the Australian Chief Justice, Owen Dixon, said of the decision, ‘it makes plain that the last vestige has disappeared of the ancient principle which placed the fact of killing first and admitted only as exculpation, or excuse, circumstances which made it no felony’. Woolmington's case was, in many ways, the apogee of criminal responsibility in English law. But in 1961 there was a not unimportant epilogue in the decision of the House of Lords in Smith's case – a decision which excited surprise and disapproval.

In that case the driver of a van refused to stop when directed by a police constable to do so and drove off dragging the constable who was eventually thrown into the course of oncoming traffic and killed. The driver was charged with murder. The House of Lords said that it was irrelevant whether or not the driver actually contemplated the constable's death as a probability; the sole question being ‘whether the unlawful and voluntary acts were of such a kind that grievous bodily harm was the natural and probable result’.

This decision seemed to negate the whole course of historical development. The High Court of Australia recoiled from this objective approach to responsibility. Not only did it refuse the follow Smith's case but it led to an announcement by the Court that it would no longer follow House of Lord's decisions which it had hitherto done as a matter of comity.

Criminal responsibility – Developments in continental law

Before the eighteenth century, it could be said that the criminal laws of most European countries were brutal and unjust. The Enlightenment transformed this. The writings of Beccaria were of enormous influence. So much have we taken for granted what Beccaria urged and proclaimed that it is difficult now to recognise the revolutionary impact of his work in 'De Delitte e della Pene'. During the latter part of the eighteenth century new criminal codes were prepared or promulgated under his influence by Frederick the Great of Prussia, Gustavus III of Sweden, Joseph II of Austria and Catherine the Great of Russia. The French Penal Code of 1791 embodied the principles which he laid down.

Beccaria introduced rationality and humanity into what had hitherto been the brutal business of crime and punishment. It followed inevitably that criminal responsibility should come to have the same significance on the Continent as it had in England. Indeed, by the end of the nineteenth century, German law was theoretically more developed than English law in its conception of mens rea, by distinguishing in the Code provisions the various forms of subjective guilt. Thus if A intending to kill aimed his pistol at B but killed C he would in English law have been guilty of murder at a time when in German law he would have been regarded as killing with negligence. It is doubtful whether, in the upshot, these refinements were of great significance. Whilst there was less talk of the ‘reasonable man’ in German law, inevitably there were circumstances in which it was necessary to hold the accused to an objective standard of forseeability and care. Nevertheless, the German approach is illustrative of the fact that the concept of criminal responsibility had become common to the western legal tradition.

Criminal responsibility accepted

By 1935 – the year of Woolmington – we may say that the principle a person should be punished only if he or she was responsible for a proscribed act or omission had become central not only to English but to western criminal justice.  It had been a long process.  The idea of criminal responsibility is now universally accepted: most people would for example find it abhorrent to punish the insane equally with the sane (7). The idea though, it is useful to recall, had itself once been unthinkable. We go back to the deodand when liability was visited upon the inanimate thing which had caused injury or to a time when the ox that had gored was stoned to death. Even as late as the middle ages animals causing injury were hanged (8).

Three things had happened. First, there had been a fundamental mental change in the idea of causation. It was no longer believed that harms were caused by volitionless things or by magical spirits inhabiting trees or other objects.  Once the significance human volition came to be recognised it became possible to distinguish liability in the cases of intentional, negligent or accidental harms. Secondly, the individual had become detached from the group. Liability ceased to be a matter of group vengeance or group reparation.  Finally, it had come to be recognised that the individual was an autonomous moral agent and that liability was to be proportioned according to the exercise of that autonomy. In commenting upon Woolmington, Sir Owen Dixon said of English law that ‘for eight centuries the course of its (English Law's) very gradual evolution has been from an almost exclusive concern with the external act which occasioned death to a primary concern with the mind of the man who did the act.’

Difficulties in ascertaining and giving effect to criminal responsibility

Giving effect to the idea of criminal responsibility does not, in theory, detract from the primary purpose of the criminal law to protect the community.  That purpose is not advanced by punishing a non-responsible person for an unintended or unavoidable act.  But we move into a much more difficult area if we insist, as the criminal law does, that the Crown prove that responsibility.

In saying that ‘the thought of man is not triable’ in the fifteenth century, Chief Justice Brian was only just exaggerating.  Establishing a person’s mental state is difficult.  As Holdsworth, in his History of English Law wrote, the problem is one of evidence not of principle:

            “We must adopt an external standard in adjudicating upon the weight of evidence adduced to prove or disprove mens rea.  That, of course, does not mean that the law bases criminal liability upon non-compliance with an external standard.  So to argue is to confuse the evidence for a proposition with the proposition used by that evidence”.

In a sense it is true, as Holdsworth remarked, that the problem is only one of evidence.  But, stating the matter more precisely, it is an evidentiary problem produced by an interaction of two principles or values and the need for compromise if effect is to be given to the primary object of the criminal law.

The imposition of the onus on the Crown is an aspect of procedural fairness.  Procedural fairness is a value of our criminal law designed to give the individual protection against the power of the State.  As such, it qualifies the primary purpose of the criminal law.  That is to say, the deterrent objective is reduced by the recognised need of the State to prove guilt, including the accused’s mental state, in order to secure a conviction.

Even in a system in which criminal responsibility is central (but in which also the Crown is required to prove it) it is not possible to give effect by way of exculpation to every individual peculiarity, infirmity or disability.  Ignorance of the law can be allowed to excuse only in limited circumstances* ; duress will be a defence if the threats would affect ‘a person of ordinary firmness’* ; necessity will afford a defence to a person ‘of ordinary fortitude’*.  The scandal of duelling could never have been abolished if the defence of provocation had been allowed whenever a person felt provoked.  That defence will avail only if it would have ‘caused in any reasonable person, and actually caused in the accused, a sudden and temporary loss of self control’*.

Challenge to the idea of criminal responsibility

In the twentieth century much effort has been directed to working out how the goals of preventing crime and recognising individual responsibility might be reconciled. Legal doctrine has fluctuated.  In areas such as the ‘strict liability’ offences, relating to conduct ‘not criminal in the real sense’, the element of responsibility has been dispensed with or substantially reduced.  Particular difficulties have arisen in isolating the criminal responsibility of corporations. Specific problems have also been encountered in the international field where superior orders has been denied as an allowable defence under the Nuremberg Principle. None of these however bear upon the principle of criminal responsibility but only upon its application.

There has been one contemporary challenge to the idea itself.  This occurred in the immediate post-war period in the area of criminal punishment.(10)

The criminological theory which we shall describe was formulated chiefly by psychiatrists, criminologists and social scientists.  In whatever form the theory was expressed, it proceeded from the proposition that all human choices are the product of forces beyond conscious control. Therefore, as one of its main proponents, Professor Glueck, said, ‘the basic premise of free will is faulty’. From this it was deduced that ‘concepts of responsibility and punishment... are theological and metaphysical anachronisms... questions of 'mad' or 'bad' with their erroneous emotional loadings do not arise’.*

The theory is not in terms logically inconsistent with the idea of criminal responsibility.  That idea does not say whether and when human beings are ‘responsible’ but only that, in so far as they are, criminal liability will attach. Nevertheless, the underlying assumption of criminal responsibility is that there is scope for free will in human conduct.

Professor Glueck was supervising psychiatrist at Sing Sing prison. His views and those of other psychiatrists became widespread and led in time to a kind of mood affecting, to a greater or less degree, criminological and legal thinking. To some extent their theories of criminal punishment harmonised with an optimism that criminals would yield to treatment and reform.

In 1959 Baroness Wooten felt able to say that ‘revolutionary though the prospect of abandoning the concept of responsibility may be it is clear that we are treading steadily towards it’.*  Baroness Wootten was not a psychiatrist but a magistrate and social scientist. She reached this opinion from her experience as a magistrate and from a study of cases recorded under the United Kingdom Homicide Act 1957. She maintained that ‘no intelligible distinction could be drawn between psychopathy and wickedness’. And in the Hamlyn Lectures which she delivered shortly afterwards (1963) she said that ‘the only solution for the future was to allow the concept of responsibility to wither away’.

These views were given force by the discovery of a category of mentally abnormal offender who was neither a neurotic nor a psychotic. Persons within this group did not suffer from any identifiable mental disorder. 'Psychopath' has now seeped into ordinary language. In the early years, the psychoanalyst, Flugel, wrote of psychopaths as ‘a curious, and as yet not sufficiently understood type of case apparently utilising deficient or disordered super‑ego development’ (the super‑ego is, inter alia, the moral critic that maintains in the ego an unconscious guilt sense). The psychopath is often quite intelligent and appreciates at the cognitive level the difference between what is regarded as right and what is regarded as wrong and also the consequences of engaging in wrong conduct. The psychopath may even take evasive action to avoid being caught or suffering punishment. All of this suggests that psychopaths are merely persons who are disposed to be bad. As Baroness Wootten has said ‘psychopaths are extremely selfish persons and nobody knows what makes them so’. In every day terms the psychopath suffers from an undeveloped conscience or sense of guilt (9).

It is against this background that the movement for by‑passing responsibility in criminal behaviour grew.

We must not however confuse this with the earlier historical denial of responsibility. There was no suggestion by criminology that non‑responsible persons should be punished for loss or injury caused by their actions as if criminally culpable. What was denied was that those engaging in criminal behaviour were 'responsible'. It was not suggested that those engaging in anti‑social behaviour, although non‑responsible, should fall outside the criminal law and go free. They were still to be subject to State control. The sole determinant though was to be treatment of the offender with the object of preventing him or her from offending again. Punishment was to have nothing to do with it.

As Professor Glueck wrote:

"The problem would be eased, however, and the question of responsibility would not have to be raised, if the concept of management of the anti‑social individual were changed from that of punishment as the major instrument of control, to the concept of the anti‑social person as a sick person, in need of treatment rather than punishment."

Baroness Wootten's position was more moderate.  She would eliminate responsibility ‘only because of the difficulty in determining whether an offender was in fact 'responsible'. ‘There are’ she said, ‘cases where persons respond predictably to various stimuli. There are cases where medical treatment works: there are cases where it fails. Equally there are cases in which deterrent penalties appear to deter those upon whom they are imposed from committing further offences, and there are cases in which they do not.’

It is necessary to comment upon some of the consequences which flow from these theories.

First, we must ask what effect the elimination of responsibility would have on the content of criminal offences. Those advocating the philosophy of non-responsibility have not been greatly concerned with this question. They have directed their attention chiefly to punishment and penalty. Some, such as Dr Glueck, have failed to perceive the logical consequences of their position for, as long ago as 1925, he said that ‘the jury could still pass upon the mental element of the crime – the mens rea’.  But if nobody is ‘responsible’ for the ‘offence’ and that which has occurred is the product of some form of sickness, of what importance is mens rea?. More logically, Baroness Wootten has pointed out that the accused's mental state ‘would be relevant not to the question of determining the measure of culpability, but to the choice of the treatment most likely to be effective in discouraging him from offending again’. The only reason therefore for examining the offender's mental state would be to diagnose the abnormality from which he or she suffered and the propensity, given that abnormality, to recommit offences. It may be doubted whether an examination of the offender's state of mind for these purposes would be appropriate for Courts rather than psychiatric examination. Thus the ‘primary job of the courts would be that of determining by whom a particular act has already been committed.’

Punishment being irrelevant, the coercion to be applied to the offender would continue so long as he or she needed treatment. This led to proposals to separate the sentencing process from the Courts and to hand it over to Administrative Treatment Tribunals composed of psychiatrists, psychologists and social workers. As prevention of the individual offender was to be the only object, detention would continue so long as treatment was necessary if the offender were amenable to treatment, or so long as incapacitation was required, where the offender was not treatable but yet showed a propensity to commit further offences. Thus if a person were to commit a murder in a sudden fit of jealousy in circumstances where, there was no reason to expect a homicidal recurrence, logically the murderer would be released. A shop stealer, on the other hand, suffering from kleptomania would, upon the same logic, remain in detention until the ‘Treatment Tribunal’ was satisfied of his or her cure.

These ideas of criminal justice, although for a time forming the prevailing mood, have been rejected and the idea of criminal responsibility has held firm under this benevolently‑intended assault. This has been so in part because the practical experience of endeavouring to treat and reform criminals has proved to be much more difficult and much less successful than had originally been hoped. But it is also due to the recognition that human beings are not, in the words of Professor Herbert Hart ‘alterable, predictable, curable and manipulable things’. Certainly our acts are affected by predisposing causes stretching in a continuum from almost absolute freedom to compulsion but the difficulty in drawing the line of responsibility does not mean that there is no line to be drawn.*

October 1994


(1)       "Criminal responsibility" and moral responsibility are not the same. The criminal law is concerned with anti­social acts. It is not nowadays concerned with the morality of the offender as such. The tendency to the contrary in Shaw v. DPP (1961) 2 W.L.R. 897 where the House of Lords recognised an offence of conspiring to corrupt public morals has been widely and sharply criticised. It is true that the ancient origins of mens rea, especially in the canon law, were associated with morality and sin but in contemporary law it has come to mean intent, foresight of the consequences and freedom of choice.


(2)       ‘Payback’ might present the appearance of anarchy but it was the threat of ‘payback’ which maintained some form of order through fear of reprisal,.  It was, as mentioned in the text, an inadequate means of doing so.  The ‘stateless’ society walks a tightrope.  It is the threat of force which maintains a degree of peace but that threat is only viable if there is a prospect of force being inflicted.  In the absence of the State, however, only the victim or kin are interested in applying force.  They though are motivated by feelings of revenge or at all events by motives unrelated to the maintenance of order.  Inevitably, violence to the offender is likely to be applied with impulsive irrationality.


(3)       Bloch instances a case as late as 1260 before the Parlement of Paris in which a knight Louis Defeux was wounded by Thomas Douzouer. Louis proceeded against his assailant in Court.  The accused explained that he had himself been attacked sometime before by a nephew of the victim. He had in accordance with the Royal Ordinance, waited forty days before taking revenge.  The knight agreed but said that what his nephew had done was of no concern to him. The knight’s argument was of no avail. The judge held that an act of an individual involved all his kinsfolk.


(4)       “… a margarada… is not held straightway after an offence has taken place, but only after people’s rage and resentment have had time to cool.  Arrangements are always made by the injured party.  The two opposing groups, painted in white clay, stand just out of spear-throwing reach, with mangrove jungle or scrub behind for protection if necessary.  Members of the aggrieved party advance toward the opposite side in a totemic dance, then walk back.  The others do the same.  Now they are ready for the ‘duel’.  Men of the accused’s group run irregularly across the ground, and with them run two men who are closely related to both sides.  Spears are flung at them – but usually with the stone or iron blades removed.  They can dodge, but must not throw back the spears, or abuse which accompanies them.  This takes the first edge off the injured clan’s anger.  After a brief lull, the accused man or men runs across the ground.  This time the blades are left in the spears, and flung at him one after another.  Elders from both sides try to restrain the participants, warning them to keep their tempers in check.  Finally, the accused man’s party dance across to their opponents.  If they spear him in the thigh the matter is at an end, and both groups join in dancing.  Ideally a thigh wound is enough but sometimes the accused is killed.  Very occasionally, a fight develops from the ritual settlement; in which case the feud continues until further attempts are made to curtail it through a magarada.  If there is no final spearing the case is still open, and further retaliation can be expected.”:  The World of the First Australians’, R.M. & C.H. Berndt, pp.293/294.


(5)       Monetisation as a means of satisfying the feelings of victim or kin preceded the existence of the State.  This applied in parts of Papua New Guinea.  Thus, in ‘Behind the Ranges’, Sinclair describes how the moga could compensate even in cases of death, ‘but the death could be paid for by a ceremonial presentaiton of pigs and mother-of-pearl shell.  These payments, or rather exchanges…were called moga’.  Compensation, as a substitute for retaliation, thus preceded the development of the distinction between the criminal and the civil law.  Once the State arrived its eventual insistence upon a monopoly of force in the community meant that it took over the awarding and enforcement of compensatory remedies.


(6)       From the third century onwards the German tribes established themselves on the ruins of the western Roman Empire. A kind of barbarised Roman law survived. But in the east a sophisticated Roman law was accepted and developed by the Byzantines. Justinian appointed a Commission of Ten, headed by the chief of the Imperial Chancery, Tribonian, to codify the laws. This huge task was begun in A.D.528 and the Code was completed in the incredibly quick time of three years. The Commission was then charged with examining existing juristic writings and putting them all together in what became known as the Digests.


The fifty books of the Digest contain the writings of 39 Roman jurists. Of the main five jurists mentioned, the earliest was Pomponius, who died in A.D.138, whilst the last, Ulpian, was murdered in A.D.223.


In the West some use was made of the code but the Digests were apparently unknown for hundreds of years. The laws of the west, other than England, were the crude barbarian codes based upon Roman law but mingled with tribal customs.


In the twelfth century, the Digests and the entire Corpus luris Civilis, as it was called, were rediscovered and this led to a renaissance of legal learning throughout Europe.


It may be said that after being in abeyance for five centuries Roman law came to life again in the twelfth century. "Within the range of human history" Vinogradoff has written "there is no more momentous and puzzling problem than that connected with the fate of Roman law after the downfall of the Roman state... the story I'm about to tell is, in a sense, a ghost story. It treats of a second life of Roman law after the demise of the body in which it first saw the light."


(7)       The United Kingdom Royal Commission on Capital Punishment (1949) said at page 98:


"We do not believe that the insane murderer should be punished equally with the sane, or that, although he ought not to be executed as a punishment, he should be painlessly exterminated as a matter of social hygiene. The argument is in each case the same – that his continued existence would be of no benefit to himself, and that he will be not only a useless burden, but also a potential danger to the community, since there is always a risk that he may escape and commit another crime. Such doctrines have been preached and practiced in National‑Socialist Germany, but they are repugnant to the moral traditions of western civilisation..."


(8)       A horse that had killed a man was hanged in Dijon in 1389; a sow with a litter of six was sentenced for the murder of a child at Savigny in 1457, but the baby pigs were reprieved because of lack of positive proof of complicity.


(9)       An extreme and barbarous example of psychopathy was the case of Andrew Norrie. In 1986 Norrie was found guilty and sent to prison for shooting two persons in what was described as a shooting spree. Whilst in prison he attacked an inmate and was charged with wounding with intent to murder. A psychiatrists told the court on that charge that the idea of being goaled for life did not bother Norrie. There are ‘plenty of people inside to stab, bash and kill’. Norrie had added ‘with a happy smile’ that the victim of his attack ‘was just unlucky’. The psychiatrist described him as an ‘intractable psychopath’ lacking any moral conscience.


Norrie had told the psychiatrist that the murders in 1986 had been ‘good fun’ and had made him feel good. The thrill in killing people about whom he felt indifferent was ‘in seeing his victims falling around, screaming and moaning’.


(10)      Whilst modern criminology provided the challenge it did not originate the idea of reform as an object of criminal punishment.


            Beccaria had introduced rationality into the criminal law, replacing the medieval treatment of crime as sin with a defined social purpose of deterring crime.  Punishment ceased to be a pretext for sadism and cruelty.  Along with social purpose, enforcement of the criminal law and criminal punishment became limited by the offender’s or alleged offender’s moral responsibility.


            At the same time serious attempts were made to use punishment and imprisonment as a means of reforming the criminal.  Howard and others were appalled by the contaminating effect of prisoners mixing in prison.  A system of solitary punishment was devised and implemented with results that were however retrograde.


            These ideas of deterrence, moral responsibility and reformation were the product of the Enlightenment.  Moral responsibility was the obverse of the Enlightenment ideal of autonomy. Theories of reformation in the criminal area were an application of the Enlightenment idea of reform based upon the view that man was perfectible. 


            Potentially some of these ideas were in conflict.  Thus the objective of deterrence given extreme effect by the doctrine of maximum severity all but obliterated the idea of responsibility as a limiting factor in punishment.  The criminological theories of reformation in the twentieth century, when taken to an extreme, overthrew individual responsibility.  They did so avowedly because the idea of responsibility rested upon the premise of free will.  This the criminologists denied.*




*           Although for a time in England, by an historical oddity arising out of an attack upon Queen Victoria, an insane person would be convicted but excused on the grounds of insanity.


*           See Endnotes.


*           Behind the Ranges.


*           We have seen very recently, in the case of the English nurses, the old and primitive idea of punishment directed to satisfying the kin of the victim so that the victim’s relative was allowed to determine whether the sentence of death would be carried out together with the acceptance of payment by the accused in satisfaction (1997).


*           The Conqueror had declared that all persons whom he had brought with him were within his Peace.


*           Feudal Society p.128.


*           Bk 1 Ch.25, Sect.89.


*           In 1140, Gratian, a Bolognese monk applied the Corpus Civilis Juris to the mass of church decrees, rules and decretals. His Decretal consolidated these. Technically, it was a private work but it was so generally used in universities and in church courts that it became the source of universal church law. The Popes and the councils convened by them, supplemented the Decretum in its consolidated form. It was authorised to be published by Gregory IX in 1234. With the latest edition of 1317 it remained in force until 1918.


*           Legacy of the Middle Ages, Oxford, p.322.


*           Op. cit. p.357.


*           1477


*           Pollock and Maitland p.478.


*           Reneger v Forgessa, Plowden at p.19


*           An Introduction to the Philosophy of Law, p.79


*           Walden v Hensler (1987) 61 ALJR 646.


*           R v Hurley & Murray (1967) VR 526 – 543.


*           R v Howe (1987) AC 417 – 419.


*           R v Duffy (1949) 1 AER 932.


*           J.E. McDonald.


*           Social Science & Social Pathology, Wootten, pp.250-251.


*           For a comprehensive discussion of criminal responsibility in relation to punishment under the present law see the decisions of the High Court of Australia in Veen v the Queen (1979) 143 CLR 469; (1979) 53 ALJR 305 and Veen v the Queen (No.2) (1987-88) 164 CLR 465; (1988) 62 ALJR 224.  The significance of intention has been reasserted in cases such as Heh Kaw Te (1985) 157 CLR 523.  In England the objective test enunciated by the House of Lords in DPP v Smith (1961) AC 290 has been legislatively repudiated. Even the ancient maxim that ignorance of the law is no excuse has been modified, Lim Chin Aik v The Queen (1963) AC 160.



*           See generally The Humanitarian theory of punishment, Vol 6 Res Judicatae, p.224, C.S. Lewis